AI in GTM

A split between back-office RevOps (systems, process, tickets, quota fixes — never touches the field) and field-operator RevOps (lives between the sales team and the machine, injecting value into forecast calls, campaigns, and programs).

Build your internal operating rhythm around the four phases of how a customer consumes you — awareness, consideration & decision, implementation, and value realization — rather than around your org chart.

Staff sales enablement by treating every non-quota role (managers, RevOps, SEs, enablement) as overhead wrapped around a $1–2M-quota AE, and asking what each role gives back. Budget it zero-based and build it as a living sales academy, not a content factory.

Stand up formal enablement once a frontline manager's span of control passes five or six reps — earlier if you sell complex, enterprise, high-consideration products.

Split the field into hunters who acquire new logos (traditional sales path) and farmers who grow the install base aggressively, then engineer the bridge so neither feels the other is interloping.

In consumption revenue, the signed PO is where the work starts. Because revenue recognizes on usage, the entire post-signature job is driving adoption and demonstrated value.

Land small as a paid pilot, prove value fast, run a ~6-month 'double-tap' true-up, then bridge to the 12-month renewal — which is really the first real deal.

Acquisition and install reps carry different numbers; acquisition sellers ideally carry no consumption quota. Build a bookings plan for hunters and a consumption plan for farmers, layered with spiffs and target-incentive mixes.

Go-to-market gathers raw materials (account plans, commercial events, product releases, macro signals); a centralized data-science function owned by finance turns them into a forecast. Sellers cannot predict consumption.

Leverage is forcing your way into the room by making leaders unprepared without you. Trust is being invited in because sales leaders want you there. Only trust builds durable influence.

RevOps sits in a strange seat: reporting to the CRO but excluded from the CRO's peer conversations, while simultaneously knowing more than most of its peers and hearing things in rooms sales never enters.

AI is a productivity multiplier that requires clean data and human oversight. A productivity gain should be reinvested in more capacity to go faster, not banked as headcount reduction.

Defend content investment internally on three fronts: (1) it outperforms traditional marketing on cost per qualified lead, (2) it generates earned-media value the company would otherwise pay Meta and Google for, and (3) it reframes and lowers the cost of the top of the funnel.

Build media brands as verticals that each map to a product hub (marketing, sales, entrepreneurship, AI), starting from a customer pain point and using TAM plus coverage gaps as the signal to add a brand or channel.

For each media vertical, decide whether to build a brand from scratch, partner with existing creators/journalists, or acquire an established brand — driven by the pace of change in the category and whether trusted voices already exist.

After acquiring a media brand, never meddle with the editorial. Meddle with the editorial → meddle with the audience → meddle with the trust. Instead, arm the brand with insights, production, distribution, and growth capital while leaving the content itself alone.

Produce video first, then repurpose the single asset: video → transcript → audio podcast, newsletter, short-form clips, and SEO/AEO blog post. A human owns the original idea at the start, AI compresses the middle (scripting, outlining, clipping, localization), and a human quality-gates the output before it publishes.

Start and perfect your video strategy on YouTube, where scale gives you the richest early signal on what audiences like, then treat platforms like LinkedIn as new distribution for that content — not as derivative afterthoughts — via a data-trained clipping strategy.

At scale, media can sit under a media-believing CMO. But early-stage media should report to the founder/CEO, not marketing — the founder best knows what the brand needs to be, and marketing ownership risks flattening editorial into product-marketing nobody wants to watch.

HubSpot's replacement for inbound marketing: a way to drive compounding growth in partnership with AI that constantly learns and compounds across paid media, creators, and content, rather than relying on keyword-ranking and hand-raiser capture.

Agentic GTM is composed from a few interchangeable building blocks: a very smart intern (the LLM), web search, a data platform, and a workflows agent that can take or suggest actions on your behalf.

A maturity ladder for AI in GTM: using skills (precise, installable instruction sets) puts you in the top 1% of users; using scheduled tasks (routines that run agents on a cadence) puts you in the top 0.1%.

A research-and-enrichment agent (running in the LLM, plugged into the web) hands its output to an engagement agent living inside the GTM platform (harnessed to owned channels and CRM), which has its own instructions for how to act.

Personalization that generates a bespoke sequence per person: the agent knows whether you're connected, whether you emailed or cold-called before, and what the CRM says, and decides both the copy of each step and the channel each step starts on.

To design agentic flows, ask what recurring work you would hand a very smart intern you just hired — then build agents around that answer, letting the LLM interview you and write the instructions.

Across Amplemarket customers since January 2026, opportunities generated by channel split roughly 40% email, 40% calling, and 20% social — evidence that multi-channel, contextual outbound still works.

AI creates a widening productivity gap between adopters and non-adopters, but market pressure keeps teams employed: if you can go faster, competitors force you to reinvest that speed rather than fire people to move at the old pace.

Design monetization in sequence: packaging first, then pricing structure, then the pricing metric (what you charge on, and whether forward- or backward-looking), and the price point dead last.

Bundle features around the outcomes a customer is trying to achieve, not around a product-team ranking of which features get used most.

The amount of pricing/packaging complexity a business can carry is capped by its ACV and the type of customer it sells to — high ACV can absorb complex enterprise pricing; low-ACV startup sales demand simplicity.

Classic SaaS had near-zero marginal cost and 80–95% margins; AI reintroduces a real, usage-driven cost to serve that most SaaS-native companies can't even measure, so pricing that doesn't follow cost loses money.

Outcome-based is what you price on; usage-based is how you meter it — and they can combine. Outcome pricing fits vertical products with a uniform outcome and is a trap for horizontal products where the same usage means different things to different users.

You're a candidate for outcome-based pricing only if: (1) you can clearly define one outcome across your customer base; (2) customers agree to and accept that exact definition; and (3) the value of that outcome is similar across all customers.

Make usage-based pricing palatable by combining a recurring base fee with usage on top, wrapped in real-time visibility, per-team spend controls, caps/notifications, and token-level billing traceability.

De-risk migrating your most important customer by working backward: test new pricing on the least-important segment or market, then new business, then run validation interviews with your 2nd–5th customers, plus internal validation with the sales team, before migrating the anchor account.

Treat pricing and packaging as a living product surface, revisited every product-release cycle or sales cycle (whichever is longer) — not set once and left for five to seven years.

A framework for evaluating a 'backwards' move from CEO to CRO: weigh product-market fit, founding-team fit, and investment thesis against the compensation, equity, and personal-fulfillment math of joining a high-growth build with strong culture.

The idea that the go-to-market motion — talent profile, process weight, and metrics — must be rebuilt at each ARR band rather than scaled linearly. Different stages leverage different areas of the process.

Gate sales hiring on two leading indicators: how well reps are ramping against a defined ramp curve, and what percent of quota (and ramped-quota capacity) they are attaining. Below threshold, pull the plan back.

The second-order damage of over-hiring sales: diluting territories and top-of-funnel demand across reps who won't stick, which starves top performers and eventually drives your A-players out.

A hiring heuristic for specialized verticals: emulate your customer base and screen for the common denominator of work ethic and mission alignment rather than a specific sales or industry pedigree.

An AI-adoption operating model: allow broad, decentralized experimentation to reduce fear and prove ease of use, then centralize the valuable skills, agents, and data pipelines — governed by RevOps — for anything mission-critical.

A pipeline philosophy that favors thoughtful, researched, use-case-specific outreach to a narrow buyer over high-volume, low-hit-rate blasting.

Treating in-person events as an operational play with three phases — pre-plan (targeting, pre-set meetings), execute (on-site, ideally with stage presence), and post-plan (structured follow-up tracked through CRM) — not as a booth you show up to.

The primary barrier to AI adoption is a mental model, not a skill gap: people onboarded in a pre-AI world treat AI as the next, harder evolution of technology and cling to point-and-click UI notions.

A model of selling as two blended disciplines: the art (psychology and influence — moving many stakeholders in the same direction) and the science (methodically progressing a deal through a rigorous process to signature).

The reframe that managing a sale is like managing a project — bringing operational and project-management rigor (sequencing, stakeholders, milestones) to progressing a deal to close.

A transformation framing in which agentic AI augments the revenue team rather than replacing it — the 'plus' signals that humans stay in the motion, owning relationships and accountability, while agents handle preparation and scale.

The higher the deal value and the more up-market the customer, the less AI belongs in the customer-facing interaction — and the further down the tail (SMB), the more the agent can own the motion with a human reviewing the output.

A two-variable decision model for how much AI to put into any motion: account value (ACV) and which product surface the customer is touching (and how mature that surface is).

Split customers into enterprise (large advertisers), mid-market (D2C brands, performance agencies), and SMB, and assign a different agentic role to each based on that segment's customer-service needs and risk tolerance.

The newer and less proven a product, the more human-in-the-loop the motion should be — because the fastest way to learn from customers experiencing something new is to talk to them, not to automate the interaction.

AI enablement swings between a centralized owning group and fully decentralized team-by-team ownership; the healthy resting point is in the middle — cost-and-tool guardrails set centrally, process redesign owned by the teams.

Whether you can decentralize AI at all is gated by the strength of your underlying data — a clean CRM and a healthy data stack are the precondition for letting functions own their own AI.

Judge AI initiatives by revenue and efficiency outcomes — speed to market, meeting volume, pipeline-stage conversion, revenue per head, ARPU — rather than by AI spend.

The operator's path to senior leadership: deliberately pursue new lines of business, international expansion, and reorgs so you see and understand the entire business, not just one function.

The point of 'carrying a bag' isn't the title — it's going through a period where you genuinely feel the pressure of contributing to the top line, a career experience you have to go collect.

Run your weeks against roughly four equal priorities, each assigned a color, and audit your calendar so it's about a quarter of each — a mechanism to keep strategic time allocation honest.

Deliberately rotate through operations, marketing, partnerships, consulting, sales ops, and product before ever carrying a quota, so you understand everything that actually affects revenue — rather than reaching the CRO seat straight up the sales track.

Stay in every role at least a year (sometimes two) to actually learn the skill, and when hiring, evaluate candidates on increasing responsibility and achievement rather than raw time-in-seat.

Convert a product-led motion into an enterprise motion by getting selective on collaboration-heavy segments, landing two or three teams or divisions, then uniting them under one executive with a combined security, collaboration, and cost case.

If you find yourself in a competitive cycle defined by a competitor's strengths, one of you is in the wrong cycle — and it's probably you. Know your weaknesses so you can avoid the fights they define, and concentrate on the ICP that values your strengths.

A private trial-run roadshow before the public IPO roadshow: executives travel separately to a low-profile event and pitch bankers who signal buy-or-pass on an app, letting the company watch the book oversubscribe and the price move before the S-1 debut.

An acquisition demands the acquirer audit every contract, approval, and pipeline metric to validate revenue durability, and then run a full integration of systems, org, and process — a burden an IPO never imposes.

Startup equity is worth literally zero until an IPO or acquisition. Secondary sales are rare, board-gated, and usually capped; in a buyout, investors are paid first, so if the exit isn't large enough your equity can be nothing.

Revenue leaders should personally build at least one or two agents (you can ask Claude to teach you) so they understand the power, scope, and correctness constraints well enough to manage AI-driven GTM — the same way understanding marketing and ops makes you a better revenue leader.

Lead with deep domain expertise and your own thinking captured on paper first — not with AI-generated first-draft language — then use AI to fill gaps and propel execution rather than to create ideas you can't defend.

A three-layer split of GTM ownership: executives (CEO, CRO) own the WHY (market, category, how we win); the VP of RevOps owns the WHAT (processes, business and operating strategy, scalable design); GTM engineers own the HOW (enrichment, automation, ICP plumbing, execution).

The RevOps role is splitting from a generalist (decent at business and systems admin) into two lanes: the deeply technical GTM-engineering lane, and the strategic decision-maker accountable for the GTM infrastructure overall.

Effective AI transformation must change all three legs at once — people (how teams work and are structured), process (re-architected end-to-end), and technology (AI-first infrastructure and data) — not just automate external workflows.

The central tension for a RevOps leader: keep running the non-stop operating machine (forecasting, pipeline, QBRs, territory and account planning, comp) while simultaneously leading an AI-first transformation — usually with the same headcount and a mandate to use fewer people.

Tessa's methodology scores an operator's or org's AI adoption from 0 to 5 — where 0 or 1 is basic use (asking questions, rewriting an email) and higher levels reach standardized workflows and autonomous agents.

Sort work by urgency and importance: do the highly-important-and-urgent first, delegate the urgent-but-low-importance, and protect time for the highly-important-but-not-urgent — always weighing level of effort per initiative.

Start in 'analog mode' — write your own thoughts, plan, and hypothesis manually using your own judgment — then use AI to find examples, metaphors, and crunch data to back it up and level it up.

A tactical first step for any operator: open a Google Sheet, list the core tasks you do daily, weekly, monthly, and quarterly, mark the level of effort and whether each is manual or automated, then map where AI or an agent could help — and how peers in your role are doing it.

Invert the usual order of operations: instead of perfecting audience cohorts and reusing one asset, start with what you're trying to say and the mindset you're reaching, then treat every ad as a blank canvas customized to that audience and context.

For one-to-many media like DOOH, target the location and mindset of the screen rather than the individual, and vary the creative per venue-context so it feels relevant to many without being invasive.

Bring a new product line to market in three moves: assess and wire up supply/partners, launch as a high-touch managed service with trusted betas to 'fail and win, fail and win,' then open self-serve, and finally close the measurement loop to prove impact.

Design channels to reinforce each other in sequence — a brand seen on a digital-out-of-home screen is retargeted on mobile — so the story continues across screens rather than each channel running in isolation.

Before opening a new channel to market, educate and certify the internal team so everyone can represent it consistently — treat internal readiness as a prerequisite for external demand.

Put three parties in the room together — the agency (strategy and deployment), the brand (positioning and creative appetite), and the platform (the creative that lands) — each owning a distinct role, compounded by shared measurement.

Spend the opening months of a new role absorbing information from two sources — the people on the ground doing the selling and implementing, and the available data — then marry those perspectives into a working theory of what's actually happening before setting priorities.

When you miss a quarter, an intellectually honest post-mortem usually ties the miss to a planning or strategy assumption you weren't honest about — not to near-term deal execution.

Charge only when the AI agent completes the entire job correctly (e.g., reads and infers every field on a document 100% right), so price tracks roughly one-to-one with the value delivered.

Report usage-based revenue to the board by stacking two clearly labeled layers: contracted ARR ('take it to the bank') plus a conservative fraction of the forecasted amount booked as 'estimated ACV' (EACV).

Start with a small, high-confidence use case you know you can nail, prove value fast, and use that win to earn the right to the next project — becoming the customer's primary consideration for what's next.

Qualify and sell the deal first, then run the POC only to confirm the solution works and the teams click — never as a desperate Hail Mary to generate intent that isn't there.

Before offering a POC, square away the MSA, legal and security, and budget, and get both IT and operations (the business side) at the table and excited. The POC then only validates the solution and the working relationship.

Have the implementation / agent-PM team scope the work during the sales cycle, so the buyer meets who they'll work with, gains confidence, and sellers can't over-promise.

Build the revenue plan so every channel has its own win rate and ASP and new-logo ('NUCO') plugs the gap to the number — robust enough that a missed quarter can be traced to a specific assumption that broke, with leading indicators warning you a quarter or two out.

Leadership sets a non-negotiable number; what's up for debate is only the resources required to deliver it. The exercise is iterative and cross-functional, and pairs with a proactive 'what would it take to 10x my org' model run before the CEO asks.

When standing up or fixing a go-to-market org, sequence your build in a deliberate order: put the systems (RevOps backbone) in place first, then the people, then the process.

In heavy industries the first contract is the audition, not the win. Delivering it at a high bar earns the right to the 'real contract' — the bigger, expansion opportunity that follows.

RevOps is 'the language in which companies test, measure, learn, and drive rapid scalability' — the first port of call at any company — not just Salesforce hygiene.

The senior CRO responsibility is demonstrating control over the number — knowing when you're behind, what the corrective actions are, and whether they're working — rather than merely landing the target.

In concentrated, capital-intensive industries, expansion doesn't come from more seats or another module — it comes from earning trust through delivery so the customer opens the aperture to bigger questions.

A hiring thesis for complex industries: recruit people who've operated in the space and can speak with credibility, then teach them the selling motion — screening above all for learning agility and structured communication.

Treat the service/delivery organization as a co-equal leg of the stool alongside sales and account management — not as a margin-enhancement play.

Patch's consulting arm embeds strategists directly with customers to navigate the complexity and information asymmetry of carbon markets — its version of the forward-deployed engineer.

Durable differentiation in complex industries comes from combining software, human expertise, and the proprietary data the product generates — not from any one of them alone.

The classic three-horizon framework (core business, adjacent bets, and future/experimental bets) becomes far more actionable when AI lets you experiment cheaply.

Three sales processes run in parallel and then fused: win the fintech that wants a banking/card product, sign a sponsor bank willing to back the program, and marry the two under a single tri-party agreement.

Bake forecasting into qualification as a trade: the customer shares projections (customer counts, average spend) and in return receives a professionally built deal model showing how the program becomes profitable — one shared document both sides work from.

Use the sales engineer's solution document — effectively a statement of work — as the artifact that holds every party accountable to exactly what was scoped and approved.

The sequence in which a founder should lay down go-to-market foundations — with RevOps placed effectively first, right after the first salesperson, before scaled AE or BDR headcount.

Treat each go-to-market segment (enterprise, mid-market, SMB, and their international variants) as its own line of business, with distinct product needs, marketing plan, ACV/LTV, conversion rate, sales cycle, and quotas.

Identify and win over the people who can kill a deal — often someone you never meet, like compliance or a bank's board — rather than over-investing only in the enthusiastic champion.

The fear that native-AI companies will displace SaaS incumbents that bolt AI onto legacy architecture — and the buyer's inability to tell a truly native-AI product from a SaaS wrapper.

Meet the customer in the middle by proving value in their own environment through a proof of concept, sharing risk, then expanding — making the POC the default go-to-market move rather than a concession.

AI deals require selling horizontally across the functional buyer (HR generalists, HR ops, HR leadership), IT, an AI committee, and security — any of whom can veto or delay the deal.

The economic buyer — the person with discretionary authority to say yes and move budget — has gone horizontal in AI deals; IT is often the new economic buyer even on an HR purchase.

Embedding technical forward-deployed engineers into the implementation team to manage LLM change, build guardrails against hallucination, and hand-hold customers through early adoption — modeled as an accounts-per-FDE/CS gearing ratio in headcount planning.

An implementation philosophy (borrowed from Indiana football coach Curt Cignetti's 'stacking days, stacking wins') of engineering a continuous drumbeat of provable metrics and success stories the champion can tell internally.

Extending the go-to-market engineer role beyond top-of-funnel prospecting into mid-funnel deal execution — automated SOWs from call transcripts, company-specific deal coaching, and CRM-plugged GTM diagnostics.

The enterprise qualification methodology Scott helped develop; in AI's chaos the two most decisive elements he stresses are Champion ('no champion, no deal') and Decision Criteria — the capability shopping list a buyer uses to evaluate vendors.

Rather than extracting the buyer's decision criteria, supply it: an editable, weighted, unbranded capability scorecard that lets the customer objectively compare vendors for the problem they're solving.

Auto-fill MEDDPICC in the CRM from Gong call transcripts while also keeping rep-entered MEDDPICC, then compare and contrast the two to triangulate what's actually happening in accounts.

When a company hires a go-to-market leader, RevOps and enablement are the first two hires; the ecosystem is built before AEs so reps ramp fast into a well-oiled machine.

Each new media/computing paradigm — radio, TV, web, social — builds a wholly new advertising infrastructure around it, and advertising never disappears. AI is the next paradigm and will get its own reinvented ad stack.

Instead of one agency-created, brand-approved ad shown to everyone, every person is served an ad generated specifically for them — the long-sought 'holy grail' that AI can finally deliver at scale.

A ladder from a fixed banner to fully generated advertising: start with dynamic text generated to match the conversation, move up to generated assets, then to dynamic UI — the ad format and interface generated per user and per brand.

The publisher's surface area is treated as sacred, and the ad platform optimizes for the end user first, the publisher second, and the advertiser third — because end-user value drives engagement, which in turn makes advertisers happy.

Click-through rate is a misleading success measure because users can click without converting; the right target is the advertiser's actual objective (conversion/value), which should hold steady rather than decay if the ads are genuinely relevant.

The early assumption that 'user asks about shoes → show a shoe ad → they buy' is wrong; buyers research with an LLM but still purchase elsewhere to price-shop and earn rewards. Closing that behavior gap takes time, not just better technology.

The more expensive or complex a purchase, the more a buyer researches first — so AI ads convert best for high-consideration, financial categories (taxes, student loans, credit cards), amplified by seasonality like tax day.

A north star of returning advertising to craft — the era when a Got Milk campaign or a full-page New York Times ad was a genuine piece of art — using generative tools to produce brand-true experiences people admire rather than block.

The market is splitting into companies that use AI to introduce precision (tighter ICP, enforced qualification, best-practice discipline) and companies that use AI to generate volume (infinite leads on top of an undefined motion).

Putting AI on top of an existing go-to-market motion exacerbates whatever is already broken — much like practicing a bad golf swing makes your game worse, not better.

When sellers carry too much pipeline, win rates drop dramatically because they engage and multi-thread less; a balanced pipeline wins at nearly twice the rate.

ICP is a small, well-understood segment defined by fit-and-timing signals — not the entire universe of companies you could theoretically sell to (TAM).

Keep the ICP you use for the fundraising/exit growth story in separate books from the tighter ICP your sellers chase every day.

The best sellers disqualify roughly three quarters of their opportunities by discovery, advancing only ~25% — which produces late-stage conversion above 70%.

Sales efficiency measured as dollars generated per day; larger deals ($70k+ ACV) are over 6x more efficient because they don't take proportionally longer and carry more expansion potential.

Deals with six or more stakeholders win at nearly 4x the rate, and buying committees keep growing — so multi-threading is increasingly decisive.

A 360-degree seller who self-sources pipeline, closes, and stays on as the commercial point of contact through land-and-expand — replacing the single-purpose relay of SDR → AE → CSM.

A recurring ~50-page audit that connects to the platform in two hours, looks back a year over won and lost deals, and reports across five chapters: sales-efficiency trend, win/loss analysis, live-pipeline risk, rep coaching gaps, and sales-process friction.

Small, stacked gains — 10% better ICP targeting, 10% better qualification, 10% more multi-threading — compound quarter over quarter into materially different results within three or four quarters.

For AI products, the old default channels — performance marketing and SEO — give way to two new engines: social video that demonstrates the product's output, and visibility inside LLMs (ChatGPT, Claude).

A hyper-compressed planning rhythm: weekly planning against a monthly calendar, no plans beyond the month, and an annual North Star kept as a vision rather than a locked-in plan.

A three-part decision model for a fast-moving market where clean data lags reality: use data where it exists, intuition to see patterns beyond the numbers, and how fast you can learn as a tie-breaker — preferring to go, do, and learn over planning to perfection.

Get surfaced inside models by writing intent-based 'how-to' content built around how a real user searches, in longer formats (a rough 30,000-character bar), and by building presence in the forums LLMs read — Reddit first.

Start with a branded subreddit treated like any social account, then engage authentically in relevant subreddits — see what people ask, help them, and inject product info where genuinely useful. Avoid the 'cheating farm' of over-promising, irrelevant subreddits, and pushy advertising.

The trap of chasing growth without delivering value — reframed as: a KPI to grow by X really means figuring out how much value (Y) you must create to earn it.

Rebrand only when it's existential — required to capture a market you must be in — not because a name feels sexier or slightly clearer. Then sequence the rollout: decide, warn loyal users early, switch the domain silently, then announce.

The three things a startup CEO looks for when hiring from big tech: ability to be hands-on (even C-level does the work), courage to own decisions personally, and agility to change direction fast.

Demystification of the vocabulary: a 'repository' is a folder, and an 'agent' is a folder containing a set of instructions saved as a file. You 'program' or 'train' the agent by writing its SOP in natural language and triggering it with an automation.

The operating system is built on four repos/folders: (1) transcript warehouse (raw call input), (2) customer warehouse (per-account intel and context), (3) GTM library (in-depth playbooks), and (4) company context (brand guidelines, customer avatars, pain points).

A relay where each agent prepares data for the next: a transcript agent annotates and routes calls, a customer-warehouse agent enriches account files from those notes, and a working agent (e.g., territory design) consumes the pre-built context to do real GTM work.

A body of enriched files — per-customer context, playbooks, company avatars and brand — authored so agents can inherit memory. The documents are written for agents to read, not humans: 'made by agents, for agents, used by agents.'

The agent platform (Claude Code, Claude Cowork, OpenAI Codex, Google Antigravity) becomes the central interface for the whole organization because, via MCP, it is tool-agnostic — pulling from and writing to HubSpot, Salesforce, Google Drive, Snowflake, and Intercom.

Because agents can write back into files, every implementation can update the source playbook with new learnings, so the system improves itself with each customer and prospect rather than staying static.

Each major lab (OpenAI, Google, Anthropic) ships three distinct products: the model (baseline infrastructure — GPT-5.2, Gemini 3, Claude Opus 4.5), the consumer app (the browser chatbot for general-population Q&A), and the agent platform (for professionals to get work done).

The capabilities that only agent platforms have and consumer apps lack: a persistent internal to-do list (so the agent works 10–40+ minutes autonomously), the ability to launch sub-agents, reading and writing files on your local machine, permission/plan modes, queued messages, and context compaction.

A repeatable structure for building agent prompts: (1) supply context files, (2) instruct it to launch sub-agents, (3) have it maintain a to-do list, (4) tell it to be token-efficient, (5) direct it to write outputs to files, and (6) frame it to think like a strategist / thought partner.

A token is the atomic unit of how AI thinks (~3–4 characters). The context window is finite working memory holding all inputs and outputs (e.g., 200K for Claude 4.5, 1M for Gemini 3); once full, the agent forgets earlier context. 'Compacting' summarizes the current context and hands it off to a fresh agent with a clean window.

A skill is a folder of files that teaches an agent how to perform a task (make a PowerPoint, an SOP, a PDF, wireframes). The labs adopted a shared skills standard, so you can download skills from the internet or build your own and point the agent at the skill's path to execute it.

Getting an agent platform running in ~5 minutes: (1) download VS Code, (2) install the official Claude Code extension from Anthropic, (3) log in with a $20/month Claude subscription. Restart, click the orange icon, authorize, and the agent is enabled.

A go-to-market model with no contracts and no renewals — pure pay-as-you-go, even on multi-million-dollar enterprise deals — so the customer can leave at any time and the company effectively re-earns the business every single day.

Anthony's framing for the startup-vs-scale-up choice: a big ship already has a direction and goes far; a jet ski is nimble and yours to steer. Oliver's corollary: at early stage you work ON the company (creating its DNA), and at scale you work IN it (executing).

Define ARR by taking collected revenue and extrapolating it forward twelve months, then stacking disciplined, repeatable consumption projections — derived from a customer's telemetry data or their prior-vendor usage — on top of the extrapolated actuals.

Pay commissions up front on a projected ACV — the customer's estimated one-to-three-year spend — protected by security-margin haircuts, a two-person review of the estimate, a rep-adjustable projection, and a consume-to-earn provision that only fully vests the commission once the customer actually consumes.

A comp plan surfaces problems but rarely causes them. If deals collapse at scale, the root cause is bad hires or a product that isn't doing its job — not the incentive structure, which only mitigates or escalates the underlying issue.

Staff a company in waves: SEALs first (operate confidently without supervision in uncharted terrain, mission evolves as they gather intel in the field), then Marines (build the foundation), then infantry (scale once the foundation exists).

Motivation is why someone joined your cause — intrinsic, and not the leader's job to install. Morale is how someone feels in a given moment — situational, and squarely the leader's job to manage by supporting people through hardship without removing it.

The real leverage from AI isn't the model — it's the discipline of organizing and maintaining company knowledge so an agent has clean, current context to draw from, turning a one-hour subject-matter meeting into a five-minute prompt that's 95% right (plus a ten-minute human review of the output).

AI Engine Optimization (AEO/GEO): the discipline of improving how a brand is ranked and cited inside LLM recommendations. The defining stat is that up to 85% of what AI recommends comes from third-party content, not your own website.

The categories of off-site content LLMs weight most: review sites (e.g., G2), small influencers and YouTube, digital publications and blogs covering your category and competitors, and integration content co-published with partners.

Run an AI-visibility analysis to surface the URLs you should be working with and the content types you lack, then use those two outputs to build the partnership and content roadmap.

There's no universal first step; you diagnose your existing content baseline and attack your biggest gap — whether that's missing technical FAQs, a better-ranked competitor, negative brand sentiment, or a story that's stale after a pivot or new product.

Model partnerships like any other channel: TAM, then persona overlap, then a per-quarter penetration assumption, pipeline conversion, and win rate into new bookings — the same funnel logic marketing and outbound use.

Partner-involved revenue retains and expands better than direct: PartnerStack benchmarks show ~130 NDR on partner-managed revenue vs. ~105 on the core business (and ~108 partner-sourced), plus better gross retention and lower cost-to-serve.

Zero-to-five (pre-PMF): usually no partnerships hire. Post-PMF: invest early but patiently, sizing partnerships as a third GTM pillar alongside sales and marketing, targeting 30-40% of total revenue over time.

The defining shift from a sales org (where you sell the vision and the full range of what's possible) to a customer-success org (where you have to deliver on promises that are sometimes bigger than reality).

The discipline, on every promotion, of deliberately stepping back to relearn a department and role rather than carrying your previous job forward under a new title.

As the cost to build applications collapses and AEO churns like the SEO-algorithm era, durable defensibility comes less from product and more from a broad, diverse ecosystem of partner-evangelists.

The single diagnostic Alex uses to decide when and how to change GTM: at any moment, identify who controls the client's decision, who controls the revenue, and who controls the margin. When the answer changes, the go-to-market must change.

A market maturity curve every industry travels: from a phase where you must educate buyers from scratch (highest margin, lowest competition), through growing awareness and competition, to full commoditization where price pressure peaks.

The 'golden era' — strong demand, high margin, still-low competition, educated buyers — is not a reward to enjoy but the starting point of commoditization, and it signals you should already be building the next product.

A talent principle (from The Science of Scaling) of evaluating people, customers, and standards by their worst-day performance rather than their potential — like a professional athlete who is great on their worst day, not just in flashes.

A portfolio-scaling model where a commoditizing, lower-margin core product is used as an entry wedge, and higher-margin products sitting earlier on the educational curve are layered on top — replicated in-house or acquired — repeating at every level.

A market-intelligence practice of tracking where venture capital — especially seed and pre-seed — allocates capital, categorized by segment, as the cheapest and smartest signal of the next big thing two to three years out.

An AI-native operating default: when you have a real need you'd pay for but can't find the right tool (or it's too expensive), build it yourself with AI coding tools rather than waiting on a vendor.

The remote-vs-office debate is a false binary. The real variable isn't where people work but how intentionally the right people are brought together — connection can be engineered without full-time co-location.

The magnet that makes an office worth showing up for is the interactions with the right people, not the space or its amenities.

Coordinating people day-to-day in space — honoring individual flexibility, team adjacencies, and the actual work being done — is a Rubik's cube problem that exceeds human capability and is well suited to AI.

A workplace failure mode where a building holds scattered pockets of two or three people with gaps in between, so it's technically occupied but feels low-energy and dead.

Distributed organizations progress through stages — a single co-located hub, a fully distributed org, then localized clusters — each requiring a different connection cadence.

When a company demonstrates tangible care for employees — above all, respect for their time — employees reciprocate that care back into the business with dividends.

Adopt AI by targeting real, painful processes and asking whether AI can do each one — or do it better, faster, or cheaper — rather than handing everyone an open-ended LLM.

Kotter's change-management allegory Brett invokes for the RevOps reality: you may be the one who spots the crack in the iceberg, but seeing it isn't enough — you have to sell the change to 'the elders' and earn consensus before anything moves.

A mentor's operating standard: if a leader asks for a square, come back with a square (or they'll discount everything else you say), but if you know a circle is what they really need, bring that too.

A boss's line — 'you're not a race car driver, but you know how to build a really fast race car' — that captured why an operator who understands funnel mechanics, handoffs, and the sales cycle can be handed the wheel of the team.

Resolve inside-vs-field conflict by defining discrete functions — specialists who do top-of-funnel work and AEs who land-and-expand existing customers — with executive-mandated boundaries nobody is allowed to cross.

Build a recurring team forum where anyone can say 'I need help with this,' because no individual coach can ever exceed the combined knowledge of the whole team — the highest-leverage part of a leader's cadence.

A meme of the modern SaaS stack — cloud, kernel, and applications neatly piled up — with AI as the Angry Bird flung in to topple the whole tower.

A productivity multiplier should be reinvested in output, not headcount reduction: if you can be a thousand times more productive, produce a thousand times more rather than gut the staff.

The load-bearing reason AI won't replace high-trust, complex sales: if something goes wrong, there's no one on the hook, no career on the line, no justice to be served.

As buyers research through AI chat, the website's job shifts: detect whether an LLM bot is visiting, serve it structured content to shape what it brings back, and push your information onto off-site links and affiliates the models cite.

Catalog every task each function performs, then sort each into two buckets — 'can I automate this with AI' versus 'this is inherent to the function itself' — alongside a competency matrix and clear career on/off ramps.

Brett's 2026 kickoff message: the only constant is change — or, in the truer Heraclitus phrasing, 'although you're standing in the same river, the water flowing through it is always different.'

In a world of unlimited information and opportunity, the scarce edge is clarity — the ability to focus on the few highest-priority problems and not over-index on the feeling of stress. Individual clarity and organizational clarity move together.

A concept coined by a16z's Martin Casado: unlike product-market fit (fitting a product to existing demand), market annealing means shaping the market itself — educating buyers, defining the demand, and shaping the product in parallel.

The platform is a world-class kitchen that can prepare any 'meal' (extract value from any image or video data). Rather than acting as waiters serving every hungry customer a different dish, you build one focused 'hamburger stand' — a single killer app sold 100% outbound — to prove a restaurant can be built on top of the kitchen.

A powerful platform doesn't create instant value on its own; it needs a killer application that lets customers get value immediately — the way Databricks needed notebooks before people could realize its value.

Instead of marking up hyperscaler infrastructure and competing on its margin, pass that cost through at parity and charge on the usage or value delivered on top of it.

A weekly leadership operating cadence with no fixed agenda or set times: the team gathers to solve the hardest problems and works until the set of things is finished, ordering pizza along the way.

A progression of accountability: an IC empowers themselves; a head of sales empowers through people and owns a team; a CRO takes accountability for the whole company — vision, strategy, fundraising, and product influence.

Just as engineering accrues technical debt, an organization accrues partner debt, customer debt, and employee debt by taking on too many things at once and failing to fulfill the promises made.

Define the outcome you want and stay agnostic about how each person reaches it. Process is a safety net and a ramp for building habits, not the objective; the way an outcome is achieved should be 'completely irrelevant' as long as the outcome is right.

A leader's job, like a coach's, is to tap into the best parts of each person's natural style and put the right people in the right positions to build the best total team — not to standardize everyone toward one form.

Recruit for demonstrated problem-solving ability and internal drive rather than credentials (Ivy League degree, MBA, finance background). A sales role is fundamentally problem-solving done all day; pedigree is rarely the requisite it's assumed to be.

Compress a hard enterprise/government sales cycle by building an external ecosystem whose desired outcome equals yours — cooperative-purchasing bodies, complementary technology partners, and lobbyists — instead of scaling a bigger direct team.

Use cooperative-purchasing organizations (Sourcewell, HGAC) — which let one public agency's pre-competed, approved purchase serve as validation that another agency can buy the same way — as the engine that manufactures trust and shortens the buy.

When a competitor's strengths complement rather than fully overlap yours, convert the rivalry into an integrated co-sell: lead with the shared outcome ('if we compete, one of us loses; together we both win') and prove a repeatable joint motion on one marquee deal.

A market signal: legacy verticals that historically treated software as a cost center or risk-mitigation expense begin treating it as a competitive advantage, and the share of enterprises in-market for software jumps from a typical ~5% per year toward ~50%.

In an AI world, the cost of being wrong has collapsed, so iteration velocity — not first-time correctness — is the dominant competitive advantage in go-to-market.

Do the reps yourself with an AI-built tool and an expert on call, so you compress the failure that an outside agency would spread over months into a couple of fast, cheap weeks.

Any task that looks like one step ('start Google Ads') is really dozens of steps, and not knowing where to start — or which step is next — is what actually stops people, not the tool.

Reverse-engineer every revenue goal from the sales cycle and pipeline coverage, then act with the urgency that math demands.

For AI-native companies at scale, the customization math tips from buy to build once one engineer can build the tool in a month and any coding agent can maintain it.

The durable value in a GTM tool isn't the code — it's the product manager's hard-won knowledge of what people actually do with it: the workflows, sequence, guardrails, and integrations.

Get a paying customer on an MVP or front-end-only demo before building the real product, so you validate that someone will actually pull out a credit card.

AI gets you ~99% of the way, but the final mile still takes human taste, focus, and follow-through — a six-month project becomes four days, not four minutes.

You can only automate a motion well if you understand it cold — which is why AI SDRs built by people who never ran a sales process miss the mark.

The quality of any tool reflects how core its job is to the business that built it — so for jobs that matter, choose the obsessed point solution over the all-in-one chasing more NRR.

Match every hire, tool, and system to your actual stage instead of copying what much larger companies do.

The best GTM advice is focus — narrow the ICP and the product, and get comfortable saying no early.

Since go-to-market process breaks whenever it depends on human compliance, the fix isn't more enforcement (mandatory fields, stage gates) but removing people from the data-capture loop entirely — letting AI listen and populate the system automatically.

Once AI captures CRM data automatically, the salesperson stops being a producer of data (data entry) and becomes a consumer of it — served a prioritized view of what's healthy, what's slipping, and what to work on next.

Adopt in stages: crawl (RevOps connects CRM, call recorder, Slack/Teams, and email, sets team structure and field mappings for a two-way sync), walk (auto-create and enrich records, remove humans from data entry), run (deal-health analysis, agents, and cross-functional data products).

Plot every opportunity on two axes: horizontal = how healthy the deal is (likelihood to win), vertical = how likely it is to close when the rep expects. The quadrants surface safe bets, acceleration opportunities (will close but not this quarter), and firm-decision-date deals where you may not be selected.

A full-reasoning agent wired to every captured touchpoint plus external research tools that executes deal tasks — building a custom ROI calculator to the prospect's own metrics, pulling industry benchmarks, and drafting the decision-maker email.

A per-contact score from -10 to +10 that identifies champions and blockers at a glance, with the reasons and specific quotes behind each. Filtering contacts by ICP persona and a high promoter score produces a live, shareable list of advocates.

Product roadmap signal should come from live sales conversations with the market — use cases, friction, competitors mentioned — not primarily from customer success and existing customers, who are biased because their problem already feels solved.

RevOps problems are hard to solve but remarkably consistent across similar-stage companies — a Series B sales-led company has the same problems and the same fixes as its peers — which is why the function outsources well while sales and product must stay in-house.

A maturity path for tying customer success to revenue: crawl (run a value cycle, lead with hard value, and book CSMs under S&M not COGS), walk (give CSMs CSQL goals and track the funnel), run (train CSMs to close simple upsells, or add an account-management layer inside the CS org for complex ones).

A way to tie business outcomes at the customer back to your product. Soft value is sentiment-based (how the customer feels); hard value is measurable — hours saved, dollars saved, headcount saved — that you can attach real numbers to.

When CSMs don't own the upsell directly, they're accountable for surfacing a set number of customer success qualified leads through normal customer work and handing them to sales; the leader tracks close rate, cycle time, revenue, and funnel shape.

Call it a bonus, not a commission, to shift the mindset. Pay 80% base / 20% bonus, split into two or three parts. The three-part version weights NPS, gross retention (an individual number), and net retention (a company/team goal) a third each; the two-part version drops NPS for individual gross plus company net, with an upside kicker above 115% NRR.

Give everyone in the company — not just customer-facing roles — a small bonus tied to NPS and NRR, so office managers and engineers alike have a stake in customer sentiment and retention.

Picture a bucket with capacity 100 (100% retention). The hose pouring in is revenue; you want to fill and overflow the bucket (>100% NRR). Every hole punched in the bucket is churn.

A CS-plus-sales/AM pod structure is justified only when the average contract value and the available 'green space' to expand support the coverage cost; otherwise a single AM covers the whole portfolio.

Have each team member list what they do daily, weekly, and monthly. Anything that doesn't require critical thinking is a candidate to hand to AI — via custom GPTs or purpose-built tools — freeing CSMs for critical thinking and human relationship-building.

As AI and no-code make building products easy, the hard problem shifts from creation to distribution — getting a great product in front of its rightful customers in an attention (eyeball) economy.

An outbound platform architected for AI from the ground up as a multi-agent system — each agent using the model it's best at — rather than a pre-AI product with AI 'slapped on top' via chatbots or plugins.

A workflow where you paste a domain, the system scrapes it, infers your ICP and buyer personas, writes them out as reusable context files, and converts them into a targeted lead list that also powers copywriting and qualification.

Use intent signals (job changes, hiring, department growth, 10-K priorities, life events) to decide who to contact and when — but keep them out of the message. Mentioning the signal wastes scarce email real estate and doesn't impress the buyer.

Replace the standard chain — Sales Navigator for lists, Apollo and other enrichment tools, a verifier, and ChatGPT deep research — with a single AI-native system on a fair, usage-scaled credit model.

Bridge the gap between sellers who understand angles but not tooling and 'GTMEs' who understand tooling but not selling by giving one strategy-fluent operator an easy-but-sophisticated execution system.

Before feeding anything into AI, do the old-school work of identifying your audience and nailing your ICP and personas; AI is an accelerant layered on top of that bedrock, never a substitute for it.

A mid-market-to-enterprise deal has 11–15 stakeholders: the executive sponsor (C-suite), the director-level champion who discovers and evaluates vendors, the end users, and the technical implementers who hold integration veto power. You must speak to every one.

A range from shallow (know the main persona in the buying committee) to deep (true one-to-one: identify a website visitor, enrich them in Clay, pull public psychographics, and send a personalized message).

Research a target account's public data, draft an insight-rich 'we admire you' breakdown, have the founder refine it in their voice and post it, then let the AE reference it live in the deal to accelerate the cycle.

Rather than over-engineering an attribution model, ask 'how did you hear about us?' on the request form, in the disco script, and again in the first or second demo — capturing self-reported touchpoints across the funnel.

Three curated NotebookLM knowledge bases: (1) competitive research (~90 competitors' value props), (2) product docs, release notes, and analyst reports, and (3) ICP/persona research plus Gong pain-point call transcripts.

Pull close-won/close-lost call transcripts into an LLM, extract pain points, convert them to topics, cross ~82 personas by ~50 topics into hundreds of monitored prompts, and let each prompt seed a blog post, white paper, or ad/email copy.

First do the blocking-and-tackling (know your customer, produce content, know your channels and measurement); only then 'abstract' into distinctive, delightful, B2C-style campaigns that break through the noise.

Andy's written checklist of the go-to-market foundations fast-growing (roughly Series A) companies forget: the data foundation, GTM tooling, the right metrics to track, efficient processes, CPQ, and enablement.

Build formal enablement when you start cloning sales teams and multiplying products and complexity. Below that — one manager, fewer than ~10 reps — the manager owns enablement and rep ops themselves.

Enablement hires come in two shapes — the former rep you train up, and the teacher-type with an ops mind. Either succeeds only if they partner with sales leaders as the prioritization function and hold an opinion on what to train.

You're a passive job-seeker — always with a role lined up or recruiters chasing you — until you get 'punched in the face': laid off, in conflict with a boss, or at a company that ran out of money, forcing a proactive, jarring search.

At the VP/C-level, the odds of a role being publicly posted are low; it's whispered to you through the network. Whispered captures the confidential company insight execs gather while interviewing (and then normally throw away) into a durable edge.

A resilience test for data architecture: if we deleted your CRM instance today, how exposed are you? Teams with a true data warehouse as source of truth could bolt on a new front end and be fine.

The core RevOps mindset: configure systems to fit the business rather than deeply customizing them into brittle, un-maintainable states. Paired with a data skill set (SQL, which AI now makes easy).

Put the GTM engineer role inside the RevOps org: first build the data foundation, then build AI agents on top of it. Keep it aligned so automation solves root problems, not just the surface problem in front of it.

RevOps spans six functions — sales ops, marketing ops, CS ops, GTM systems, strategy, and enablement. You won't be great at all of them, so build a full-funnel operator by rotating across them, ideally under a leader who moves you around.

A three-stage progression: prove genuine product-market fit once (a real problem, consumer-first), then achieve go-to-market fit by making the motion repeatable, then scale it into a 'dynasty' rather than a single lucky win.

Build around a real consumer problem and behavior, deliberately not leading with the novel technology (crypto/blockchain) or a get-rich-quick token.

A trading mechanic where a player offers unwanted items for a desired one; the system real-time-buys the target from one seller and real-time-sells the offered items to multiple buyers worldwide, netting a cashless swap.

Layering successive, lower-friction on-ramps into the economy — crypto, then credit card, then cashless quick trade, then an AI NPC negotiator — so each new rung pulls a broader persona into trading.

A staged handover of asset control from the studio to the player: start closed (assets exist but access is tightly controlled), move to shared responsibility ('you have a key, I have a key'), then to full player control (take it and go, even revoke the game's access).

When AI is integrated properly into how code is deployed and how engineers work, each person effectively gains a code reviewer, a junior programmer, and a security analyst — roughly tripling their output.

Using different AI models to challenge each other's work — e.g., writing a piece of code with Claude and having ChatGPT analyze it in the role of a security compliance officer.

Mythical's build sequence: (1) economic tech — changing the economies inside games; (2) social interaction — new ways for players to compete and play together (Pulse Arena tournaments); (3) open platform — letting outside studios build on Mythical's stack.

A reframe of investor rejection: a no usually reflects the VC's own vision or their LP-mandated 'deal box,' not a flaw in the idea — 'it's not a bad idea, it's just not the idea they have.'

A composite health metric combining the number of deals a rep works, the average deal size (ACV), the win rate, and the length of the sales cycle. Ebsta uses it to quantify the gap between top and average performers.

A seller who influences top of funnel, generates their own opportunities, and continues to own the relationship after the deal is signed — the opposite of the single-purpose-vehicle / hunter-farmer model where customers are handed from one specialist to the next.

A relationship-health score built from observable transactions — meetings, email traffic (inbound worth more than outbound), and call data (longer calls worth more) — deliberately excluding intent and sentiment analysis.

The difference between a firmographic, one-line ICP ('Series A–C startups') and a layered one that adds persona, buyer maturity, investors, and growth rate — and never confuses ICP with TAM.

Requiring every opportunity to carry written, scored qualification, with explicit gates and triggers to move from one stage to the next — and not allowing sellers to skip stages or self-score their own qualification.

The top-performer discipline of converting the fewest opportunities out of discovery on purpose — ruthlessly killing deals that won't close so time and resources flow to deals that will.

AI crawlers split into three types: retrieval bots (take a live user prompt, fetch your content, and pull the answer back — RAG-style), training bots (crawl broadly to train a model over time), and traditional index bots (classic search indexing like Google).

Reframing AI retrieval bots not as noise but as human buyers acting through a machine: the person sends the LLM to read your site on their behalf and receives the answer inside the chat.

Serve two versions of your website: the visually rich experience for humans and a separate, machine-optimized version for AI bots — clean markdown/JSON, schema markup, and supplemental content like extra FAQs — delivered by intercepting bot traffic at the CDN.

AI Engine Optimization (also called GEO) is the practice of optimizing how LLMs retrieve, represent, and cite your brand — a successor discipline to SEO that targets AI answers rather than search-result rankings.

Win the AI channel by creating content for long, specific, bottom-funnel prompts that only your brand (or a very few players) can credibly answer — instead of chasing high-awareness, top-of-funnel keywords.

A two-step operating loop for a brand-new channel: first establish a measurement baseline (which bots, which pages, which prompts, cited or not, versus competitors), then start making and testing changes to learn what moves visibility and conversion.

A growth-marketing foundation — merging the analytical, data-driven side (attribution, reporting, infrastructure) with the creative side (offers, channels, audience) — as the strongest stepping stone to running all of marketing.

The organizational cancer where no one is empowered to make a decision — approval chains, forms, and red tape turn every needed action into a slow, frustrating reaction that bleeds talent, customers, and revenue.

Jack's practical blueprint 'from the janitor to the CEO' for building a high-performance, innovative culture by empowering every level of the org to decide, take initiative, and add value.

Every role, from CEO to individual contributor, gets a clearly defined 'fence' — a sphere of influence — inside which they can innovate, fail, and decide without asking permission.

Empower frontline people to act in the customer's interest without asking permission — do what you'd do for your grandma — and keep asking 'how do we do better?' to surface and implement their solutions.

Culture compounds from what you reward, formally or informally. Reward outcomes and customer-proud work and you spin a flywheel of initiative; reward hours and optics and you manufacture performative productivity.

Model your own mind as a large language model: a database with a learning layer whose training data is the people you interact with. The more people you meet, the richer the dataset and the better your predictions, analysis, and instincts.

Because your brain-as-LLM has no safety team to filter bad inputs, deliberately curate what you consume: stay away from the noises of no's and stay close to the noises of yes, filling your dataset with people who believe things are possible.

Any capacity you stop using atrophies — a finger immobilized for months, eyes covered for a year, and the brain the same way. Offloading creative and critical thinking to AI quietly weakens the very muscles you depend on.

The outreach-spam era is a supply glut: pre-2019, ~50 companies chased ~100 US/Europe customers; the tech boom pulled ~500 sellers from across the globe toward a shrinking buyer base, so desperate automated outreach was inevitable.

There are only two ways to succeed: consistent, disciplined long-term work, or compromising ethically to appear successful overnight. Growth hacks are nonsense; you have to move through the natural process with patience.

Modern life moves people between three boxes — office, home, and club — which caps thinking and energy. Humans were designed for the outdoors, so escaping into nature once or twice a month restores perspective, energy, and ideas.

A VC framework Anthony relays: evaluate a startup on the wave (market), the surfboard (product), and the surfer (founder) — and the surfer matters most, because a great surfer finds the right wave even on mediocre equipment.

Treat your network as a decade-long asset instead of a job-hopping byproduct. Build history with people and do it with integrity, and those relationships become your most durable, lowest-cost pipeline.

A founder raises capital only three to five times in a lifetime while an investor does it every single day — so the founder is structurally the amateur. Closing that gap with structure, data, and network intelligence is the mission.

Flowlie's positioning: a behind-the-scenes operating system for a raise — not a marketplace, broker, or middleman — that helps founders uncover the right investors and the right people in their own network to reach them.

The two pillars of Flowlie: a predictive fit-scoring model (version five) that ranks how likely a firm or partner is to be interested, and a network-analysis engine that maps warm-intro paths and ranks each with a 'path impact score.'

The core fundraising philosophy: the outcome is decided mostly by the preparation — target lists, investor updates, relationship-building, and warm-path lining-up — that happens before you ever say you're raising.

Deliberately forward-loading warm-intro requests — scheduling connectors to introduce you weeks out — so investor meetings cluster into a single window instead of trickling in one at a time.

Valley's model for a full outbound loop: (1) find high-intent leads, (2) research them across 60–70 data points, (3) score them for ICP fit, (4) craft messaging in your voice, and (5) send it on LinkedIn autonomously at scale.

V1 outbound: take a filter-based list, enrich the contacts, drop them into an email or LinkedIn sequencer, and wait for meetings at the bottom of the funnel. V2 outbound: intent signals, warm outbound, and enrichment/research/qualification — send 30 hyper-relevant messages and book 10 instead of blasting 30,000.

A converting outbound message must make the recipient feel two things: relevance (it's clear why you reached out to me specifically, not the person next door) and investment (that real time was spent on me, not a one-to-many blast).

Person-level website-visitor identification is only valuable when paired with a layer that researches, personalizes, and sends. Intent should be weighted by visit frequency (repeat visitors over single visitors) and by high-intent pages like pricing, customer, and case-study pages.

Run Valley on autopilot (fully autonomous research, drafting, and sending) for broad, high-volume ICPs and transactional sales; run it as a copilot / human-in-the-loop 'AI SDR intern' (research and drafting, human approves send) for narrow, high-value, account-based motions.

Because LinkedIn caps monthly message volume, the only levers to improve outbound results are acceptance rate and reply rate — which forces teams to improve the relevance of their prospects and their messaging rather than sending more.

Because every public SaaS company answers to the same SEC rules, quote-to-cash should be a standardized, out-of-the-box process — not a uniquely engineered snowflake per company. A 'unique' process is a problem to fix, not a competitive advantage.

Instead of a separate CPQ, billing system, and revenue-recognition system integrated between CRM and ERP, run a single platform that handles CPQ, AR/billing, and ASC 606 rev rec — sitting between the CRM and the GL with no reconciliation and one product catalog.

An AI agent that lets any seller generate a compliant quote by typing a plain-English request into Slack (or mobile, email, or the CRM). The agent parses the request, asks for any missing policy-required inputs, applies product rules, and returns a quote PDF.

A business-focused Q&A layer that asks a seller simple questions (where is the customer located, what segment) and converts the answers into the right products, compliance, and discounting — instead of making the rep understand how the CPQ is configured.

Usage/consumption billing comes in distinct models: pure pay-as-you-go (no commitment, invoice on actual use), pre-committed plus overage (commit to a volume like 200,000 API calls/month, pay extra above it), and credit pools (buy a $100k pool and draw down across products, AWS/GCP-style).

A policy-driven method for turning variable consumption into a defensible ARR: for pay-as-you-go, take average consumption over a trailing 3-6 months and recognize a set percentage (e.g., 80%); for committed-plus-overage, the commitment is fixed ARR and overage recognition depends on how straight-line it is and what the auditor will accept (from ~95% down to ~20%).

When a customer adds licenses and renews early, you cancel the current term (crediting the unused period, like dropping a car lease) and restructure into a new term. Done right it's one opportunity, one order form, with credits and proration auto-calculated and reporting that shows it as upsell — not churn plus a new deal.

The design principle that the primary consumer of a CPQ should be the seller, not deal desk or RevOps. Reps should be able to run even complex deals (multi-year ramps, partner margins, special payment clauses) and the entire post-signature lifecycle themselves.

The seller creates an 'order' (draft during the sale cycle, confirmed once closed) and that same object generates the invoice and feeds finance. There's no separate quote-to-invoice re-keying, so numbers can't diverge between what sales sold and what finance bills.

A staged operating model for taking a company from nothing to a running revenue engine: first establish foundations and first principles, then instrument and stabilize, and only then layer in advanced and modern techniques (including AI) to sprint.

On joining, learn the existing systems by using and pushing them to their breaking point, then ship a working V0/V1 before soliciting input — collaborating afterward to fill in scope and context.

A CRM-based revenue-intelligence system resting on three pillars: (1) volume/activity — meeting depth and self-sourced pipeline; (2) accounts — tiering and account quality; and (3) accuracy/validation — clean, correctly-tagged data with automated backstops.

A single consolidated system — often an automated spreadsheet with 50-60 metric tiles rather than a visual 10-12-metric dashboard — organized in three levels: North Star KPIs (board/investor), functional KPIs (six to ten per team, in lockstep), and hyper-specific activity metrics.

Measure sales on the inverse of marketing's volume: only opportunities that pass a hard gate from discovery into 'prove value' count — deals genuinely closeable, and closeable within the quarter — and marketing's targets are pegged to that same gate.

Build custom, proprietary 'hubs' from scratch (e.g., in Replit) that solve a precise business problem and eradicate vendor spend; then transform their outputs into an agent-readable format (JSON) so agent 'spokes' (n8n, Manus, computer use) can run the downstream mission — with a human at the tail.

For any process, first ask whether AI can do the entire thing (path one: hardest but most efficient). If it can't be done cleanly, default to human-first with AI as augmentation (path two).

A GTM platform should be designed from the ground up as one system spanning data, engagement, and machine learning — not assembled by bolting point solutions together — the way a self-driving car is engineered whole rather than by strapping cameras and radar onto an ordinary vehicle.

AI augments the seller rather than replacing them. Duo, launched September 2024, is a human-in-the-loop companion — the rep's Pokemon or Iron Man suit — that learns each individual through reinforcement learning and grows with them.

Amplemarket is 'in the business of matchmaking' — connecting buyers who have problems with sellers who have solutions, so that every time a problem exists the buyer is made aware of the best possible solution.

In non-transactional, high-consideration buying, the purchasing experience — the craft, the care, the reverence for the product — is part of the value itself, and that care transfers to the buyer.

AI lets far more people build, so there will be more companies ('planets') to connect, each with smaller sales teams, and the space between them grows more opaque as creating information drops to near $0.

Every 24 hours the rep lands on a fresh feed of the most relevant accounts and buying signals in their book of business (the Spotify Daylist), paired with a recommended action for each — swipe the lead in or out (the Tinder system of action).

Low-quality, high-volume outbound is not a small positive but an active negative — it burns your domain, your leads, and your single chance at a first impression, signaling that your company doesn't care.

The highest-value trigger is timing — reaching a buyer when the problem you solve is already the last thing on their mind before sleep. You find that moment by composing signals (e.g., 100%+ team growth plus ten open AE roles) rather than relying on any single one.

The intelligence layer is a paradigm shift bigger than the internet, but it's raw power like the discovery of fire. Value comes only when you invent the 'pan' — thoughtful, purpose-built tooling — to cook something with it.

Vendors sell inside-out ('what I know, therefore you must know, therefore I'm best'), but buyers grade you outside-in through their own set of lenses — their pain, prior tools and research — most of which you can't see.

A B2B buyer moves from curiosity/pain through searches, forums, reviews and peers, forming impressions the whole way. Around 60–70% of the journey they build a shortlist of 3–4 vendors, and the best-impression 'number one' vendor at that point wins ~84% of the time.

Two obstacles to becoming number one: marketing complexity — the buyer is anonymous and you can't see when or where their journey starts; and sales complexity — by the time they engage they're highly informed on problem, solution and competitive landscape.

A large language 'thinking' model trained to think like a specific customer's buying committee, wrapped in a synthetically generated company, org hierarchy and pain scenario — a high-fidelity representative of your real buyer.

A buying journey of ~400–600 touch points is organized into discrete 'episodes' (a Google search, asking the community, exploring a vendor website, reading a testimonial); episodes roll up into a journey, and journeys attach to a buying persona.

A sales manager coaches to internal process ('you forgot to book the next meeting,' 'you skipped MEDDIC'); the buyer scores you on impression, not process. Dodging a pricing question the buyer just saw a competitor answer is what actually costs the deal.

AdamX's three products. TopJourney scores your buyer journey episode-by-episode against competitors; TopRep has the synthetic buyer analyze every recorded sales call for patterns; RolePlay lets reps practice live against the trained synthetic buyer.

The moments buried in call recordings where buyers express what they love ('sold on the solution,' 'this beats your competition hands down'). The synthetic buyer captures, de-dupes and ranks them, filtering out noise like 'thanks for waiting' or 'love your Zoom background.'

Reject the freshman-classroom model where every rep takes the same required courses. Coach like athletics: don't teach a strong server to serve — drill each rep's actual weakness (the 'backhand').

Your buyer journey score only matters relative to competitors in the same shortlist. An average journey wins if everyone else is worse; a good journey loses if a rival is better. Winning is measure-improve-measure-improve against the competition.

Reframe underperformance as a people problem, not just a revenue problem: focus on the individual rep and their manager, and stitch together the data (calendar, CRM, enablement) that reveals where each is struggling.

PeopleLens' four-step loop: (1) unify siloed rep-touchpoint, org, and people data into one connective tissue; (2) run proprietary models over structured and unstructured data; (3) render a persona-specific lens (exec, manager, rep); (4) push personalized performance nudges and agents to the front line.

The same underlying data rendered three ways — an exec lens for strategic bets and stack-ranking, a manager lens that diagnoses why a specific rep is struggling, and a rep lens that gives each seller a 360 view of their own outcomes, competencies, time allocation, and nudges.

For decades GTM data centered almost entirely on the customer (spouse's name, pet's name, endless fields). True first principles put the customer on one side, the product at the center, and the rep on the other — bringing the 'forgotten' rep into the equation with their own data lens.

Grow-or-go decisions are usually driven by anecdote in a QBR, not by facts about where a seller breaks down. The biggest, cheapest ROI is the 'massive middle' B-pool; because letting a rep go is roughly 18 months of revenue, personalized coaching that lifts the middle beats cutting.

Consolidate every revenue signal — email and calendar from the mail server, conversation intelligence from calls, and CRM history — into the Salesforce opportunity, account, lead, and contact records, rather than scattering them across ten systems.

A score, tracked over time, that aggregates communication frequency, depth, and stakeholder engagement across an account or opportunity to indicate the strength of the relationship and the likelihood the deal closes.

Use an organization's own closed-won and closed-lost deals to set benchmarks — time-in-stage, deal age, stakeholders per stage — then flag opportunities that deviate from what winning normally looks like.

Analyze call transcripts with AI to auto-populate a qualification framework (e.g., MEDDIC) — recommending a score per element plus supporting notes the rep can accept, edit, or ignore — without the rep manually entering it.

A composite score where 0 equals closed-lost and 100 equals closed-won; it should rise as a deal moves through the pipeline and reacts to all positive and negative signals mapped against a 12-month benchmark of won deals.

A selling discipline of always securing the next meeting while you are still in the current one, so an opportunity never sits without a scheduled next step.

Reps submit a data-backed forecast (pipeline / upside / commit) weekly; managers then submit their own adjusted view, hedging a rep's commit to upside when qualification is thin. Coverage ratios and pacing roll up by the Salesforce hierarchy.

Compare a rep's actual pipeline coverage (e.g., 6.8x) to the coverage they historically require to hit target (e.g., 3.6x) to decide whether they need more pipeline or should focus on closing what they have.

Treat the accuracy and structure of your underlying data — the ontology — as the foundation for any AI strategy, because AI is only as good as the data it can access, and swappable models matter less than the data feeding them.

Attio's product philosophy: the CRM should adapt to how your organization already does business, not force you to change your process to fit the tool.

By syncing your inbox and calendar on signup, Attio auto-builds your network of companies and people, enriches it, and layers on last-touch, contact ownership, and relationship strength — with no separate tool.

A custom record attribute powered by an AI prompt: you write your ICP in plain language and Attio evaluates every inbound lead against it, flagging fit for the rep.

Five standard objects plus unlimited custom objects and attributes, including Workspaces and Users objects that pull product data in, so the schema mirrors your actual business.

An automated workflow that, on every new signup, uses a research agent to summarize and ICP-tag the company, then routes: enterprise to round-robin, non-ICP to self-serve, and ambiguous mid-market/startup leads to a Slack channel for a human to route via buttons.

Attio is both where customer data lands and where you take action on it — you can report on live data, drill into the underlying records, and immediately sequence, task, list, or route them without leaving the tool.

Train a single AI model on a company's own data, then deploy it across every channel a buyer wants — chat, email, and voice — so context and quality carry seamlessly between modalities.

As more teams use AI to send high-volume 'fake personalized' outbound, each individual email becomes less effective — a network effect that runs in reverse, degrading the whole channel as adoption grows.

Remove the email-first form gate from website conversation. Provide genuine value and answer questions first, then weave pre-qualification into the conversation — 'give to get,' classic sales.

The specific questions a buyer asks are first-party intent data that reveals what they care about and are trying to solve — a 'roadmap to close' you can't buy from any data vendor and can only earn by opening the conversation.

Respond to buyers in real time in whatever channel they're using — chat, email, or voice — because the latency in most GTM motions is the human, not the medium. AI can reply within seconds and continue the buyer's exact conversation.

Run qualification logic live in the conversation and route each buyer to the right next step: a highly qualified buyer to an AE calendar, a mid-tier buyer to an SDR, a low-tier buyer to self-service.

Use AI voice around the human moment, not in place of it: after a buyer books a demo, an AI call gathers a few tailoring questions so the human demo is more valuable to the buyer and the rep is better prepared.

Train the model on data a company already has — website, docs, academy, good call transcripts, sales-training material — then refine it in a testing interface where you role-play your own customer and thumbs-up/down responses to tune tone and accuracy.

'Lean' should mean intentional, agile, right-sized structure for your stage — not the scrappy, disorganized, under-structured state most early teams actually describe when they say they're lean.

Import product engineering's agile operating system into RevOps — standups, definitions of done and ready, boards, user stories, QA and UAT stages — as the default way the team works.

A documented onboarding 'course' — tech stack, who-owns-what map, the agile working agreement, definitions of done and ready, systems, and roadmaps — that makes an incoming contractor or agency productive on day one.

Because RevOps has no fixed blueprint and fits differently into every company, assemble your function by borrowing proven patterns from more mature functions.

Divide the customer journey vertically into segments (four, from growth/brand marketing through sales, onboarding, CS, and support) and give each a product owner who obsesses over improving that stretch for customers, the company, and employees.

Dedicate a help-desk-and-comp role (backed by contractors) to absorb the daily end-user questions and recurring commission/quota cycles so developers and admins stay focused on the roadmap.

The recurring cycle where point tools proliferate around the CRM, category winners emerge and go vertical, the stack consolidates into a few big players — and then a new layer (now AI) fractures the ecosystem again.

Own the company growth model and go-to-market performance-to-plan — fully segmented, every way the business can be cut — as the source of strategic leverage that earns RevOps a seat in the room.

A short list of the company's top 'must-be-true' initiatives that the RevOps leader relentlessly surfaces cross-functionally — in every doc, roadmap, and prioritization call — to keep the whole organization aligned.

The career path out of the RevOps 'yes-too-much / no-too-much' trap: treat high-quality technical work as table stakes and win the next level on leadership — building a function that runs without you controlling every part of it.

Structure B2B support around the account as the centerpiece — its timeline, sentiment, history, and stakeholders — rather than around individual, disconnected tickets the way horizontal ticketing platforms do.

In B2B, AI's role is to assemble and surface the full context of an account — pre-sales data, call recordings, CRM history, previously-approved human answers — rather than to generate a single reply to a single question.

For technical, high-context B2B questions, AI should draft a documentation-grounded suggested response that a human reviews and sends — keeping a person in the loop instead of auto-replying.

Automatically convert existing Loom (and demo) videos into complete, screenshot-rich documentation, turning the thin, unowned docs AI draws from into high-quality source material — closing the data loop that makes AI answers good.

A workflow engine (triage, condition-based routing by time zone and ticket type) combined with AI-driven workflows (sentiment-based escalation, SLA-breach alerts) — the layer Tony argues actually constitutes a B2B support system.

Measure the business by revenue per full-time employee rather than by headcount hired or money raised. Top performers run $500K+ per head (versus an old $150–200K benchmark), driven by AI-leveraged operators.

Build marketing, brand, a reliable pipeline channel, and your own sales process before hiring a salesperson. Reps are harvesters of pipeline and closers — not creators of demand.

Land your first sales inside your existing network, then narrow to a hyper-specific micro-niche for whom the product is an absolute no-brainer, and make the economics the best deal of their lives early on.

In an AI-driven sea of sameness, brand generates demand. Aesthetics signal seriousness and a content strategy (written, tutorials, or podcasts) is the modern equivalent of commercials and billboards.

Start on HubSpot as an affordable, pre-built, scalable CRM; stand up Snowflake as the data warehouse for sales, product, and financial data; and report from there (e.g., Looker) rather than overloading the CRM.

Before buying any onboarding or CSP tooling, define exactly what first-time-to-value is for your product and sprint to reach it as fast as possible.

Let AI take work to roughly 90% and reserve the last mile for a human, so output sounds authentic and nothing goes out that doesn't resonate. The goal is producing better, not just producing more.

Treat go-to-market like health and fitness: track leading-indicator 'biomarkers' (onboarding speed, churn by segment, new-rep ramp, pipeline created, conversion) instead of reacting to lagging results after they break.

Each stage of growth — validation, product-market fit, product-channel fit, scale — is exponentially harder than the last, and you can lose product-market fit at every technology wave (on-prem to cloud, cloud to SaaS, SaaS to AI-native).

Design your offering as the thing you personally wished existed in your prior role, then scale the 'love' by hiring people better than yourself, guarding culture and integrity, and getting process and finances tight early.

RevOps is the business's family-clinic generalist — no single specialty, but a stream of problems from every function daily. Its job is to diagnose root causes by stepping into each function's shoes, not to treat the presenting symptom.

Take a reported symptom and break it into workflows and steps from first principles — for a conversion drop: lead source, count, region/quality, marketing activity, routing, scoring, and product pitch — then benchmark whether it's isolated (~20% of reps) or across the board.

Step 1: give the person comfort and let them talk (avoid seeding your bias). Step 2: validate the hypothesis quietly against the data in the background. Step 3: talk to other stakeholders of the platform, process, and functions to triangulate where the problem truly lies.

RevOps solutioning is a blend of people, process, and platform — never numbers alone. The revenue outcome can come through personal relationships, process, or systems, and usually a combination.

Four defenses that stop problems before they surface: (1) automation and AI to keep leaders out of low-value work, (2) learning and development so the team understands how the GTM machine fits together, (3) data hygiene with restrictive write-access to core systems, and (4) weekly/biweekly checks with real-time reports and fix-on-the-spot remediation.

A deliberate, agenda-less block of time spent exploring the data — the opportunity module, lead behavior, Slack signal — just to sense how the business is behaving, without a specific question to answer.

The two skills that carry a RevOps career: being a genuine people person who can build relationships with extroverted sellers and senior cross-functional leaders, and curiosity paired with a doer attitude — because the problems are new every day.

Sales velocity = (number of deals x average deal value x win rate) / time to close, expressed as a normalized dollars-per-day contribution per seller. The velocity delta is the multiple separating top performers from B/C players (11x in the 2025 report).

A view of the revenue motion where the left side is acquisition (lead to close) and the right side is post-sale retention and expansion. The insight: the right side must be multi-threaded and instrumented as deliberately as the left.

Compare the average number of days a deal spends in a stage when it wins versus when it loses. Once a deal exceeds ~14 days in a stage, win rate drops sharply; by four weeks it falls to about 5%.

Top performers close off roughly 30% of opportunities at the discovery stage, refusing to advance deals that were never properly qualified on budget, stakeholders, timeline, mutual close plan, and security/legal review.

Quantify what top performers do (e.g., six engaged stakeholders and a finance persona above a set engagement score by stage two), visualize it simply, and enforce those benchmarks as gates a deal must clear and triggers that prompt sellers and managers inside the CRM opportunity record.

A machine that connects to email, calendar, and phone systems to reconstruct every customer relationship, create and maintain CRM contacts, score engagement out of 100 (with trend and relationship-owner), and write it all back to Salesforce automatically.

Divide marketing and go-to-market into commoditized components (asset production speed, message scaling, channel operation, lead flow) that any marketer can run, and the non-commoditized secret sauce (the story you tell, one-to-one personal interaction, and how your audience feels) that is genuinely differentiated.

Technology makes the pipes that reach an audience faster and more efficient, but you and your team decide what flows through them. Because attention can't be bought — only earned — an efficient pipe carrying a boring payload still fails.

Q1: 'How do we feel about the way we brand ourselves and interact with the market today?' Q2: 'What do you really want to be known for?' The gap between leadership's private answers and what the market actually perceives defines the brand problem to solve.

A single-page document capturing 'this is who we are, this is how we want to be perceived in the market, and this is what we want people to know about us,' agreed by everyone with a stake in the brand before any external-facing work begins.

Three barriers keep companies from investing in brand: (1) trust — it's hard for a founder to let a newer leader or agency lead a transformation of their baby; (2) the brand-vs-growth tension — a ~6-month brand bet competes with demand channels that pay back today; (3) scar tissue from agencies that over-promised and under-delivered.

Bring in an agency when (1) they have specific expertise you lack in-house and you want to capture that knowledge, (2) you simply need helping hands and bandwidth you can't hire fast enough, or (3) you need a credible third party to validate a strategy you're pitching to an executive or founder.

You cannot look like, market like, act like, and operate like every other company and expect outsized (10X) results. The size of the outcome you want dictates the size and non-obviousness of the investment you must be willing to make.

Split marketing spend into three distinct buckets: a one-time brand revamp (tuning the 'Formula One car'), ongoing brand marketing (billboards, out-of-home, high-funnel video with no direct-response path), and demand gen (PPC, SEO, conversion-driven paid social, outbound SDR).

Spend nearly all your budget on demand while more dollars still pay back through payback period, CAC, and CLV:CAC. When additional demand spend stops producing proportional value — you've maxed the intent that exists in the market — that's the signal to invest higher-funnel in brand to grow total searches and eyeballs.

After a successful brand overhaul, three things are true every time: (1) internal excitement spikes — employees advocate and share; (2) recruiting efficacy jumps — inbound interest in working there rises; (3) core marketing metrics get more efficient — more direct traffic, higher conversion, higher funnel velocity and close rate.

Outsized outcomes come from luck multiplied by preparation. You can't will the lucky break, but you don't get the fortuitous opportunities without having done the work that makes you ready to seize them.

The replacement for servant leadership: lead from the front, believe no job is too small, and do the high-context work yourself instead of managing away from it.

A lens (adapted from Ben Horowitz's peacetime/wartime CEO) that treats the last 12 zero-interest-rate years as peacetime — stable, predictable, growth-at-all-costs — and today as wartime, defined by speed, precision, and survival.

A change-management ritual: every team member keeps a Post-it on their main screen that reads 'How can AI help me do what I'm about to do?' — retraining individual behavior before restructuring the org.

The missing middle between micromanaging and absentee leadership (credited to Rippling COO Ian McInnis): get close to a work stream to build context and coach, then step back and grant autonomy once you see consistency.

Michael's operating equation for trust: consistency over time equals trust. You earn the right to grant autonomy by observing consistent delivery, not by title or tenure.

Jim Collins's Good to Great bus metaphor (get the right people on the bus, in the right seats), extended with Michael's addition: you must design the seats themselves — the actual jobs — not just fill them.

A mentor's rule that a new leader has roughly 100–120 days to make their people and structure decisions; after that window, the team's output is the leader's own fault or benefit.

Taste is the human judgment to know whether AI's output is actually good. AI takes prompts and shows you a thing; determining if that thing is good is a nuance and sophistication AI doesn't have.

A model for what to outsource: well-defined work is a neatly wrapped present you can hand to an agency (or junior talent); ambiguous, high-context work is a plate of spaghetti where the noodles are snakes and you need the plate back.

Career market fit is the idea that the market may see your value more clearly than you see it yourself (Michael leads go-to-market but the market thinks of him as a marketer). Paired with it: know the neighborhood you're heading toward, not the exact destination, and take any avenue pointed that way.

A three-layer operating system: quarterly OKRs with an above/below-the-line priority cut and built-in slack time; two-week to-do/doing/done sprints with a Monday plan, Friday check-in, Thursday review + retro, and a Friday 20% block; and a monthly company-wide all-hands to prove what shipped.

A phrase from Rita McGrath's Seeing Around Corners: those closest to the work make the best, fastest decisions, while decisions made far from the work are colder and slower.

A research-derived methodology that distinguishes two selling environments — low-complexity/transactional and high-complexity/enterprise — and, in the complex environment, wins by Teaching the buyer something new, Tailoring it to their situation, and Taking Control from an advisory, expert frame.

The Challenger research profiled sales-rep archetypes and measured which won. Weiss names four: the Hard Worker (outworks everyone), the Challenger (teaches and pushes change), the Relationship Builder, and the Lone Wolf.

Neil Rackham's 1970s discovery framework: understand the buyer's Situation, the Problems inside it, the Implications of not solving them (cost of inaction), and the Need-payoff of solving them, to drive urgency around why change now.

A riff on SPIN designed to combat product-led, feature-and-benefit selling by first understanding the buyer's problems before offering a solution — best used to help a buyer see a problem they didn't know they had.

The discipline of reading where a buyer is in their own process and matching your motion to it: a mature buyer knows the problem, the solution criteria, and the competitors; an immature buyer knows little and needs co-creation.

A holistic system organized around engaging the buyer, qualifying, and closing — including 'upfront contracts' (pre-agreeing to a next step) and 'pain funnels' that probe first-, second-, and third-level pain.

The view that sales mastery is the deep execution of a full set of fundamentals — hunting, reaching decision-makers, discovery, creating needs, business-case building, presenting, multi-threading, negotiating, handling rejection — assembled in your own way, rather than any single methodology.

Gartner's framework, rooted in research on buyer decision fatigue and decision confidence, in which the seller's job is to help an overwhelmed buyer make sense of an overload of information and competing options so they can decide with confidence.

The time it takes a new user to 'get it' after logging in — a core PLG success metric Ocean actively drives down by putting the product's aha moment directly on the landing page.

Vectorize both companies (65M) and LinkedIn profiles (230M), then combine them in one search: input an example person's LinkedIn handle and find lookalike people, by role and context, inside the lookalike companies of a target account.

Target by a contextual understanding of what an individual actually does for a company, not by their title — because titles vary with company size (CMO vs. head of growth vs. VP marketing) for the same real role.

Build, filter, and preview the target list in Ocean without spending a single credit; export to Clay for enrichment only once the list is validated.

Gen 1 is an LLM wrapper around an analog/non-normalized database — a pretty face on messy data with broad, imperfect targeting. Gen 2 models the actual GTM process and automates the flow end-to-end, with human validation between steps.

Automation's real strength is micro-targeting: overlay intent and third-party data on a tightly defined audience so every message is highly relevant, producing 5–10% conversion instead of 0.1%.

A build-from-scratch playbook for customer success operations: (0) Breathe and triage for impact; (1) Learn the lay of the land — roles, journey, and where time goes; (2) Bring in the right tech once the process is aligned; (3) Build KPIs and a customer health index; (4) Use the data to drive decisions; (5) Stay connected to the customer.

High product adoption and green dashboards do not guarantee a healthy customer. A single metric (logins, courses created, items assigned) measures activity, not the value the customer is actually extracting.

Automated Salesforce alerts that fire at six, three, and one month before a renewal date, pinging the right people to confirm conversations have started, questions have been asked, and adoption is on track.

Three gates that determine when a company is ready to buy a Customer Success Platform: (1) established processes for outreach, QBRs, and handling at-risk vs. healthy accounts; (2) trackable product-usage data (e.g., via Snowflake or a BI tool); and (3) an inability to stay proactive by hand at leadership's bar.

A composite health score that aggregates multiple signals — support (first-response and resolution times), satisfaction (NPS/CSAT), adoption, and usage — rather than relying on any single isolated metric.

The core BCG method: read a broad problem statement closely to extract its keywords and clues, branch into a small set of hypotheses using judgment and calculated guesses, then validate or nullify each with data, experiments, and conversations — under real time and resource constraints.

Treat the stated problem as a symptom. Bring the right functional owners into the room, go deliberately broad first, and pull several years of historical data so the true root cause reveals itself layer by layer before you narrow.

Before analyzing, classify the work as either a backward-looking diagnostic (what went wrong?) or a forward-looking strategy question (how do we grow or break into a new segment?). The mode changes which hypotheses you form and how much value the analysis returns.

A small-and-mighty RevOps team protects strategic bandwidth by interrogating every meeting invite, pushing back on low-value asks, leaning on leadership for air cover, and delegating only when it serves the team — freeing time (and AI-reclaimed minutes) for deep thinking.

There are seasons to accelerate the business and seasons to maintain — to hold the speed limit rather than push the gas. Sustainable performance requires knowing when to brake, because it's genuinely hard to stand still and all-gas/no-brakes leads to disaster.

GTM Fund's early-stage evaluation model. The wave is the macro trend / 'why now' (falling AI costs, regulatory tailwinds, distribution shifts); the surfer is the founder's skill, vision, and tenacity; the surfboard is the product — important but the most flexible because it evolves.

A deep, non-obvious insight into a problem space that gives a founder an unfair advantage. It comes from either lived experience (having operated in the space and felt the pain intimately) or obsession (diving so deep into the problem you discover truths others miss).

A way to read early traction that ignores headline revenue in favor of predictive signals: concentrated, evangelical customers; enterprise validation; founder-led sales; usage depth and retention; and shipping velocity. The real question is never 'how much revenue?' but 'does this traction predict you'll find product-market fit?'

Fundraising is go-to-market pointed at investors. How a founder runs the raise — target lists, warm intros, tailored pitches, a disciplined intro-to-close funnel — is treated as direct evidence of how they'll run sales, partnerships, and customer acquisition.

Sophie's personal method for finding fulfilling work: reflect on what you keep returning to with curiosity and what consistently energizes you, write it down, look for patterns, and identify the two or three core forces (a 'triangle') that keep pulling you back. Aim for a role at the center of all of them.

Luster's core operating loop: first diagnose proficiency at the atomic skill level, then predict where a lack of proficiency is about to impact performance in the next 24–48 hours, then prescribe the specific practice or content to close that gap in real time.

The principle that you must objectively measure a team's competency gaps before deploying any learning, development, or training — otherwise the enablement is a waste of time and money.

The failure mode of the consultant-led skill audit: after three-to-four months and hundreds of thousands of dollars analyzing the team on poor CRM data and self-reported interviews, the firm 'plops' a diagnosis with no mechanism to fix it — and it's already last quarter's problem.

Grounded in behavioral and cognitive psychology, Luster offers two practice modes: full-call simulations that mimic an entire sales conversation (prospecting, discovery, QBR, proposal, negotiation), and isolated skill drills with a built-in AI coach that repeatedly tests one skill such as objection handling.

Two ways to build an AI product. 'Quick tech' is a user interface layered on a single shared LLM instance — fast to demo, but unable to control data sharing, latency, or per-customer context. The 'platform' approach builds a trained, closed-off instance per customer behind proprietary layers, trading feature speed for control, security, and stability.

Luster's proprietary stack that sits between the raw LLM and the user interface. Layered bottom-up: a per-customer trust-and-security layer, a custom ingestion model of the org's people and behavior, a company-specific insights/persona/goals layer trained on first-party plus third-party web data, a conversational-AI layer (latency, personality, context), and an output layer of predictive skill insights and prescribed actions.

Instead of sending from a massive shared server (a shared IP pool) full of thousands of unvetted senders, give each user an isolated mini-server ('cluster') with its own IP address so one sender's behavior can't affect the others.

Regularly send emails from a customer's mailboxes to known reference mailboxes to observe where they actually land — inbox, spam, promotions, or undelivered — as the true indicator of email infrastructure health.

Simulate natural, two-way activity across real corporate mailboxes to balance the unnaturally low response rates of cold outreach, so email service providers don't flag the account.

Cap daily send volume per mailbox to what platforms now tolerate (15–25/day, down from hundreds), have the platform control the cap rather than the rep, and scale volume only after messaging is validated on a small sample.

Treat cold outreach like a paid-ad platform: give the system many message variations, test each against a small subset of the audience, and scale only the versions that generate positive engagement.

Use reinforcement learning — a distinct branch of AI from LLMs — as the optimization layer that looks at what has and hasn't performed to predict which hooks, lead magnets, and offers will resonate, and recommends new variations over time.

Blend autonomous AI (which ingests large data sources) with human review checkpoints — a sales rep reviews certain AI-generated copy before it reaches a prospect, and an admin reviews certain content before it reaches the rep.

A maturity model for the function. RevOps 1.0 is the tactical, reactive service center — implementing the tech stack, formatting sales calls, planning territories, comp plans, and CS playbooks, and managing requests. RevOps 2.0 is an internal management consultant that participates in corporate planning, sits shoulder-to-shoulder with finance on the board plan, and leads with insights and recommendations.

A planning discipline in which the analyses behind each metric, the plan assumptions themselves, and actual performance against those assumptions are all kept visible and updated in real time — rather than being computed once for the annual plan and filed away until the next board meeting.

The set of five or six drivers a well-built revenue waterfall contains — normalized prospect volume, sales cycle, time-based conversion distributions, close-won production, SQLs and MQLs — that you should have a pulse on at all times and be able to segment 20–30 ways (enterprise vs. SMB, region, product line, service center).

Use ChatGPT as the technical collaborator you turn to when a human peer is unavailable — paste a broken formula or a stuck problem and get an immediate diagnosis and a testable fix.

Describe a Salesforce business rule in ordinary human language and let ChatGPT translate it into the validation rule or formula, then review the output for business context before saving.

Paste an existing, complex formula and ask ChatGPT 'what does this formula do?' to get a plain-English explanation you can understand and pass on to others.

A mental model for reacting to disruptive technology: history doesn't repeat but it rhymes, and past waves (like email) augmented and grew work rather than eliminating it.