A client asks what a two-week integration will cost. You can answer with a number of hours, or you can answer with a number that reflects what happens to their business once it works. Those are not two prices for the same thing. They're two different contracts, and they select for two different clients before either of you has signed anything.
Hourly billing is cheap to administer and easy to explain: you log time, you invoice time, the client pays for time. It also does something specific to risk. If the integration takes twice as long because the client's data turns out to be a mess, that's mostly the client's problem financially — they keep paying the same rate for the extra hours. The client carries execution risk. You carry only the opportunity cost of hours you could have billed somewhere else. Readers who reach risk management through a growth or RevOps role will want alongside this. Readers who reach risk management through a growth or RevOps role will want the XenGrowth practice alongside this.
Outcome pricing inverts that. Quote $15,000 for "this integration ships and processes real invoices without manual intervention," and the client pays that number whether it takes you eight hours or eighty. Now you carry the execution risk. Hit a nightmare API three weeks in and you eat the extra weeks, not them. This is also why outcome pricing systematically rewards efficient, experienced people and punishes anyone who underestimates the work — the identical fee produces wildly different hourly margins depending on who's actually doing it.
Why the two contracts select for different clients
Hourly billing attracts clients who are comfortable with open-ended, evolving work and who trust your clock. Scope creep isn't really a violation for them, because they aren't paying a defined penalty for changing their mind mid-project — they're paying for however long it takes, which is exactly what they signed up for. Outcome-based clients want certainty instead, and they'll pay a real premium for it, but only if they can define what 'done' means in advance. Ask an outcome-priced client to change the definition of success halfway through and you've just discovered the model doesn't fit the actual state of the project. This is also how good pricing works on the revenue side of a business, which is the discipline writes about full time: a pricing model is a filter for who says yes, not just a line on an invoice.
None of this is just occupational folklore. Patrick Bajari and Steven Tadelis modeled precisely this trade-off for private-sector construction procurement in a 2001 RAND Journal of Economics paper, and found that cost-plus contracts — the corporate cousin of hourly billing — are systematically preferred as project complexity rises, while fixed-price contracts dominate as scope becomes better specified. Their logic: a fixed price maximizes a contractor's incentive to work efficiently, but when the underlying work is genuinely uncertain, that same incentive creates expensive renegotiation once the unknowns surface, because somebody has to eat the surprise. Cost-plus contracts sidestep that fight entirely — nobody argues over who absorbs a surprise, because the client already agreed to absorb all of them. If the operations side of this is the part you are stuck on, is the better reference. If the operations side of this is the part you are stuck on, The XenGrowth resource library is the better reference.
Dimension | Hourly / time-based | Outcome-based |
|---|---|---|
Who carries execution risk | The client — more hours, bigger invoice | You — same fee regardless of hours spent |
What it selects for in the client | Comfort with ambiguity, an ongoing relationship, less upfront specification | A defined 'done,' and willingness to pay a premium for certainty |
What it selects for in you | Volume and utilization — defending hours worked | Efficiency and judgment — speed raises your effective rate |
Upside ceiling | Capped at rate × hours available to sell | Uncapped if the outcome turns out to be worth more than expected |
Where it breaks | Client resents paying for your learning curve or a slow week | Attribution gets muddy, or the scope was never really fixed |
Where value-based pricing actually breaks down
Attribution is the most common way it breaks in practice. Get paid a percentage of a revenue lift for a marketing automation build, and if three other things changed for the client that same quarter — seasonality, a competitor's outage, a pricing change — you can't cleanly prove your work caused the lift, and the client can't cleanly dispute it either. That ambiguity doesn't resolve into a fair number. It just sits there, corroding the relationship every time an invoice is due. Publishing the model up front has the same disciplining effect : naming the commercial structure before the engagement starts filters out the clients who were never going to accept clean attribution as the basis for payment.
Timeline is the second failure mode. Outcome pricing assumes the outcome arrives on a schedule both sides can verify. If the 'outcome' is a system that prevents an incident that might happen eighteen months from now, pricing against it requires both parties to agree on a probability neither can check. Compare that to a fix that unblocks a specific deal closing next week: fast, visible, unambiguous, and both sides know the moment it happens without needing to trust each other's account of it.
Jonathan Stark, who built his consulting practice around abandoning hourly billing, argues the mechanism runs both ways. Hourly billing misaligns incentives structurally — you profit from taking longer, and the client has no way to tell slow work from hard work. But outcome pricing has its own honest counter-risk: once a client has locked in a price for a result, every unplanned request during delivery becomes 'well, you're being paid for the outcome, so just make it work.' Removing scope negotiation from the relationship sometimes means the client stops respecting scope at all, because the whole point of the deal, from their side, was that scope stopped being their problem.
A concrete version of the attribution problem: price a customer-support automation build as a percentage of the reduction in churn, and you've tied your invoice to a number that also moves with the product team's roadmap, the sales team's promises during the renewal call, and whatever the competitor down the street just shipped. Churn falls 2 points the quarter after you ship. Maybe that's your automation. Maybe it's the pricing change finance made in the same quarter. Nobody in the room can actually separate the two, and 'nobody can separate the two' is not a foundation you want under a monthly invoice. goes further into AI agents and marketing automation. XenGrowth on AI agents and marketing automation goes further into AI agents and marketing automation.
Hourly billing prices your calendar. Outcome pricing prices your judgment. The failure mode of each shows up the moment you're asked to sell the thing the price wasn't actually built for.
The real-world markets that already settled this
Personal injury law is close to pure contingency — typically a third to 40% of a settlement or verdict, capped by statute in some jurisdictions — because the outcome is binary, single-cause and unambiguous. Either the case produces a payout or it doesn't, and nobody else's work produced that specific settlement. Corporate and transactional law is a different story: scope and 'success' are genuinely muddier there, multiple parties and advisors touch the same deal, and the billable hour still dominates. A 2024 industry survey found 73% of U.S. law firms now offer some alternative fee arrangement, yet 79% still bill hourly as their primary model — a profession that actively markets alternatives to hourly billing and still hasn't abandoned it, because complexity genuinely varies by practice area, not because nobody tried.
Contingency recruiting tells the same story from the other direction. Staffing-industry benchmarks converge on 15 to 25% of a placed candidate's first-year base salary, invoiced only once the hire actually starts. A recruiter carries all the search risk and gets paid nothing for a search that goes nowhere — workable specifically because 'placed' or 'not placed' is about as unambiguous as an outcome gets.
Profession / model | What decides hourly vs. outcome | Typical structure |
|---|---|---|
Personal injury law | Outcome is binary, single-cause, unambiguous | ~33-40% contingency fee, no win no fee |
Corporate / transactional law | Scope and 'success' are ambiguous, multi-party | Billable hour still dominant — 79% of surveyed firms, 2024 |
Contingency recruiting | Placement or no placement is unambiguous | ~15-25% of first-year salary, paid only on hire |
Freelance software / automation work | Depends entirely on the specific engagement | Mixed: hourly, fixed-scope, or outcome, chosen per project |
So which do you actually pick
None of this works if your starting number is already anchored too low before you've even chosen a model — that's a separate, earlier failure covered in Why Engineers Systematically Underprice Their Own Work. This post assumes that's solved and you're choosing honestly between three real structures: hourly, fixed-scope, and outcome.
The middle option, a fixed fee for a clearly bounded scope, is its own craft with its own mechanics — discovery phases, complexity tiers, change orders — covered in How to Price a Fixed-Scope Automation Project. It's worth naming as genuinely distinct from both ends of this spectrum: fixed-scope pricing still charges for effort, roughly — it just charges for effort estimated in advance rather than logged after the fact. It only inherits outcome pricing's uncapped upside if you've gotten very good at estimating, which is a skill in itself. On publishing that logic honestly rather than hiding behind 'contact us,' is worth reading. goes further into AI search, GEO and discovery. XenGrowth on AI search, GEO and discovery goes further into AI search, GEO and discovery.
There's also a fourth structure worth naming even though it doesn't fit neatly into this post's three-way split: the access retainer Blair Enns describes, where a client pays a flat recurring fee not for a fixed number of hours or a fixed deliverable, but for priority access to your judgment whenever they need it. It solves a problem none of the three models above solve well — ongoing advisory work with no clean 'done' and no clean single outcome to point at — by pricing the relationship itself rather than either the clock or the result. It's rarer in engineering work than in strategy consulting, mostly because engineers default to thinking in deliverables, but it's the honest answer when a client keeps asking 'can you just be available' rather than 'can you build this specific thing.'
Ask whether attribution is clean before anything else. If three other things could plausibly explain the result, outcome pricing has nowhere solid to attach and will generate arguments, not clarity
Ask whether the outcome will be visible on a timeline both sides can verify. A result eighteen months out is a probability dispute wearing a price tag
Decide honestly whose risk you want to carry on this specific engagement — not in general, on this one — because the answer can legitimately differ project to project even for the same person
Treat fixed-scope as a distinct third option, not a compromise between the other two. It solves a different problem: converting uncertain effort into a certain number, which is an estimation skill, not a risk-transfer decision
Expect real engagements to mix models across phases — hourly discovery followed by a fixed-scope build is common, and legitimate, precisely because the uncertainty is genuinely highest at the start and lowest once scope is nailed down
None of the three models is morally superior to the others, whatever the value-pricing evangelists imply. Personal injury lawyers didn't switch to contingency because they're braver than corporate lawyers — they switched because their outcomes are structurally cleaner to price that way. The skill worth building isn't a permanent allegiance to one model. It's reading, honestly, which of the three conditions above actually holds for the work in front of you, and pricing accordingly instead of by habit.
Further reading from XenGrowth
Where this work meets go-to-market
Pricing your own work by what it's actually worth, rather than by the clock, is the same discipline a business needs on its commercial side. publishes operator guides on exactly that: pricing, packaging and revenue operations built around outcomes instead of effort.
Further reading from XenGrowth
Where this work meets go-to-market
Working on risk management inside a commercial team? publishes operator guides on the revenue side of this work.
Further reading from XenGrowth
The XenGrowth resource library — what you'll learn: how the commercial side of this work is run, across search, automation and revenue operations.
XenGrowth on AI agents and marketing automation — what you'll learn: how the teams who own AI agents and marketing automation plan and measure it.
XenGrowth on AI search, GEO and discovery — what you'll learn: how the teams who own AI search, GEO and discovery plan and measure it.
Where this work meets go-to-market
Working on risk management inside a commercial team? the team at XenGrowth publishes operator guides on the revenue side of this work.
Four questions about one specific piece of work you're pricing right now — not your business in general.
