For a founder still running land-and-expand playbooks alongside new enterprise or mid-market motions, how should commission/quota structure differ to prevent cannibalization?

Two ways to structure the plan: one blended quota or two fenced pools

Every founder running land-and-expand playbooks next to a new enterprise or mid-market motion eventually lands on the same fork in the road. There are really only two structures, and the difference between them is not cosmetic — it determines which motion gets worked and which one quietly dies.

Option A: the blended quota. One rep, one number, all revenue counts. A rep carrying $900K in annual quota can hit it with new logos, expansions, upsells, cross-sells, or any combination. Commission is a flat rate — say 9% — on everything that lands. This is the default structure for a reason: it is simple to administer, it is easy to explain to a rep in an interview, it requires no routing rules, no deal desk, and no arbitration when an account is ambiguous. It also matches how most founders instinctively think about revenue, which is that a dollar is a dollar.

The problem with the blended quota is that a dollar is not a dollar in terms of effort. An expansion inside a happy land-and-expand account is a warm conversation with a champion who already knows the product, usually a 2–5 week cycle, often a 60–75% close rate, sometimes no security review at all because the account cleared security on the original land. A net-new enterprise deal is a cold or lightly-warmed account, a buying committee of six to twelve people, a security questionnaire, a legal redline cycle, procurement, and a four-to-nine-month calendar — with a close rate in the 15–25% range on qualified opportunities. If both retire the same quota at the same rate, the expected commission per hour of rep effort on the expansion is several multiples of the enterprise deal. A rational rep — not a lazy one, a *rational* one — works expansions and lets the enterprise pipeline age. The enterprise motion does not fail loudly; it fails as a slow leak of deals that never advance past stage two.

Option B: two fenced pools. New-logo quota and expansion quota are separate numbers, each with its own attainment curve, its own accelerators, and its own commission rate. An enterprise AE carries, say, $700K of new-logo quota and zero expansion quota. An expansion AE or account manager carries $1.4M of expansion/renewal quota and zero new-logo quota. Neither can substitute. If the enterprise AE lands $500K of new logo and stumbles into a $300K expansion, the expansion does not fill the $200K gap — it pays into a separate, lower-rate line or gets credited to the account owner entirely.

Fenced pools cost more to run. You need routing rules, a graduation trigger, a deal desk or RevOps function to arbitrate ownership disputes, two comp documents, and two quota-setting models. What you get in return is that no rep is ever choosing between motions, because no rep has both motions in their plan. The structural fix for "reps chase the easy motion" is not exhortation or pipeline reviews — it is making sure the choice never appears in front of a rep in the first place.

Option C, the compromise nobody should love but many need: one rep, two sub-quotas, gated payout. The rep carries both numbers but neither accelerator unlocks until the new-logo sub-quota is retired. This is the structure for a company that cannot yet staff two teams. It is worse than Option B and better than Option A, and its main failure mode is that reps who fall behind on new logo in Q1 write off the accelerator for the year and revert to pure expansion behavior. If you use it, reset the gate quarterly, not annually, so a bad quarter does not disarm the incentive for eleven months.

Deciding which structure your stage actually supports

The decision is not "which is better in the abstract" — Option B is better in the abstract, always. The decision is whether your revenue base and headcount can carry the specialization overhead. Below roughly $8M ARR with fewer than six quota-carrying reps, splitting into two teams usually produces two under-staffed teams; above roughly $15M with ten or more reps and validated demand in both motions, blended quotas are actively destroying your enterprise motion whether or not you can see it in the numbers yet.

The honest diagnostic runs in this order. First, is there validated demand for the second motion, or do you have one motion plus a handful of inbound enterprise deals that happened to show up? A few opportunistic logos is not a motion, and building fenced pools to govern occasional good luck is over-engineering. Second, can you fund at least two dedicated enterprise AEs plus the sales engineering support an enterprise motion requires? One enterprise AE is a science experiment, not a motion — a single rep's territory has too much variance to tell you whether the motion works. Third, is the interference already visible: enterprise pipeline that does not advance, forecast slippage concentrated in enterprise deals, reps whose activity logs show enterprise accounts touched once a month?

If the answer to any of the first two is no, the right move is not a more elaborate comp plan. It is to sequence rather than parallelize: run land-and-expand to a durable base, then layer the enterprise motion deliberately as a funded second phase. Comp structure cannot fix a resourcing problem, and a sophisticated fenced-pool plan wrapped around two under-resourced motions is overhead on top of a strategy problem.

The trap in this decision is measuring your way into complacency. Blended metrics conceal exactly the divergence you need to see: a blended net revenue retention of 115% can be a land-and-expand motion at 130% averaged with an enterprise motion at 95%, and the 95% is a serious problem the blend erases. Compute new-logo attainment, expansion attainment, win rate, and cycle length per motion before you decide anything about comp — because if you cannot report per motion, you also cannot pay per motion, and the whole fenced-pool design has no data to stand on.

The numbers behind each structure

Comp design is arithmetic before it is philosophy. Here are the ranges that matter and how they interact, with the caveat that every one of these is a starting point to be calibrated against your own deal data, not a benchmark to be copied.

Commission rate spread. The core mechanic is that new-logo dollars must be worth materially more per dollar than expansion dollars. A common shape: 10–12% of first-year ACV on new logo, 5–8% on expansion, and 1–3% (or a flat retention bonus) on flat renewals. The spread has to be big enough that a rep who could pursue either does the math and picks new logo. A 9% versus 8% spread is decorative — it does not change behavior. A 2× spread does. Where founders get nervous is the cost: paying 11% on new logo against a blended 9% feels like a raise. It is not, if the expansion side comes down to 6% at the same time — the blended cost of sale often lands flat or lower, because expansion is the larger volume line in most land-and-expand companies and it is being paid less.

Quota multiples. The usual planning heuristic is that a quota-carrying rep should produce 4–5× their on-target earnings in new ARR. An enterprise AE at $160K OTE therefore carries roughly $650K–$800K of new-logo quota. An expansion AE is a different economic animal: expansion revenue is cheaper to produce, so the multiple runs higher — often 6–8× OTE — putting a $130K-OTE expansion rep somewhere in the $800K–$1M range on expansion and a separate renewal number on top. Setting the enterprise quota using land-and-expand multiples is the single most common and most fatal quota error: it makes an enterprise motion that is performing normally look like it is failing, which triggers a panic reorg that kills the motion for real.

Ramp. An expansion rep working a warm book is productive in 30–60 days. An enterprise AE selling into a four-to-nine-month cycle is not producing closed revenue for two full quarters, and their first full-productivity quarter is usually quarter three or four. Enterprise plans therefore need a ramp schedule — a common shape is 0% quota in month one, 25% in months two and three, 50% in the second quarter, 100% from the third quarter on, with a guaranteed draw covering the ramp period. Without a draw, the enterprise AE's take-home in their first two quarters is variable-comp zero, and they either leave or start hunting for the fastest close available, which is exactly the behavior the fence exists to prevent.

Accelerators. Accelerators are where you buy the behavior you actually want. On the new-logo plan, run a meaningful kicker above 100% attainment — 1.5× the base rate from 100–125% and 2× above 125% is a common shape — so that the marginal enterprise deal is the most valuable hour in the rep's week. On the expansion plan, accelerators should be flatter, because the goal there is consistent coverage of a book, not heroic overperformance on a handful of accounts. A flat 1.25× above quota on expansion is usually enough.

Graduation and sourcing credit. When a self-serve or land-and-expand account grows into an enterprise platform deal, both sides must get paid or the account owner will hoard it. A workable split is that the enterprise AE receives full new-logo-rate credit on the incremental platform ACV, and the land-and-expand owner receives a sourcing credit of 15–25% of the enterprise AE's commission on that deal — paid on top, not carved out of the enterprise AE's number. It costs a few points of blended commission and it buys you the graduation path, which is the highest-leverage thing in a two-motion company: the land-and-expand motion seeds accounts at low acquisition cost that the enterprise motion later expands, so the two motions feed each other as a pipeline rather than fighting as competitors.

Clawbacks and payment timing. Land-and-expand deals churn differently from enterprise deals. A reasonable structure pays expansion commission monthly on booking with a 90-day clawback for accounts that churn or downgrade inside the first quarter, and pays enterprise commission 50% on signature and 50% on first invoice payment, with a 6–12 month clawback on early termination. Founders under-use clawbacks and then discover a rep was paid 11% on a logo that churned in month four.

One number to watch. Track commission cost as a percentage of new ARR, per motion, every quarter. If total cost of sale on the enterprise motion runs above roughly 25–30% of first-year ACV once you include the AE, the SE, and the SDR support, the motion's unit economics need attention — but judge that number against enterprise payback expectations (often 18–24 months), not against the 8–12 month payback a healthy land-and-expand motion produces. Applying land-and-expand payback standards to an enterprise motion is another way founders talk themselves out of a motion that is working exactly as it should.

Implementing it without breaking the quarter

Comp changes are disruptive by nature, and a badly sequenced rollout does more damage than the cannibalization it was meant to fix. The sequencing below is the one that survives contact with a real sales team.

Step one: segment your reporting before you touch the plan. You cannot pay per motion if you cannot report per motion. That means every opportunity carries a motion tag, every closed-won record attributes to new logo or expansion cleanly, and you can produce win rate, cycle length, average deal size, and attainment split by motion for the trailing four quarters. This is RevOps work and it usually takes four to eight weeks. Skip it and the first commission dispute — and there will be one in week three — has no data to resolve it.

Step two: write the routing rules and the graduation trigger before the comp plan. The fence is what makes the comp plan enforceable. Write down the segmentation thresholds explicitly: which employee count, which expected ACV, which industries, which arrival channel routes to which motion. Then write the edge cases, because edge cases are where fences fail — what happens when a self-serve account starts adding seats and a security questionnaire arrives, what happens when an enterprise rep discovers a small team inside their target already self-serve, and who adjudicates when both teams claim a logo. Name a single routing authority, usually RevOps or a deal desk, with the power to decide fast so field conflict does not fester into a comp dispute.

Step three: model the new plan against the last four quarters of actual deals. Run every closed deal from the trailing year through the proposed plan and compute what each rep would have earned. You are looking for three things: whether total commission spend lands where you expect, whether any individual rep takes a severe pay cut (which is a retention event you need to plan for, not discover), and whether the plan produces the behavior you want on the actual deal mix you actually have. A plan that looks elegant on a whiteboard and cuts your best rep's earnings 30% is not going to survive.

Step four: transition mid-year with protection, or wait for the fiscal boundary. The clean answer is to launch at the start of a fiscal year. If you cannot wait, transition at a quarter boundary and offer a one-or-two-quarter earnings floor — reps earn the greater of the old plan or the new plan during the transition window. It costs real money and it buys you a team that does not spend the transition quarter updating their résumés.

Step five: reassign books explicitly, in one motion, with the customer in mind. When you split teams, accounts move. Do it once, cleanly, with a warm introduction to each affected customer rather than a silent CRM reassignment. The account owner who is losing an account should stay involved through the transition, which is exactly the behavior the sourcing credit is paying for.

Step six: instrument the seams and review at 90 days. After launch, watch the specific signals that tell you whether the fence is holding: enterprise pipeline progression by stage, whether new-logo attainment is actually moving or reps are just booking the same expansions under different labels, discount depth on enterprise deals (a sign the published land-and-expand price is being used as an anchor against you), and the volume of routing disputes. High dispute volume means the fence has holes; near-zero enterprise pipeline movement 90 days in means the incentive spread is not wide enough.

The discipline that matters most after launch is stability. Comp plans that change mid-year teach reps that the plan is a suggestion, and reps who believe the plan is a suggestion revert to whatever behavior paid last year. Fix routing holes and quota errors as they surface, but hold the rate structure fixed for a full fiscal year so the incentive has time to actually change behavior.

What the comp plan cannot fix

It is worth being explicit about the limits, because founders routinely ask comp to solve problems that are not comp problems.

Comp cannot fix a pricing conflict. If your published land-and-expand price is $25 a seat and your enterprise motion is quoting $55, no commission structure stops procurement from throwing your own price list back at your rep. That requires separate price books, real feature fences — SSO, SCIM, audit logs, advanced admin, compliance certifications sitting only in the enterprise edition — and an enforced enterprise floor below which the motion does not transact.

Comp cannot fix a roadmap conflict. If enterprise deals are dying in security review because SSO was never built, paying the enterprise AE 12% instead of 9% changes nothing. That requires explicit capacity allocation between the motions on a fixed cadence, decided deliberately rather than by whoever advocates hardest in the weekly product meeting.

And comp cannot fix an absent core motion. If "two motions" is a euphemism for a GTM strategy that never developed conviction about a single center, a fenced-pool comp plan just makes the incoherence more expensive. The diagnostic question is honest and uncomfortable: do you have two validated motions each deserving dedicated resourcing, or one motion plus drift? Comp is the enforcement layer for a strategy that already exists. It is not a substitute for having one.

Related questions

Should the founder personally carry a quota during the transition?

No. A founder carrying a number competes with the reps for the same deals and distorts every forecast. The founder's role in early enterprise deals is strategic credibility and executive relationships — participate in deals without owning quota credit for them.

How do you handle a rep who legitimately works both motions early on?

Use Option C: both sub-quotas in one plan, with all accelerators gated on retiring the new-logo number first. Reset the gate quarterly. Treat it as a bridge to team separation, not a permanent structure.

Does the SDR or BDR comp plan need to change too?

Yes. Enterprise-targeted SDRs should be paid on qualified meetings that advance past discovery, not raw meeting count, because raw meeting count pushes them toward easier SMB targets. Longer cycles also mean their opportunity-based bonus pays out one to two quarters later.

What about customer success — should CS carry expansion quota?

Only if CS actually owns the commercial conversation. If CS is expected to drive expansion but is comped purely on retention and health scores, expansion falls between the teams. Either give CS a real expansion number and rate, or route expansion to a dedicated commercial owner.

How often should the plan be recalibrated?

Rates and structure hold for a full fiscal year. Quotas can be reset annually, with mid-year adjustments only for territory changes or headcount shifts — never as a response to overperformance, which is the fastest way to lose a top rep.

FAQ

Should expansion revenue count toward new-logo quota at any ratio?

No. Partial credit — counting expansion at 50% toward the new-logo number, for example — sounds like a reasonable compromise and it reintroduces exactly the arbitrage you were trying to eliminate. A rep who can retire new-logo quota with discounted expansion dollars will do the arithmetic and find that two expansions beat one enterprise deal in expected value. Keep the pools genuinely separate. If you want to reward a rep who does both, pay a separate bonus, not quota credit.

What commission rate spread is actually large enough to change behavior?

The spread has to be large enough to overcome the effort and probability difference between the two deal types. Given that an expansion might close at 65% probability in four weeks and an enterprise deal at 20% in six months, a one- or two-point spread is noise. A ratio of roughly 2:1 between the new-logo rate and the expansion rate is the practical starting point, tightened or widened based on your actual cycle length and win-rate data.

How do you set an enterprise quota with no historical enterprise deals to model from?

Set it conservatively from OTE and work backwards: quota at 4–5× OTE, with a full ramp schedule and a guaranteed draw. Treat the first two quarters as data collection rather than performance assessment. Once you have eight to twelve closed enterprise deals, you have enough to compute real average deal size, cycle length, and win rate, and you recalibrate at the next fiscal boundary — not mid-year.

Who owns the account after a land-and-expand customer graduates to an enterprise contract?

The enterprise AE owns the commercial relationship going forward, but the transition should be a warm joint conversation, not a reassignment. The prior owner stays engaged through the handoff and receives a sourcing credit on the enterprise deal. If the customer experiences graduation as being reassigned to a stranger, you have converted a growth account into a churn risk.

Does this apply to a mid-market motion or only true enterprise?

The same structure applies to any motion with materially different cycle length, deal size, and close rate than your land-and-expand base. Mid-market typically sits between the two — shorter cycles than enterprise, higher win rates, smaller committees — so the rate spread and quota multiples land between the two extremes rather than at the enterprise end.

What is the first sign the fence is failing?

Discount depth on enterprise deals climbing quarter over quarter, and enterprise opportunities sitting in the same stage for more than a full cycle. The first says your pricing fence is leaking; the second says reps are logging enterprise pipeline for the forecast without actually working it. Both show up in RevOps reporting a quarter or two before they show up in revenue.

Sources

• https://openviewpartners.com/blog/land-and-expand/
• https://www.saastr.com/how-to-set-quotas-for-sales-reps/
• https://www.bridgegroupinc.com/saas-ae-metrics
• https://hbr.org/2015/04/motivating-salespeople-what-really-works
• https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights/sales-compensation-a-key-to-driving-growth
• https://www.forentrepreneurs.com/saas-metrics-2/
• https://a16z.com/the-saas-metrics-that-matter/
• https://www.gartner.com/en/sales/topics/sales-compensation
• https://www.pavilion.com/
• https://www.winningbydesign.com/resources/

Related on PULSE

• How to build a graduation trigger that moves self-serve accounts into the enterprise motion
• Setting enterprise sales quotas when you have no historical enterprise deal data
• Pricing architecture for running self-serve and enterprise editions side by side
• Segment-level GTM reporting: why blended CAC and NRR hide broken motions
• When to split one sales team into two specialized motions
• Sales compensation clawbacks and payment timing for long-cycle deals