How to design a sales compensation plan

Set OTE before quota

Most broken comp plans start with a revenue number and work backwards to pay. That order guarantees a mismatch, because the quota you need is rarely the quota the market will accept for the pay you are offering. Start with on-target earnings instead. Decide what you must pay to hire the profile you want in the segment you sell to, then back-calculate the quota that delivers that variable at a commission rate finance will sign off on. The Bridge Group's 2026 AE report puts median AE OTE at $200,000 and median AE quota at $960,000. Applying the 53:47 base-to-variable split from its 2024 SaaS AE study, that is roughly $106,000 base and $94,000 target variable, an implied commission rate near 9.8% of bookings. If you want to pay $240,000 OTE against the same $960,000 quota, your rate jumps past 12% and your cost of sale moves with it. Nobody notices that in a planning meeting; everybody notices it in the annual comp cost review. Two numbers deserve a decision before anything else. First, the OTE corridor: a range per role and segment, plus a written policy on whether remote pay is location-adjusted. Second, the attainment assumption. The Bridge Group found 48% of AEs at or above quota in its 2026 data, so a plan that only pencils out if 80% of the team hits plan is already wrong. Model your comp cost at the attainment you actually expect, not at 100% across the board.

Pay mix by role

Pay mix should track how much direct control a role has over closed revenue. The more a person can personally move a deal to signature, the more of their pay can sit at risk. Full-cycle AEs own the number end to end, which is why the Bridge Group's split sits near 53:47 and why 50/50 and 60/40 are both common. SDRs do not control whether a deal closes, so loading them up with variable pay taxes them for outcomes they cannot influence. The Bridge Group's 2025 SDR study puts median SDR OTE at $80,000 with a 68:32 split, roughly $55,000 base and $25,000 variable. Sales engineers, customer success managers and partner managers influence revenue without owning it, and typically land between 70/30 and 80/20. Segment matters as much as role title. An enterprise AE working eighteen-month cycles with three deals a year has far more variance in any single quarter than an SMB rep closing twelve deals a month. Heavy variable pay on a low-volume, long-cycle role produces reps who earn nothing for two quarters and then a fortune in the third, which reads as a lottery instead of a plan. Flatten the mix toward base for long cycles and push it toward variable for high-volume transactional motions. Model each combination before you commit. The OTE pay mix worksheet is built for exactly this: role, segment, base, variable, quota and resulting commission rate side by side. Getting the mix wrong is expensive to correct, because reducing anyone's base mid-year is a legal and morale problem in most jurisdictions.

Quota to OTE ratio

The quota-to-OTE ratio is the single number that tells you whether a plan is affordable and achievable at the same time. It is quota divided by OTE. The Bridge Group's 2026 AE data lands at 4.6x, meaning a rep carries about $4.60 of quota for every dollar of on-target pay. Ratios in the 3x to 5x band are common in B2B software. Below 3x, your cost of sale is high enough that finance will push back. Above 6x, you are asking reps to produce more than the role and the territory can reasonably deliver, and attainment will confirm it within two quarters. Use the ratio as a sanity check on both ends of the plan. If sales leadership wants a $1.5M quota on a $200,000 OTE, that is 7.5x and worth a hard conversation before the plan is published rather than after reps have started missing. If marketing is only going to source enough pipeline for $700,000 per rep and you are planning 4.6x, the quota is a fiction and you should fix pipeline coverage or headcount first. Run the ratio per segment and per territory rather than as a company average. An average of 4.6x hides an enterprise team at 6x and an SMB team at 3x, and the enterprise team will churn. The sales commission calculator is a quick way to check the implied rate at 80%, 100% and 120% attainment before you lock anything. Also account for the ramp period: a rep in month three should carry a prorated number, not a full one.

Accelerators and decelerators

Accelerators raise the commission rate above a stated attainment threshold, usually 100%. They exist to keep strong reps selling in Q4 instead of sandbagging deals into January. Without them, OTE is also the earnings ceiling, and high performers notice. A typical structure pays the base rate to 100%, then 1.5x the rate from 100% to 150%, then 2x above that. Some plans add a second threshold at 200% or cap total payout, and capping is where most of the argument happens. An uncapped plan occasionally produces a payout that makes the CFO uncomfortable, which is a good problem. A capped plan produces a rep who stops working in October, which is not. If you must cap, cap high and say so in writing on day one rather than discovering the cap when someone hits it. Decelerators work the other way, paying a reduced rate below a floor such as 60% or 70% of quota. They protect comp spend on chronic underperformance, but they double as a resignation trigger, so use them sparingly and never in a rep's first full year. Two mistakes recur. The first is applying accelerators to the wrong measure: a rate that accelerates on bookings while the business is measured on net revenue rewards discounting. The second is setting thresholds against a quota that was never achievable, which turns the accelerator into decoration nobody reaches. Model the cost of your accelerator at the top decile of last year's attainment distribution using the commission accelerator model, and confirm you are willing to write that check.

Crediting rules and splits

Crediting rules decide who gets paid for what, and they cause more disputes than rates do. Write them before launch, because retrofitting them after a contested deal means picking a winner in front of the whole team. Five questions need written answers. What event triggers credit: closed-won, signed contract, first invoice, or cash collected? Whose number does a deal count against when the account changes owner mid-cycle? How are split deals divided when two reps work one account, and does the split total 100% or can it exceed it? What happens to credit on a renewal or an expansion inside an account another rep originally won? And does an SDR keep credit on a meeting that closes seven months later? Cash-collected crediting protects the business from paying commission on revenue that never arrives, but it delays rep payouts and reps hate it. Closed-won crediting pays fast and shifts collection risk onto the company, which is why it usually pairs with a clawback. Most teams land on booking-triggered payment with a recovery clause. Overlay roles need their own crediting logic. A sales engineer or a partner manager who receives 100% credit on a deal the AE also gets 100% credit for is double-counted on purpose, and that is fine as long as the comp cost model reflects it. Document every rule with a worked example, including the ugly cases. The commission crediting rules template covers the standard set. Keep the finished rules in the same document as the plan, not in a manager's inbox.

Clawbacks and draw recovery

A clawback recovers commission already paid when the underlying revenue disappears. The usual triggers are a customer who cancels inside a stated window, a contract that is never invoiced, an invoice that goes unpaid past a set number of days, or a deal that turns out to have been booked incorrectly. Clawbacks are legitimate, but they are also the fastest way to lose trust in a comp plan if the rules are vague or applied inconsistently. Three decisions make the difference. Set a window and hold to it: 90 or 180 days from booking is typical, and an open-ended clawback is unworkable. Decide the recovery mechanism, which is almost always a deduction from future commission, not a demand for repayment, and cap the deduction at a percentage of any single payout so a rep never receives a zero-dollar payout. And state the exceptions, because a cancellation caused by a product failure or a service outage is not the rep's fault and charging them for it teaches everyone to distrust the plan. Draws follow similar logic. A recoverable draw is an advance against future commission and must be repaid from later earnings; a non-recoverable draw is a floor the company absorbs. New hires in ramp usually need a non-recoverable draw for three to six months, because a recoverable draw just builds a debt a ramping rep cannot clear. Whichever you choose, write the repayment schedule and the end date into the plan. Start from the clawback policy template and adjust for your billing terms and jurisdiction, then have employment counsel read it before it goes out.

SPIFFs without plan damage

A SPIFF is a short, targeted bonus that sits on top of the core plan to push a specific behavior: selling a new product, booking multi-year terms, clearing aging pipeline, or filling a slow month. Used well, it is a cheap steering mechanism. Used badly, it becomes a second compensation plan nobody documented. Keep four rules. Time-box it, with a published start and end date, so it never becomes an entitlement reps plan their year around. Budget it separately from the commission plan, so your comp cost model stays honest and nobody discovers an unbudgeted six figures in Q3. Make the measure something a rep controls directly and a system can count without a manager's judgement, because a SPIFF settled by argument costs more goodwill than it buys behavior. And never let a SPIFF pay more than the core plan on the same deal, or reps will optimize for the bonus and the quota will suffer. Watch for cannibalization. A SPIFF on a new product often pulls reps off deals that were already going to close at a higher value, so measure total bookings during the period rather than just SPIFF attainment. Also check whether the behavior persists after the SPIFF ends. If it collapses immediately, the problem was never motivation, it was enablement or product fit, and paying again will not fix it. The SPIFF program template covers the structure, eligibility and payout timing. Announce the terms once, in writing, to everyone eligible.

Plan documentation and rollout

The plan document is the product. Everything above is design work that only becomes real when a rep can read it, calculate their own payout, and sign it. A usable plan document states the role, the effective dates, OTE, base, target variable, quota and its measurement period, the commission rate and any accelerator or decelerator thresholds, crediting rules, clawback terms, draw terms, payment timing, and the conditions under which the company can change the plan. Include at least three worked examples, at 70%, 100% and 130% attainment, with the actual arithmetic shown. If a rep cannot reproduce their own number from the document, the document is not finished. Rollout has a rhythm that works. Design in Q3, socialize the structure with sales leadership and finance before anything is written, then run a modelling pass against last year's actual deal data to see what each current rep would have earned under the new plan. That backtest catches the outliers that break plans, such as the rep whose single largest deal would now pay triple. Publish plans before the period starts, not six weeks into it, and require a signature. Brief managers first and separately, because they will field every question and a manager improvising an answer creates a precedent. Build from the sales compensation plan template instead of a blank page, and keep a dispute process in writing: where a rep raises a disagreement, who decides, and by when. Logging disputes in a commission dispute log turns individual arguments into evidence for next year's design.

When to buy comp software

Spreadsheets handle comp fine up to a point. That point arrives when payroll close takes more than a day or two, when disputes are frequent enough that someone is reconciling by hand, when you run more than two or three distinct plan types, or when finance needs an audit trail it cannot get from a shared file. Below roughly 15 to 20 payees on a single plan, dedicated sales compensation software usually solves a problem you do not have yet. QuotaPath publishes its pricing, which makes it the easiest starting point for smaller teams: $35 per user per month plus a $525 per month platform fee on Growth, or $50 per user plus $800 per month on Premium, both billed annually, with the platform fee covering the first five users, as of September 2026. Salesforce lists Spiff at $75 per user per month billed annually, with additional non-Salesforce connectors at $250 per connector per month, also as of September 2026. The enterprise end quotes every deal. CaptivateIQ, Everstage, Performio, Xactly Incent and Varicent all price on payee count and plan complexity and publish no list price. For scale, Vendr reports a median Everstage contract of $44,220 a year, a median Xactly contract of $79,991, and a median Varicent contract of $28,350. Implementation is the cost buyers underestimate. Your plans have to be rebuilt and tested in the tool, and that is a one-time fee on top of licenses at every vendor in this group. Ask for licenses and services quoted separately. If you are down to two, the CaptivateIQ vs Spiff comparison covers plan modelling and Salesforce dependency.

What breaks after rollout

Comp plans fail in predictable places, and most of them are data problems rather than design problems. The first is crediting drift. Account owners change, territories get carved, and the CRM field that drives commission stops matching who actually worked the deal. By month four, someone is correcting payouts manually every cycle. Fix it by deciding which CRM field is authoritative for credit and locking who can edit it. The second is quota that no longer matches the book. A rep who loses three named accounts in a mid-year rebalance is carrying a number set against a territory that no longer exists, and nobody adjusts it because the plan was signed. Write the adjustment rule into the plan: what happens to quota when accounts move, and who approves the change. The third is untracked exceptions. A one-off approval for an unusual deal becomes the precedent for the next five, and within two quarters the real plan is a set of emails. Log every exception with the reason and review the log at plan design time. The fourth is silent arithmetic error. Statements go out, reps check them, and a rep who finds a mistake tells the whole team. Reconcile the first three payout cycles line by line before trusting any process, whether it runs in a spreadsheet or a platform. Review attainment distribution at the end of the first quarter. If almost everyone is above 120% or almost nobody clears 80%, the quota was wrong, not the reps. Treat incentive compensation management as an ongoing operating process with a named owner, not an annual project.

Buyer checklist

• Set the OTE corridor per role and segment against market data before touching quota numbers.
• Pick the base-to-variable split from how much direct control the role has over closed revenue, not from headcount convenience.
• Check the quota-to-OTE ratio per segment and flag anything outside roughly 3x to 6x.
• Model comp cost at your realistic attainment distribution, not at 100% for everyone.
• Write crediting rules, including splits, overlays and renewals, with a worked example for each ugly case.
• Set the clawback window, the recovery mechanism and the exceptions, and have employment counsel review the wording.
• Backtest the new plan against last year's actual deal data to find the reps whose earnings swing most.
• Publish signed plan documents before the period starts and brief managers separately first.
• Reconcile the first three payout cycles line by line before trusting the process.
• Review attainment distribution at the end of quarter one and fix quota rather than blaming the plan.

Related