Module 4 · The Operations Disciplines · Chapter 16 of 32
Compensation and Incentive Design
A comp plan is a bet on which growth engine drives your next dollar. Point it at acquisition when the business runs on new logos, split it out for expansion when it does not, and never let a dollar of expansion cost what a dollar of acquisition does.
9 min read
I once inherited a comp plan at a company north of $120M ARR where an account executive earned the same commission rate on a renewal-plus-upsell as on a brand-new logo. Nobody had questioned it. The AEs liked it, because expansions were easy money against a warm account, and the number on the board kept climbing. Then finance ran the cost of a dollar of expansion against the cost of a dollar of new business and found they sat within a few points of each other. We were paying acquisition prices for expansion revenue, and expansion was already the largest of our three growth engines. The plan was quietly taxing the single biggest advantage that recurring revenue is supposed to hand you.
That is a different failure from a badly shaped payout curve, and it is the one this chapter is about. The plan mechanics (pay mix by control, the quota-to-OTE red line, how a payout curve rewards the last dollar) live in the compensation reference topic, and you should read it for the arithmetic. Here we work one level up: not how to shape a plan, but where to aim it. In Territory and Quota Design we set the quota from capacity instead of gut. The quota is the target. Comp is the instruction for how to hit it, and the instruction can point at the wrong engine even when the target is right.
Comp is a bet on which engine drives the next dollar
Back in the factory chapter we split ARR into three streams, acquisition plus retention plus expansion, and watched the growth engine shift as the business scales: new logos carry the first few years, retention overtakes acquisition around $10M ARR, and expansion overtakes it around $100M. Those are illustrative crossover points from Revenue Architecture, not laws, but the direction is real and it is one-way.
A comp plan is a bet on which of those three engines produces your next dollar. Write the AE a rich new-logo accelerator and you have bet on acquisition. Pay a CSM on gross retention and you have bet on keeping the base. Stand up an expansion team on its own number and you have bet on growing the customers you already have. The mistake I keep seeing is a company that crossed into the expansion-led phase two years ago still running an acquisition-led comp plan, pouring its richest incentives into the smallest of its three engines.
The cost-of-expansion rule
Revenue Architecture puts illustrative economics on each engine, and the ratios matter more than the exact figures. A dollar of acquisition costs roughly 100% of the first year’s revenue to win, which is why a new logo throws off no first-year profit. A dollar of retention costs about 10% of the revenue retained. A dollar of expansion costs about 20% of the revenue expanded. Treat those percentages as order-of-magnitude, not as your actuals. The lesson is the shape: retention and expansion are structurally cheaper to produce than acquisition, and that discount is most of the reason a recurring-revenue business compounds at all.
The catch is that the discount only exists when the right team produces the revenue. Hand renewals and expansion to your acquisition team, on the acquisition plan, and their cost drifts straight back toward 100%, because you are paying an acquisition-priced seller an acquisition-priced commission to do work that should have cost a fraction of that. That is exactly what my $120M company had done. So Revenue Architecture sets a target for the whole engine, drawn from its twelfth revenue breakpoint:
cost of a dollar of expansion ≈ (1/4 to 1/6) × cost of a dollar of acquisition
Illustrative ratio from Revenue Architecture (Chapter 9, Breakpoint 12). It holds only when expansion runs on its own team, plan, and metric. Hand expansion back to the acquisition team on the acquisition plan and the cost drifts toward parity with acquisition, which erases the advantage.
Separate acquisition comp from expansion comp
When expansion growth begins to outrun acquisition growth, the book’s prescription is not a plan tweak, it is a phase shift of roughly two years: split expansion into its own motion, with its own team, on its own comp. Two roles run that side of the house, and they come with real OTE anchors from the Go-To-Market Model. The Account Manager carries expansion across hundreds of accounts at around $180K OTE. The Strategic Account Manager owns the seven-figure accounts at north of $1M OTE, where the deals are large enough to fund a dedicated relationship. The AE is paid on new ARR. The AM is paid on net revenue retention and expansion ARR. The SAM is paid on the growth of a handful of named accounts. Three engines, three metrics, three plans.
| Acquisition seller (AE) | Expansion (AM) | |
|---|---|---|
| Paid to grow | New logos | The net revenue of the existing base |
| Primary metric | New ARR | NRR + expansion ARR |
| OTE anchor | Market, by segment | ~$180K |
| Book | A pipeline | Hundreds of accounts |
| Cost target per $ | Baseline (~1x) | 1/4 to 1/6 of acquisition |
Accelerators and SPIFs point behavior in the short run
An accelerator steepens pay above target so the last dollar of the quarter is the most rewarding one to chase, and the compensation topic covers how to shape that curve and why capping it is an anti-pattern. The course-level question is where you aim it. If expansion is now your growth engine, piling a richer accelerator on the AE’s new-logo number just buys more of the engine you are trying to move past. Put the accelerator on the AM’s expansion number instead, so the intensity lands on the dollar you actually need more of.
A SPIF is the surgical version: a short, time-boxed, single-behavior nudge for something you need right now, like a fourth-quarter push on a new product attach, a multi-year term, or a lagging segment. Used sparingly it works. The failure is letting SPIFs accumulate until three or four of them run permanently, which is a second comp plan nobody modeled and nobody can read.
Governance: the plan is part of the operating system
Comp moves behavior and it moves valuation, so it deserves the same discipline Revenue Architecture asks for on the data model: changes run through a process, not a hallway. This is the order I hold every plan to.
- 1
Write the crediting rules before the quarter opens
When an AE sources an expansion and an AM closes it, who gets credit, split or full? Settle it in writing up front. Ambiguity here is where rep trust dies fastest.
- 2
Route every plan change through a change request
A change needs a named owner, a documented reason, and a model of the cost impact before it ships. No mid-quarter edits by feel.
- 3
Never change the plan retroactively
Reps built pipeline against the plan you published. Rewriting it after the fact is the surest way to lose the people you most want to keep.
- 4
Check the plan against the sustainability rule
Comp is the largest line in GTM cost, and the Go-To-Market Model capped total GTM cost at about 20% of a customer lifetime revenue. If the plan blows that ceiling, the motion is unsustainable no matter how well it motivates.
- 5
Administer through a tool with a dispute process
A rep shorted once distrusts every statement afterward. Automate the payout and give disputes a real channel.
Point comp at the engine driving your next dollar, keep expansion off the acquisition plan, and govern changes like the valuation-moving decisions they are. That is the highest-leverage plan work you will do, and it is worth more than any clever accelerator.
There is one assumption underneath all of it: that the pipeline actually reaches the rep you comped to work it. The cleanest plan in the world pays nobody if a lead lands on the wrong team, or reaches the right rep a day too late, or falls between acquisition and expansion with no owner. Making the pipeline arrive at the right desk, fast, with a clean handoff, is the next discipline. In Lead-to-Revenue: Routing, SLAs, and Handoffs we move a lead through the funnel so the behavior you just paid for has something to act on.