The LeanScale Podcast · Episode 91

Why Outcome-Based Pricing Is a Trap for Most AI Companies

Roee Hartuv on pricing & packaging, the jobs-to-be-done approach, and how AI broke SaaS unit economics

Roee Hartuv · Pricing & Packaging Advisor (ex-Winning by Design) · Winning by Design Hosted by Anthony Enrico
Published Updated 00:41:09 30 min read 6,071 words
Pricing & Packaging Consumption Revenue AI in GTM GTM Strategy
▸ Listen on Spotify Watch on YouTube Apple Podcasts
SummaryTakeawaysFrameworksQuotesAdviceAIEntitiesMetricsFAQTimelineTranscript
Executive Summary

The one-paragraph brief, extended

Why this conversation matters — and who should spend the hour.

Outcome-based pricing is the buzzword of the year — walk into any RevOps conference and three speakers will tell you it's the future of B2B SaaS. Roee Hartuv, a pricing and packaging advisor based in Berlin who has redesigned monetization inside B2B SaaS and AI companies on both sides of the Atlantic, thinks it's a trap for almost everyone chasing it. He spent nearly four years at Winning by Design — the firm that pioneered the Bowtie model — leading the revenue architecture team and teaching its flagship course to operators worldwide. About eighteen months ago he made a sharp call: of every GTM lever an operator can pull, pricing and packaging produces the fastest and biggest results, so he walked away from being a generalist and now works on nothing else.

The spine of the conversation with LeanScale co-founder Anthony Enrico is an order of operations most founders get backwards. You start with packaging — how you bundle features around the customer's jobs to be done, not around a product team's ranking of which features get used most. Then you move to pricing structure, then the pricing metric (what you charge on), then, dead last, the price point. In B2B, Roee argues, the number matters least: a good seller wins or loses on value, not on whether the deal is $12,000 or $13,000. What actually governs the design is the 'complexity budget' — your ACV determines how much complexity your pricing can carry, so an AWS seven-figure motion and a startup package live in completely different worlds.

The middle of the episode is a clinic on how AI broke SaaS pricing economics. Classic SaaS ran on near-zero marginal cost and 80–95% gross margins; AI reintroduces a real cost to serve that most SaaS-native companies don't even have the tooling to measure. Roee tells the cautionary tale of a bootstrapped $40M-ARR market leader that launched AI on a seat-based model in November and was 'losing money like crazy' by February — the kind of quiet bleed that follows any company that bolts AI onto seat-based packaging without re-pricing to follow cost.

That pressure pushes companies toward usage-based and then outcome-based pricing — and this is where Roee draws his sharpest line. Outcome-based pricing works when the outcome is uniform and definable across a vertical (his canonical example is Fin, which charges per closed support ticket priced against ~20 minutes of a support manager's time). It's a trap for horizontal products: the same ChatGPT or Claude subscription is worth a Google search to his mother and a two-week financial model to him, so 'you can't price on different outcomes.' His three-question test — can you define one outcome, will the customer agree to that definition, and is the value the same across your base — decides fit. When outcomes don't qualify, the answer is disciplined usage-based pricing made palatable to CFOs with a recurring base fee plus usage on top, real-time visibility, spend controls, and per-token metering.

The close is a practical repricing playbook: a validation journey that de-risks moving your most important customer by first testing the new model on your least-important segment and net-new business, then running validation interviews and internal alignment before migrating anyone. And the meta-point — pricing and packaging is now a living product surface, not a five-to-seven-year set-and-forget decision. Who should listen: founders, CROs, CFOs, product leaders, and RevOps operators who suspect their pricing is leaving money on the table or quietly losing it, and want a grounded framework instead of the LinkedIn outcome-based-pricing hype.

Key Takeaways

12 things worth stealing

The load-bearing ideas, each with the business implication and who should care.

01

Pricing and packaging is the most overlooked growth lever — and it's usually addressed far too late

Roee gave up being a generalist because pricing and packaging consistently produced the fastest, biggest results of any GTM lever. Most other levers depend on humans and take time; fixing pricing unlocks the whole customer journey — acquisition, onboarding, renewal, expansion — and touches go-to-market, product, and operations at once. Yet most companies don't hire a dedicated pricing expert until roughly $500M ARR, leaving it part-time to a CRO, CFO, or product marketer before that.

Why it matters: Treat pricing and packaging as a first-class discipline early, not a founder side-project you revisit under duress. Starting years earlier is one of the highest-return moves an operator can make.

FoundersRevenue ExecutivesRevOps Leaders
02

Fix the order of operations: packaging first, price point last

The sequence Roee runs is packaging → pricing structure → pricing metric → price point. Founders instinctively jump to the number, but the number is the least consequential decision. You first decide how you bundle value, then how you structure the model, then what you meter on (and whether you charge forward- or backward-looking), and only then the price.

Why it matters: If pricing feels broken, don't start by changing the number. Re-examine packaging and metric first — that's where most of the leverage and most of the friction actually lives.

FoundersRevenue ExecutivesRevOps Leaders
03

Package around jobs to be done, not feature-usage rankings

The classic packaging exercise is a product-team ranking: sort features by how many customers use them, put the top ~60% in Basic, the next ~30% in Better, and the premium ~10% (AI, advanced analytics) in Best. Roee's approach bundles around what the customer is trying to achieve. Everyone claims to do value selling, but if packaging isn't aligned to jobs to be done, sellers inevitably fall back to talking about features instead of the customer.

Why it matters: Redesign tiers around customer jobs so a buyer can point to 'that's the job I need' and self-select. It removes most of the scoping friction from the sales process and the renewal.

FoundersRevenue ExecutivesRevOps Leaders
04

In B2B, the price point is the least important lever

Unlike B2C or retail where every dollar can move conversion, in B2B a deal rarely turns on whether it's $12,000 or $13,000. A strong seller who leads with value will close at 12, 13, or even 15. That's why Roee ranks the price point last of all the pricing mechanisms.

Why it matters: Stop over-indexing on finding the 'perfect' number and invest that energy in packaging, metric, and value articulation — the decisions that actually move win rates and expansion.

Revenue ExecutivesFounders
05

Respect the complexity budget — your ACV caps how complex pricing can be

Every business has a 'budget of complexity' set by who it sells to. Higher ACV correlates with larger customers who can absorb more complexity, so an AWS seven/eight-figure enterprise motion can carry an intricate pricing scheme. A startup with low ACV needs a dead-simple buying process — hence startup packages. Selling to multiple segments means running multiple levels of complexity, which is itself a packaging problem.

Why it matters: Match pricing complexity to segment. Over-engineering pricing for small buyers kills velocity; over-simplifying for enterprise leaves value uncaptured.

FoundersRevenue Executives
06

AI broke SaaS unit economics — cost to serve is now a line item most teams can't measure

Classic SaaS had near-zero marginal cost (build once, run on cheap compute/storage) and 80–95% gross margins. AI reintroduces real cost to serve — some peg AI-company margins near 50%, and frontier labs like OpenAI run on average-negative margins. In SaaS, teams only tracked top line; now there's a third line item most SaaS-native companies lack the tooling to even measure, and often can't tell you the unit economics of a given customer type.

Why it matters: Before repricing, stand up the operational and product instrumentation to measure per-customer cost to serve. Pricing that doesn't follow cost — and cost is usage — will lose money down the road.

FoundersRevenue ExecutivesRevOps Leaders
07

Bolting AI onto a seat-based model quietly bleeds money

Roee's live example: a bootstrapped, profitable $40M-ARR category leader introduced AI on a seat-based model and launched in November. By February they called in a panic — 'we're losing money like crazy, we need to stop the bleeding now.' The fix was re-pricing to follow the cost of AI. Not every user is unprofitable, but the unprofitable ones scale exponentially.

Why it matters: If you've added AI capabilities to seat-based packaging, assume you're leaking margin and audit it now. Peg the AI-driven cost to usage before the losses compound.

FoundersRevenue Executives
08

Outcome-based pricing fits vertical, uniform outcomes — and traps horizontal AI products

Outcome-based is what you price on; usage-based is how — you can combine them (Fin charges per closed ticket, an outcome billed by consumption). It works when you sit in a specific vertical and the outcome is basically identical across customers. It fails for horizontal products: the same subscription is a near-free Google-search substitute for a casual user and a two-week analysis replacement for a power user. You literally can't price on outcomes that mean different things to different users.

Why it matters: Ask whether you're horizontal or vertical before you attempt outcome-based pricing. If horizontal, don't try to force it — it's why the buzzword is so rarely executed well.

FoundersRevenue Executives
09

Use the three-question test to decide if outcome-based pricing fits

Roee's qualification: (1) Can you clearly define one outcome across your customer base? (2) Can your customers agree to and accept that exact definition — including edge cases like a ticket reopening? (3) Is the value of that outcome similar across all customers? Only if all three are yes are you a candidate. He notes he's seen very few companies do outcome-based pricing well, and hasn't yet seen one package multiple distinct outcomes successfully.

Why it matters: Run every 'we should do outcome-based pricing' idea through the three questions before designing anything. Most agentic-AI pitches for outcome pricing fail question one or three.

FoundersRevenue ExecutivesRevOps Leaders
10

Make usage-based pricing palatable to CFOs with a base fee plus guardrails

The core downside of usage-based is that customers hate unpredictability — 'bill shock' is a real term, and CFOs are scared of it. But the industry is heading there anyway. The best mechanism is a recurring base fee with usage on top: the base gives both sides predictability, and the usage portion can be capped, notified, and controlled. Add real-time visibility, per-team spend controls, and token-level metering so every charge is traceable — because the first big invoice will trigger an audit.

Why it matters: Don't sell raw metered usage. Wrap it in a base commitment, spend controls, real-time dashboards, and airtight billing lineage so both the customer and your investors get the predictability they need.

Revenue ExecutivesRevOps LeadersFounders
11

Reprice your biggest customers through a staged validation journey

The end goal — migrating your most important (usually biggest) customer to new pricing — is de-risked by working backward. Test the new model first on your least-important segment or a new market (e.g., a European company entering the US, where there's little to lose), then on new business, then run validation interviews with your second-through-fifth most important customers ('we're not migrating you, we just want your thoughts'). Layer in internal validation with the sales team who weren't in the design room. Test, iterate, and build confidence before touching the anchor account.

Why it matters: Never reprice your marquee account cold. Sequence the rollout from lowest-stakes segments inward so you arrive at the big migration with evidence, not hope.

Revenue ExecutivesRevOps LeadersFounders
12

Treat pricing as a living product — revisit it every product or sales cycle

Gone are the days you set pricing when you build the company and don't touch it for five to seven years. Because products now ship value weekly, pricing and packaging should be treated as product itself — continuously improved, not overhauled every year on a whim. Roee's cadence: revisit at each product-release cycle or each sales cycle, whichever is longer (a six-month sales cycle implies a six-month pricing review).

Why it matters: Put pricing on a standing review cadence tied to your release and sales rhythms. But don't thrash — every change carries a mental burden on customers, so iterate deliberately toward a goal rather than flip-flopping between models.

FoundersRevenue ExecutivesRevOps Leaders
Frameworks Discussed

9 named models

Every framework Jimmy names, defined and time-stamped.

The Pricing Order of Operations

06:18

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.

Founders instinctively start with the number, but Roee argues it's the least important decision. Getting the sequence right forces you to resolve how you bundle value and what you meter before you argue about dollars — which is where the real leverage and friction live.

Jobs-to-Be-Done Packaging

04:38

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

The classic method ranks features by usage and slots them into Basic/Better/Best. Roee's method organizes tiers by customer jobs so buyers self-select ('that's the job I need'). Without this alignment, sellers who claim to do value selling always fall back to features — and scoping friction returns.

The Complexity Budget

08:12

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.

AWS selling seven/eight-figure enterprise deals can run intricate pricing; a startup buyer needs the simplest possible purchase. Selling to multiple segments means running multiple complexity levels at once, which itself becomes a packaging decision.

How AI Broke SaaS Cost-to-Serve

10:46

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.

SaaS teams only ever tracked top line (marketing and development). AI adds a third line item — cost to serve — and unprofitable customers scale exponentially. Introducing AI is therefore an operational and product exercise, not just a pricing one: you have to build the instrumentation to measure per-customer unit economics.

Outcome vs. Usage; Horizontal vs. Vertical

17:50

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.

Fin prices per closed ticket (an outcome billed by consumption). But a horizontal assistant is a near-free search for one user and a two-week analysis for another, so a single outcome-based price can't capture value fairly. This is why the buzzword is rarely executed well.

The Three-Question Test for Outcome-Based Pricing

20:42

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.

Question two forces resolution of edge cases (does a reopened ticket still count?). Question three is why tickets work — a closed ticket maps to roughly 20 minutes of a support manager's time, a value you can put a price on. Fail any question and outcome-based pricing won't hold.

Base Fee + Usage (CFO-Friendly Usage-Based Pricing)

26:36

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.

Customers hate usage-based pricing because of unpredictability ('bill shock'). The base fee gives both sides a predictable floor for forecasting; the usage layer is bounded by guardrails. Full metering lineage is non-negotiable because the first large invoice triggers a customer audit.

The Validation Journey (Repricing Without Losing Customers)

34:41

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.

Roee frames it as risk mitigation toward a fixed end goal. Testing in a low-stakes market (e.g., a European company's new US entry) and telling customers 'we're not migrating you, we just want your thoughts' builds evidence and confidence so the marquee migration doesn't lose face — or the customer.

Pricing as Product (Revisit Cadence)

35:36

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.

Because products now ship value continuously, pricing must reflect that pace. Roee's rule: a six-month sales cycle implies a six-month review; a new feature set is a trigger to ask whether you should reprice. But change deliberately — each change imposes a mental burden on customers.

Best Quotes

17 lines worth clipping

Pulled verbatim. Copy or share any of them.

“Why outcome-based pricing, despite being the buzzword of the year, is a trap for most horizontal AI companies.”
Anthony Enrico 00:44
“I do believe that pricing and packaging is like a magic growth lever we have, and it is mostly overlooked.”
Roee Hartuv 02:18
“If they have this intuition that their pricing and packaging is not working, where do you even begin? I begin with packaging.”
Roee Hartuv 03:50
“You can say that you're doing value selling, but if your packaging is not aligned to values and jobs to be done, then at the end of the day, you will always fall back to talking about features and about the product itself and not around the customer.”
Roee Hartuv 05:24
“In B2B, it doesn't matter for our customer if it's 12,000 or 13,000. We won't win or lose the deal. It's not like retail or B2C where every dollar could have an impact.”
Roee Hartuv 06:42
“We have a budget of complexity. The higher your ACV, which correlates with the type of customers you're selling to, the more complexity you can add into the sales process or your pricing.”
Roee Hartuv 08:12
“In SaaS we only cared about top line. All of a sudden we're introducing another line item, which is cost to serve — and SaaS-native companies don't even have the tools to measure that.”
Roee Hartuv 13:05
“They were a profitable bootstrap company. Since we did this launch, we're losing money like crazy. We need to stop the bleeding now.”
Roee Hartuv 14:26
“The outcome for me and the outcome for my mother were both getting the results we're looking for. But for my mother, it's a Google search, almost free — and what I'm doing would have taken me two weeks of work.”
Roee Hartuv 17:41
“When you're selling to a horizontal segment, outcome means different things for different users. And how can you price based on different outcomes? You can't.”
Roee Hartuv 18:27
“Can you clearly define it as one across your customer base? Can the customer agree with you and accept that as what you're pricing on? And then, is the value the same across your customer base?”
Roee Hartuv 21:31
“Everybody wants to say, hey, when you introduce agentic AI you should price based on outcomes. Great — show me a company that does it well. I haven't seen that much.”
Roee Hartuv 22:17
“People hate to buy usage-based products because it's not predictable. Bill shock is a real term, and CFOs are really scared of it.”
Roee Hartuv 23:59
“One of the best mechanisms is to create some sort of recurring base fee and usage on top of that.”
Roee Hartuv 26:36
“Can you find yourself — which jobs to be done describes you the most? If they're able to pick without any hesitation, that's the best result we can find.”
Roee Hartuv 30:34
“Gone are the days that you set pricing and packaging when you build a company and then don't revisit it for five to seven years. We have reached a point where pricing and packaging should be referred to as product.”
Roee Hartuv 35:36
“If you can nail this, you can make your customers happy while also extracting the right value that makes sense for what you're offering.”
Anthony Enrico 38:13
Practical Advice

What should you actually do?

The playbook, split by the seat you sit in.

Founders

  • Start pricing work at packaging, not the price point — bundle around your customers' jobs to be done so buyers can self-select the tier that matches the job they're hiring you for.
  • If you've bolted AI onto a seat-based model, assume you're bleeding margin: instrument per-customer cost to serve and re-peg pricing to usage before the unprofitable accounts compound.
  • Don't wait until $500M ARR to take pricing seriously. It's the fastest, biggest GTM lever and the earlier you build the discipline, the more it compounds.
  • Treat pricing and packaging as product: put it on a standing review tied to your release and sales cycles rather than setting it once and leaving it for years.

Revenue Executives

  • Run every 'we should do outcome-based pricing' idea through the three-question test — one definable outcome, customer agreement on the definition, and uniform value across the base — before designing anything.
  • Know whether you're horizontal or vertical. If horizontal, don't try to force outcome-based pricing; design disciplined usage-based pricing instead.
  • Stop optimizing the number. In B2B the price point is the least important lever; invest in packaging, the pricing metric, and value articulation that actually move win rates.
  • Reprice your marquee account last: sequence the rollout from least-important segments and new business inward, using validation interviews and internal sales alignment to arrive with evidence.

RevOps Leaders

  • Make usage-based pricing CFO-friendly with a recurring base fee plus bounded usage, real-time spend visibility, per-team controls, and caps or notifications.
  • Build airtight metering lineage — trace every token to who consumed it, when, and how much — because the first large invoice will trigger a customer audit you must be able to answer.
  • Match pricing complexity to segment ACV (the complexity budget); if you serve multiple segments, expect to run multiple levels of complexity and manage them through packaging.
  • Stand up the cost-to-serve instrumentation SaaS-native stacks usually lack, so pricing decisions can actually follow unit economics.
AI Takeaways

How AI actually changes GTM

LeanScale's signature read on the AI-in-GTM question this episode wrestles with.

The thesis

AI didn't just add a feature to SaaS — it broke SaaS pricing economics. Near-zero marginal cost is gone, so pricing must follow usage-driven cost to serve. Outcome-based pricing is a trap for horizontal AI products; the durable path is disciplined usage-based pricing wrapped in predictability, with outcome pricing reserved for narrow verticals with a uniform, definable outcome.

Cost to serve is back

AI reintroduces a real marginal cost most SaaS-native companies can't even measure. If your pricing doesn't follow usage-driven cost, you'll lose money as unprofitable customers scale exponentially.

Seat-based + AI = margin leak

Bolting AI onto seat-based packaging is the single most common trap — a profitable $40M bootstrap went from profit to 'bleeding' in three months by doing exactly this.

Outcome-based is a horizontal trap

The same AI usage is worth a Google search to one user and a two-week analysis to another, so a single outcome price can't capture value. Only uniform-outcome verticals (per closed ticket, per resolution) qualify.

Usage-based is the destination

Customers hate unpredictability, but the industry is going to usage. A recurring base fee plus bounded usage, real-time visibility, and per-token metering makes it survivable for CFOs and investors alike.

Pricing is now a living product

AI ships value weekly, so pricing must be revisited on a release/sales-cycle cadence rather than every five-to-seven years — Claude itself changes its pricing and promotions almost weekly.

Agent & automation ideas

  • A cost-to-serve instrumentation layer that attributes token/compute spend to each account, segment, and feature so teams can see per-customer unit economics before repricing.
  • An outcome-qualification agent that runs a product/segment through Roee's three-question test (definable outcome, customer-agreed definition, uniform value) and flags whether outcome-based pricing is viable.
  • A billing-transparency agent that gives customers real-time usage dashboards, per-team spend controls, caps/notifications, and full token-level lineage to pre-empt bill-shock disputes and audits.
  • A packaging copilot that clusters customers by jobs to be done (from usage, CRM, and interview data) and proposes value-aligned tiers instead of feature-usage rankings.
Operations Takeaways

By function

The same conversation, filtered for RevOps, pipeline/marketing ops, and customer ops.

Revenue Operations

  • Sequence the design. Packaging → pricing structure → pricing metric → price point. The number is last and least; the metric and bundle carry the leverage.
  • Instrument cost to serve. SaaS-native stacks can't measure per-customer AI cost. Build that instrumentation before repricing so decisions follow unit economics.
  • Guardrail usage-based billing. Recurring base fee + bounded usage, real-time visibility, per-team controls, caps and notifications — the mechanisms that make usage pricing predictable for both sides.
  • Metering lineage is non-negotiable. Trace every token to who consumed it, when, and how much; the first big invoice triggers an audit you must be able to answer.
  • Complexity follows ACV. Match pricing complexity to segment; multi-segment businesses run multiple complexity levels and manage them through packaging.

Customer Operations

  • Migrate the anchor account last. Work backward from the marquee migration: test on least-important segments and new business first, then validate with your 2nd–5th customers.
  • Validation interviews, not surprises. Put new packages in front of customers as design input ('we're not migrating you'), and treat a hesitation-free tier pick as the success signal.
  • Change deliberately. Every pricing change imposes a mental burden on customers — iterate toward a goal rather than flip-flopping between models.
  • Predictability protects retention. Bill shock is a churn risk; give customers spend controls and real-time visibility so they trust the model.
Metrics Mentioned

The numbers, with context

~$500M ARR
When companies hire a pricing expert

Pricing/packaging expertise usually arrives far too late; before that it's owned part-time by a CRO, CFO, product, or product marketing.

80–95%
Classic SaaS gross margins

Near-zero marginal cost to serve made classic SaaS margins extremely high — the economics AI is now breaking.

~50% (or negative)
AI company gross margins

'Some say margins for AI companies is 50%' — and frontier labs like OpenAI run on average-negative margins.

$40M ARR
Bootstrapped company that bled out

A profitable bootstrapped market leader launched AI on a seat-based model in November and was losing money 'like crazy' by February.

$20/month
Consumer AI seat price

A flat $20/month subscription under-captures a heavy FP&A power user and over-charges a casual user — the horizontal pricing trap.

~20 minutes
Support time per closed ticket

Fin can price a closed ticket against roughly 20 minutes of a support manager's time — a uniform, defensible value metric that makes outcome pricing work.

$12K vs $13K (even $15K)
B2B price insensitivity band

In B2B a strong seller wins or loses on value, not a ~10% price delta — why the price point is the least important lever.

every 5–7 years
Obsolete pricing-refresh cadence

The old set-and-forget norm; Roee argues pricing/packaging should now be revisited every product-release or sales cycle.

Entities

Companies, people & tools mentioned

Auto-extracted and linked into the knowledge graph.

Companies

Winning by DesignGTM Consulting / Revenue Architecture

Roee's prior firm (~4 years), where he led the revenue architecture team and taught the flagship revenue architecture course to operators globally; pioneered the Bowtie model.

00:00 · 00:44Company →
OpenAIFrontier AI

Cited repeatedly as the AI-economics reference point: even the best/second-best AI company runs average-negative margins, and its flat $20/month seat-based plan under-captures a heavy power user while over-charging a casual one — the horizontal-outcome problem.

11:33 · 15:22 · 17:41Company →
AWSCloud Infrastructure

Used twice — as the archetype of high-complexity enterprise pricing (seven/eight-figure deals) illustrating the complexity budget, and as the compute/storage cost base that made classic SaaS marginal cost near-zero.

08:12 · 10:46Company →

People

Roee Hartuvguest

Berlin-based pricing & packaging advisor for B2B SaaS and AI companies; former revenue architecture lead at Winning by Design.

Pricing & Packaging Advisor (ex-Winning by Design)Profile →
Anthony Enricohost

Co-founder of LeanScale and host of The LeanScale Podcast.

Co-FounderProfile →

Tools & software

FinAI Support Agent

Roee's canonical example of well-fit outcome-based pricing: an AI service-agent product that charges per closed customer ticket — a clearly defined, uniform outcome priced against ~20 minutes of a support manager's time.

Website ↗
ClaudeAI Assistant

Cited as a product iterating its pricing and packaging almost weekly (free design, credit-based code, Cowork promotions) and expected to move toward usage-based pricing as it ships features.

Website ↗
Claude CodeAI Dev Tool

Referenced ('code is credit', code 'is supposed to go into usage at some point') as an AI coding product expected to shift onto usage-based pricing.

Website ↗
Methodologies referenced Bowtie ModelJobs to Be DoneValue Selling
Related LeanScale Services

Where this shows up in the work

Consumption & Usage-Based Modeling

Quota design, hunter/farmer structures, and unit economics for usage-based businesses.

The episode is a deep clinic on usage- and outcome-based pricing, metering, and unit economics — the core of LeanScale's consumption-modeling work.

Fractional RevOps

Embedded, field-facing RevOps leadership and execution — the operator model Jimmy describes, delivered as a service.

Repricing, packaging redesign, and the validation journey are exactly the operator-led engagements LeanScale embeds to run.

AI for GTM

Agents and workflows that multiply operator throughput — the 'make humans superhuman' thesis, operationalized.

Roee's thesis on how AI broke SaaS cost-to-serve is the economic backdrop for LeanScale's AI-for-GTM monetization work.

Forecasting & ARR Definition

Define a raise-worthy ARR and build a reliable consumption forecast across finance, data science, and go-to-market.

Making usage-based revenue predictable for CFOs and investors — base fees, guardrails, and defensible ARR — ties directly to LeanScale forecasting engagements.

Frequently Asked Questions

Straight answers

Generated from the conversation, marked up for search and AI extraction.

What is outcome-based pricing, and why is it a trap for most AI companies?

Outcome-based pricing charges customers for a result the product delivers (for example, a closed support ticket) rather than for seats or raw usage. It works only when the outcome is uniform and definable across a vertical — Fin can charge per resolved ticket because a ticket maps to roughly 20 minutes of support time for every customer. It's a trap for horizontal AI products because the same usage is worth a near-free Google search to one user and a two-week analysis to another, so a single outcome-based price can't capture value fairly. That's why the buzzword is rarely executed well.

Should you start with pricing or packaging?

Start with packaging. Roee Hartuv's order of operations is packaging first, then pricing structure, then the pricing metric (what you charge on), then the price point last. Packaging — how you bundle features around the customer's jobs to be done — carries most of the leverage and removes most sales friction. The actual number is the least important decision, especially in B2B where a deal rarely turns on whether it's $12,000 or $13,000.

What is the complexity budget in pricing?

The complexity budget is the idea that how complex your pricing and packaging can be is capped by your ACV and the type of customer you sell to. High-ACV enterprise buyers (think AWS selling seven/eight-figure deals) can absorb intricate pricing; low-ACV startup buyers need the simplest possible purchase, which is why vendors offer startup packages. If you sell to multiple segments, you run multiple levels of complexity and manage them through packaging.

What is Roee Hartuv's three-question test for outcome-based pricing?

It decides whether outcome-based pricing fits: (1) Can you clearly define one outcome across your entire customer base? (2) Will your customers agree to and accept that exact definition, including edge cases like a reopened ticket? (3) Is the value of that outcome similar across all customers? Only if all three answers are yes are you a good candidate. Most companies selling to varied use cases fail question one or three.

How did AI break SaaS pricing economics?

Classic SaaS had near-zero marginal cost to serve and 80–95% gross margins because you build the product once and run it on cheap compute and storage. AI reintroduces a real, usage-driven cost to serve — some peg AI-company margins near 50%, and frontier labs like OpenAI run on average-negative margins. Most SaaS-native companies don't even have the tooling to measure per-customer cost to serve, so pricing that doesn't follow that cost quietly loses money as heavy users scale.

How do you make usage-based pricing acceptable to CFOs?

CFOs resist usage-based pricing because it's unpredictable — 'bill shock' is a real fear. The best mechanism is a recurring base fee with usage on top: the base gives both sides a predictable floor, and the usage layer is bounded with caps, notifications, and per-team spend controls. Pair that with real-time usage visibility and token-level billing traceability, because the first large invoice will trigger a customer audit you need to answer.

How do you reprice your biggest customers without losing them?

Use a staged validation journey that works backward from the goal. Test the new model first on your least-important segment or a low-stakes new market, then on net-new business, then run validation interviews with your second-through-fifth most important customers framed as design input ('we're not migrating you, we just want your thoughts'). Add internal validation with the sales team who weren't in the design room, then migrate the anchor account last — arriving with evidence and confidence instead of hope.

How often should you revisit pricing and packaging?

Treat pricing as a living product rather than a five-to-seven-year set-and-forget decision. Revisit it every product-release cycle or sales cycle, whichever is longer — a six-month sales cycle implies a six-month review, and a major new feature set is a trigger to ask whether you should reprice. Change deliberately, though: every change imposes a mental burden on customers, so iterate toward a goal instead of flip-flopping between models.

Timeline

Jump through the conversation

Click any chapter to open it in the transcript.

00:00Cold open + intro 01:42Why pricing is the most overlooked growth lever in B2B 04:27Why you start with packaging, not pricing 06:18Packaging by jobs to be done vs. by features 07:16Why the price point matters least in B2B 09:20The complexity budget framework 10:46How AI broke SaaS pricing economics 15:51The $40M bootstrap company that bled out on AI 17:50Why outcome-based pricing is a trap for most companies 22:47Roee's three-question test for outcome-based pricing 26:10Designing usage-based pricing CFOs will accept 31:12Building packaging around jobs to be done in practice 34:41The validation journey: repricing without losing customers 37:50How often you should revisit pricing 38:52How to work with Roee + wrap
Full Transcript

The whole conversation

Broken into chapters, searchable, verbatim from the audio. Speakers inferred (not diarized).

Transcript as markdown ↓ Whole episode as markdown ↓
00:00Cold open + intro

0:00 Rui Hardev is a pricing and packaging advisor based in Berlin who has quietly become one of the sharpest voices in B2B Saas on how to actually capture value. He spent nearly four years at Winning by Design, the firm that pioneered the bowtie model, where he led the revenue architecture team and taught their flagship revenue architecture course to operators around the world. About a year and a half ago, he made a very sharp call. Of all the levers a GTM operator can pull, pricing and packaging consistently produced the fastest and biggest results. So he walked away from being a generalist and focused on

0:44 nothing else. Today he runs pricing engagements with B2B Saas and AI companies on both sides of the Atlantic, helping them rethink how they package, meter, and price as the category moves away from seats and into consumption and outcomes. In this episode, we get into why most operators are working on the wrong lever, why the days of pure seat-based pricing are over, and why outcome-based pricing, despite being the buzzword of the year, is a trap for most horizontal AI companies. A year and a half ago, you decided to stop being a generalist, GTM advisor, and only work on pricing and packaging. What were you seeing across your

1:27 engagements that made you say, "This is the lever, and I'm giving up on the others for now." That most of the other levers take time to see results. A lot of them are focused on humans, especially when we're seller-led and customer success-led, human-led, not self-serve, and making all these changes take time, and you can train and teach and introduce tools, but as soon as somebody leaves the organization, then it's back to square one. I came to the conclusion that if we fix pricing and packaging, that unlocks many different opportunities along the customer journey, from the acquisition, to the onboarding, to the renewal and expansion,

01:42Why pricing is the most overlooked growth lever in B2B

2:18 and it touches everything. It touches the go-to-market, but it also touches product, and it touches operations, and by getting pricing and packaging right, you can really fix a lot of different aspects of a company. I do believe that pricing and packaging is like a magic growth lever we have, and it is mostly overlooked. There are not a lot of experts out there, or companies don't have a pricing and packaging expert until they reach that 500 million ARR. Before that, sometimes it's a CRO, sometimes it's a CFO, sometimes it's product or product marketing. There is not a lot of knowledge, and I believe that

3:01 if we start working on pricing and packaging much earlier, we can get much better results. I can absolutely attest to my journey here at LeanScale. I started the company in 2021, and pricing and packaging has been and continues to be one of the biggest challenges that I have had in our business. It hasn't been the messaging, it hasn't been the sales, it hasn't been the delivery, it hasn't been the recruiting. Those things have had decent paths forward, but I personally have had a really, really tough time with this topic, and I'm curious, when you approach one of your clients, what is the order of operations? If they have this

3:50 intuition that their pricing and packaging is not working, where do you even begin? I begin with packaging. Packaging is how we bundle our different features and functionalities of our product. The classical way of doing packaging, of bundling, is usually it used to be considered a product exercise, where the product team just looks at our existing customers. Let's look at who uses what features, and let's rank all the features and how many customers are using it. This takes the first 60%, the first batch, and put that under the basic. The better package gets the 60 and additional 30. The best is that 10% of features

04:27Why you start with packaging, not pricing

4:38 that are maybe the premium features, that's the AI, that's the advanced analytic capabilities. This is how the classical way of doing packaging worked. The new way or how we believe it should be done is around jobs to be done. It's around what your customer is looking to achieve using your product. It's that value that everybody talks about. To give you an example, before I became a consultant, I was a software seller. I climbed my way out from a solution engineer, account executive, to a couple of sales leadership roles. Every time we do value selling, of course, everybody wants to do value selling. Only when I got into pricing and packaging,

5:24 I discovered that you can talk and you can say that you're doing value selling, but if your packaging is not aligned to values and jobs to be done, then you can talk all you would want, but at the end of the day, you will always fall back to talking about features and about the product itself and not around the customer. If we can bundle and package around jobs to be done and what the customer wants to get out of the platform, that solves a lot of friction across the sales process and the renewal. If a customer can see the jobs to be done, the different packages and say, "Hey, this is the job that I'm looking

6:02 to achieve. This is the right package for me." It solves most of the scoping aspects of the sales process. And one of the things I know you mentioned is you leave the actual price point for the last thing. What are your opinions on pricing high, pricing low? How do you come up with a number that makes sense once you've gone through the packaging exercise? This is where we try to understand the customer's willingness to pay. You're right that we leave it to the last step. So we start off with packaging and then we do the pricing structure and then the pricing metric. The pricing metric is what we price on. And then how do we price?

06:18Packaging by jobs to be done vs. by features

6:42 Do we price on front or backward looking? So they pay only after they consume, et cetera, et cetera. And the price point comes last. In B2B sales, which I assume is most of our listeners today, it doesn't matter for our customer if it's 12,000 or 13,000. The deal will not, we won't win or lose. It's not like retail or B2C where yeah, every dollar could have an impact. It's B2B. We all know this. We all know that if we put our best sellers, they will be able to sell at 12 and 13 or even 15 because it's all about value. So that's why the price, I wouldn't say it's less important, but compared to all the other

07:16Why the price point matters least in B2B

7:25 mechanisms that we have when defining pricing and packaging, usually that's the least important. Yeah. And on the B2C side, typically the packaging part is pretty easy. You have a single product or something very simple that you will put a price on. So then a lot of the science ends up being on the number. But in most B2B, we're selling very complex products or services to solve very complex problems and to very different industries that may value certain things differently. So there's a lot of complexity for us to walk through that. I have been in B2B SaaS my whole career, and it's always been one of the toughest things for us to

8:12 figure out. Yeah. Complexity is we have a budget of complexity. Now, obviously the higher your ACV, which correlates with the type of customers that you're selling to. So the bigger they are, the more complexity you can add into the sales process or your pricing. So for example, if I'm AWS and I'm selling big seven eight figure deals with enterprise accounts, of course it is very complex. There's so much things in the pricing and packaging. But when I'm selling to a startup that is just starting off and the ACV there is very low and the complexity has to be very low as well. So that's why they have the startup packages.

9:05 So to keep it simple, to have the buying process as simple as possible. So that's we call that complexity budget. How do you sell? Who do you sell to? And what is the level of complexity that you can introduce into the pricing mix? And by the way, if you're selling to multiple segments and multiple type of customers, you might have multiple levels of complexity. That's also where packaging might come into the play as well. Okay. And let's get into pricing and packaging in the new world of AI. There's so many new things. I had worked in companies that had consumption based pricing before. So it's something I was familiar with.

09:20The complexity budget framework

9:54 So many things to figure out on that side. But we're also being introduced to outcome based pricing as well. How are you helping companies navigate this whole new environment? So outcome based pricing is the what you price on. You price on outcomes. So that's one topic. Usage based is how? How do you price on that? And you could do both, right? I could do consumption based on outcome. An example of that FinAI, that's a company that does service agents and they just close customer tickets. There's a consumption element. So you pay based on how many tickets were closed. And those tickets are outcomes. So we only charge you. It is

10:42 the what we only charge you on closed tickets. So you can have both consumption. Let's talk about what's changing in SAS. Our cost to serve is almost zero. The marginal cost to serve the 10th client or the 10,000 client is the same. It's practically zero because in SAS you build the product once you have infrastructure, AWS compute and storage. And it's basically zero. That's why our margins were so high. Our margins was in classic SAS 80, 90, 95%, right? 95% the best companies out there. But we all know that AI comes with the costs. Some say that margins for AI companies is 50%. And we all know that companies like

10:46How AI broke SaaS pricing economics

11:33 OpenAI are losing money. In OpenAI, it's clear they just released or their financials were released. So they're losing a lot of money. So their margins are actually on average, they're negative. And when that happens, when your cost is not zero, all of a sudden your cost to serve as there's a cost to it. So every customer that you bring with your onboard, it adds more costs. So if the way you price does not follow your cost and cost is usage, then you're probably going to lose money down the road. And I think a lot of these companies are not used to that dynamic. They're used to having

12:21 these almost near 100% gross margins. And now that they have to manage the cost side of it, it's creating a lot of complexity in how they package. And we have so many classic SAS companies that have bolted on AI capabilities. So they're already in market with a certain offering with certain promises and certain packaging that's typically revolved around seats or user access or something. And then you bolt on these things that start exponentially increasing the cost side of the equation. How are they navigating that change? So first of all, we can, again, coming back to OpenAI, even the best company in the world

13:05 or second best depends on which camp you are part of, they're losing money. So even them, they are losing money. So all the rest of us probably doing not as good as they are. So we're all struggling to navigate that. And it's very hard. And again, coming back to SAS, in SAS, we only cared about top line. So our cost usually serves on marketing and development. All of a sudden we're introducing another line item, which is cost to serve. SAS native companies, they don't even have the tools to measure that. And when I come in and work with my clients, it's like, what is the unit economics of selling to this type

13:46 of client? We don't know. We can give you the total support and customer success team. But we don't know. We never measure that. They don't even have the infrastructure to measure that. Every company that now wants to introduce AI needs to understand that it's not only a pricing and packaging exercise, it's an operational and a product exercise as well. To give you another example, I recently did a work for a company that they bootstrapped 40 million in ARR. They're probably the leader in their space. And they introduced AI into a seed based model. They launched that in

14:26 November of last year. In February, they called us up and told us, hey, we were a profitable bootstrap company. Since we did this launch, we're losing money like crazy. We need to stop the bleeding now. Again, they were a SAS company, introduced AI capabilities and kept the seed based model. So we worked and we finished the project a month ago to redefine their pricing and packaging, mostly the pricing, to follow that cost of the AI and to make sure that they're not losing money. And again, it's not all the users, right? We will still have some users that are profitable, but the ones that are not, it goes exponentially.

15:22 So if you're doing a seed based, for example, let's take again, open AI cloud. I'm using 20, I pay $20 a month. I'm sure that I'm costing them five times that with the amount of tokens, et cetera. So even they don't want to limit that. So a lot of other companies are struggling with that as well. So in that case, you have a company seed based model. They bolt on some AI capabilities, costs increase. You have to start pegging some of this to usage. And then I think the next layer that comes in, because then sometimes your customers will say, Hey, I want you to be efficient with my usage or certain jobs

15:51The $40M bootstrap company that bled out on AI

16:06 I'm trying to get done with the usage that you're bolting on. And then that opens up this world of outcome based pricing. And for some companies, it feels like it's the perfect fit, but for others, it's a trap. And for those listening who are probably scrolling through LinkedIn and getting the outcome based pricing feeds constantly, I was at a RevOps event in London. There were three presentations on outcome based pricing, and it's usually framed as everyone should do it. For those listening, what does a good profile for outcome based pricing look like? And what does a trap look like?

16:48 A profile is that you're setting into a very specific vertical and that the outcomes are identical between all your customers and all your users. Coming back to the Fin AI example. They outcome is ticket closed. That's what they do. Although they're serving different type of customers, the outcome is basically the same. We're closing tickets. There are some what exactly is a ticket? What happens if a ticket gets reopened? Do you get paid, etc? But those are the edge cases. So I think they nailed it. But let's back to open AI. I'm using open AI or Claude. I do very advanced financial planning and analysis using Excels,

17:41 etc. And I'm really wasting tokens on that. And I still pay 20 bucks a month. I discovered that my mother uses open AI. But she uses the same thing to how to make a turkey sandwich recipe. Same thing that you can Google, right? So the outcome for me and the outcome for my mother were both getting the results that we're looking for. But for my mother, if you try to capture the value there, it's a Google search, almost free. And what I'm doing, I'm replacing what would have taken me two weeks work to do the analysis and the financial planning. The value for me, the outcome, I would have been willing to pay much more than

17:50Why outcome-based pricing is a trap for most companies

18:27 that. But they can't charge us differently. They don't differentiate. And that's why when you go above that 20 bucks or the 100 bucks, the next one is to pay based on usage. And they're going like Claude is already introduced and getting used to that. They are like, I think the design is based on usage and code is supposed to go into usage at some point. So yes, that's why outcome based is different when you're selling to a horizontal solution or horizontal segment. Different use cases, outcome means different things for different users. And how can you price based on different outcomes? You can't. And that's why most companies

19:15 who are selling to different use cases can never define what the outcome is. And that's why I think you said it. It's a buzzword. Everybody talks about it, but it's really, really difficult to really nail it. And again, you need to be very specific or vertical needs to be very specific to really be able to do that. I think that's really good guidance. Horizontal versus vertical. So horizontal, it sounds like, hey, there's probably don't even attempt it because the use cases are so broad. It's not really worth the exercise. In the vertical realm, what are some good frameworks or ways of thinking about if your

20:02 outcome is uniform enough across your customers? Because even in the example you gave with Finn, okay, there are some edge cases, ticket sizes might be different, but I'm sure they, hey, we have the calculus to figure out that we have a good median mean of what this look like and we feel comfortable pricing in here. When does the variance look like too much to where, hey, even if you're in a vertical, it might still not be a good fit for you. I want to switch that around. When it's good, if you can ask what is the outcome and you can clearly define that, that's the first question. And if you are able to answer that,

20:42 let's go into the next question. Can your customers define the same outcome and can you agree with your customers on that outcome? So if Finn defines close ticket, can they define exactly what a close ticket means? And again, coming back to the example, if it reopens in two weeks, if it's closed on your end or the customer end, et cetera, we need to make sure that the customers accept that definition. And then the third question is the value similar to all your different customers? And in tickets, for example, probably yes, because closing a ticket on average, your support team needs to spend on it on

21:31 average 20 minutes and then you can put a price tag next to 20 minutes of a support manager. So that is something that we can do. So first of all, can you clearly define it as one across your customer base? Can the customer agree with you and accept that as what you're pricing on? And then is the value the same across your customer base? And if you answer all these as yes, then you probably are in a good situation to do outcome-based pricing. And within those outcomes, are you seeing any differentiation of like in Finn's example, it sounds like it's mostly these cases are

22:17 solved, get these tickets are resolved kind of uniformly, but are there different types of outcomes? Have you seen a company package it well, where there's more than one outcome in their offering that they're pricing against? No, it's a good idea. I haven't seen that yet. And again, that goes, I haven't seen a lot of good examples of outcome-based pricing out there. Everybody wants to say, hey, when you introduce agentic AI, you should price based on outcomes. Great, show me a company that does it well. I haven't seen that much. So if it's hey, this might be for a very select few, very repeatable, very well defined outcomes.

22:47Roee's three-question test for outcome-based pricing

23:11 That's what that pricing model is for. Do we then mostly revert back to nailing down a usage based pricing model with maybe some other fees involved? How? Maybe we can dive into that because I know I've seen all kinds of different usage based models. There's ones that ramp, there's stair steps, there's commitments and then overage, penalties for overage, blah, blah, all the mechanisms. For a company that's introducing a usage based model or has one and could refine it, what are some best practices for company's pricing that way? Let's start off with the downside of usage based. And the biggest downside is customers hate it.

23:59 People hate to buy usage based products because it's not predictable. Because it's not predictable. That's the main reason. I don't know what's my bill going to be at the end of the month or at the end of the quarter. Thereby Bill shock is a real term. So the CFOs really see, are really scared of scare. They don't like to commit to usage based models. However, there's no other way around it. I think that the industry is going there. And that's why you need to introduce different guardrails and different capabilities to increase that Bill shock. Sorry, let's go back. So the customers don't like to, sorry, just hearing some noise.

25:00 Can you ask that again? And let me start off. Yeah, no problem. So for companies that are not really a good fit for outcome based pricing, it sounds like we really need to refine what their usage based pricing model is. And I've seen all kinds of different levers, accelerators, decelerators, stair steps, minimum commits, overage. What's the guidance for a company that is launching or refining their usage based pricing model? Yeah, let's talk about the reason why CFOs hate usage based billings. It's because it's not predictable. Because on usage, you might get different bills based on how you use it, obviously. And then companies

25:54 in order to create that visibility and predictability, companies introduce different techniques to do that. At the end of the day, you want to create as much predictability to your customers as much as possible to guard them and to also guard you, right? So they want to understand how much they're paying and to cap it or create guardrails or notification, etc. And you on the other hand, you need to report to your investors and forecast how much revenue you're going to make. And that's why you also need that predictability. And predictability is creating those different mechanisms. One of the best mechanisms is to create some sort of recurring

26:10Designing usage-based pricing CFOs will accept

26:36 base fee and usage on top of that. So that allows you to be comfortable or to be confident you're going to get that base from your customers and that's predictability. And then the usage on top might fluctuate, but then you can define how much. And then you need to create visibility to your customers. Do they know how much they're using in real time? Can they control who is using what? So for example, if I'm buying certain usage based product for my entire team, I want to make sure that my IT are restricted and the marketing team doesn't use more than these, what they're budgeted for, etc. So you need to create that control. And then you

27:28 need to be able to create all the operations around that. So you send billing and you sell your invoices based on how much they used it. And you're able to track every single token back to when was it consumed by whom and how much because if you don't, every bill is going to create that back and forth. Why is it so high? Is it so low? You need to be able to trace all that. I think the first time you send out a massive bill to a customer, they're going to demand the audit anyway, and then you're going to have to go lay down all of that infrastructure to make sure you can report against it. One of the things we're

28:13 talking about earlier that I think is an interesting approach is really going deep on the jobs to be done. And for a company or a founder or maybe a CRO who's working on this, what's the best way to make sure you're thoughtfully going through the jobs to be done and then use that information and mining to inform your packaging? What do you recommend they actually do? I had a workshop yesterday with a new client and we were in this workshop, which is the packaging workshop, and we focused on jobs to be done. And actually, I did not do a good job. I wasn't able to take them from the mindset of, hey, this is the product

29:03 and this is the features and these are the features that we want customers to buy because these features allow them to use it differently, etc. So this is like product perspective. I wanted them to or tried to take them, hey, let's look at the customer and let's think of what the customer is trying to achieve. So for example, they're doing a marketing automation platform, working with the biggest retails, airlines, these sort of companies, US companies. One of the biggest retail companies is their customers and they just want the most basic functionality. It's the biggest customer and they just use one basic functionality

29:47 of the product. And it's like I said, that's fantastic. That's our first package. That is our entry package because it's only just this basic stuff. And let's build packages that grow. So we're hoping that we can take them from just using that basic feature into more advanced features as we move along and package that. So the first thing, let's look at it from a customer's perspective and try to think what are they trying to achieve and what are the jobs that we're addressing. The ultimate test that I do as part of the validation is to put, to create those new packages and to put those packages in front

30:34 of our customers. So I do validation interviews and then I present the different packaging and say, Hey, these are the jobs to be done. Tier number one, this is what it allows you to do. Tier number two, this is package number two. This is the jobs to be done here. And the job to be done for the third package is this. Can you find yourself which jobs to be done describes you the most. And if they're able to pick without any hesitation, that's the best result that we can find. But if they're saying, I don't know, this is me, but that's also me, et cetera. And that's, that's where we fail. So good packaging structure around

31:12Building packaging around jobs to be done in practice

31:23 jobs to be done. You put it in front of your customer and you say, Hey, where do you see yourself? And then they pick a package based on that. And it has to be one, one to one. Like there's no binary yes or no, it shouldn't be, I don't know. Yeah. And I think a lot of companies we've worked with, I know they haven't gone through that exercise of bringing that validation from their customer base. How much data do you typically recommend? Like how many interviews should you be doing? Is there a certain size of company that might change what those numbers are? But it sounds like a lot of work. And

32:02 I don't know if people realize how much you should be putting into this and investing into this process. What does it look like kind of full scale doing this properly? Full scale. Let's start off with the end goal. The end goal is to bring or to introduce the new pricing and packaging and migrate our most important customer, which is usually our biggest customer to do new pricing and packaging. So that's the end goal. How do we get there? So in order to move our most important client before we do that, we want to test it out with maybe the second, third and fourth and fifth most important customers.

32:46 Let's do how we do that. Let's take the second segment, which is all the second most important segment, and then go back until we do go to the first customer that we want to migrate them to. How do we have the confidence that we have the right pricing and packaging to migrate our existing customers with test it out with new business? How do we test it out with new business? Let's start off with the least important segment region that we have. So I have a client. They're just rolling into the US market, the European company moving into the US. We're going to test it out, the new pricing and packaging in the US market.

33:31 They don't have a lot to lose there. And once they feel confident, they will go back and do this in their most important markets, London, UK market. How do we get to that? We talk to our customers and do these validation interviews that you just mentioned. We test it out with them first and we tell them, "Hey, we're not moving you. We're not migrating you. We're just designing and we want to hear your thoughts before make a decision on this." How do we go in front of our customers? We go internal validation and we test it out with the sales team, with the sales leaders, with everybody that was not part of the design process. And

34:11 how do we build confidence to go in front of that group? We make a design with the right people in the room to make sure that we can introduce it to the rest of the organization and not lose face. So what I've done here is we are mitigating the risk. The end goal is to move our most important client to the new pricing and packaging. And this is how we do that. We test and iterate and build confidence as we go. I have the impression and if this is a wrong take, let me know that you really shouldn't be messing with your pricing and packaging too much, too often. If there's like an end goal and you want to

34:41The validation journey: repricing without losing customers

34:54 iterate and get to that, but it's not something that you should maybe be constantly changing every year. Oh, we're usage based, now we're seat based, now we're outcome based. Do you have any guidance on how long you should let your current plan maybe sit or there are certain triggers that say, "Hey, nope, this signal means you need to change your pricing now." In terms of timing, what guidance do you give your clients? So yes and no. You should not change because every change creates some mental burden on your customers. But I do believe that we are in a stage where pricing and packaging gone are the days that you set pricing and

35:36 packaging like when you build a company and then don't visit it for five to seven years. Because our products change so fast, because the pace in which we deliver more and more value to the customers is so fast, I believe that pricing and packaging should reflect that. We have reached a point where pricing and packaging should be referred to as product. We keep on improving, adapting. I'm not saying completely changed, but we need to revisit our pricing and packaging. And again, I'm a cloud user. I don't know if you are, but if you are, how do you notice every week, they're changing something about their pricing

36:26 and packaging. Design is free. Code is credit. No, code is not credit, or you get more credits if you use co-work. That was last year's last week's promotion. So they keep on changing and iterating. And they know why because they're launching new features in cloud on a weekly basis. And by the way, they haven't adjusted their pricing, but they are doing different changes to try to capture more value from us. Yes. And in terms of how much, I would say revisit your pricing and packaging, either every product release cycle or sale cycle, whatever is comes, usually I would say whatever takes longer. So if I'm selling six months

37:24 of sale cycle, every six months, I would say, okay, let's look at the last six months and let's see this cohort and what worked well, what did not, what do we need to change? And same thing goes to the product. If we launch a new set of features, should we revisit our pricing? Because we're now delivering more value, etc. So yeah, I believe that pricing and packaging should be looked more often than it used to be done in the past. Yeah, I really like that framework to thinking about it. It's hey, when your product evolves, that's a good time to revisit. And I do think the products are evolving at a much quicker pace,

37:50How often you should revisit pricing

38:13 which opens up the need to revisit packaging every time, where maybe the innovation cycles were just longer. So you didn't have the reason to go back to that. Well, I think those cycles make a lot of sense. And I really appreciate you walking through this because this is unbelievably complex. But I also agree with you, this is a huge growth lever. If you can nail this, you can make your customers happy while also extracting the right value that makes sense for what you're offering. And I don't think this is something you just cook up in a couple meetings, it takes the research, it takes the thoughtfulness, it takes the pragmatic

38:52How to work with Roee + wrap

38:52 approach that you outlined, and making sure you have the packaging that's fit for the product that you're offering into the market that you're focusing on. So I think I know a lot of companies that are struggling with this. What's the best way to reach out to you and your firm? And how do you work with companies to make sure they are getting the most out of their pricing and packaging? Best if somebody is interested, LinkedIn. Yeah, reach out on LinkedIn. I also try to post some interesting facts from the world of go-to-market and pricing and packaging in particular. We engage with the B2B companies

39:32 that either the most common use case nowadays is exactly what we talked about. So traditional SaaS companies introducing or baking in AI components and how exactly are we supposed to price on it? Whether it's because we're losing money or because we believe that there's huge value to capture, and we don't know how to capture that. So these are the common use cases that we do. Our engagement is in demand. So we come in, look at the data, build together in a series of workshops. Usually we engage with the executive leadership, the people that can make decisions, series of workshops. And we go through the process that I explained,

40:16 packaging, pricing structure, pricing metrics, and then pricing points. And then we do the validation, rolling it out to new customers and migrating our existing customers. That takes time, obviously. Fantastic. Well, I appreciate everything you shared on the show today. And I'm sure there's lots of people who need help with this. And as we were mentioning, this is only going to continue to change as dynamics and AI offerings change, as pricing dynamics change. So I think thinking of this as a living breathing thing, you need to maintain and take care of as part of your business. I think it's so important and I appreciate

40:56 what you've done and can't wait to follow some of your content and learn more in the future. Thank you for having me. It was my pleasure.