Sales Comp Design for Multi Product B2B Motions

The first comp plan a company writes is almost always a single product plan. One quota, one rate, one accelerator. It works beautifully until the company launches product two. Then the plan starts producing the wrong behaviour: reps quietly deprioritise the new product, the cross sell motion fails to compound, and leadership ends up arguing about SPIFFs every quarter. The root cause is rarely rep effort. It is a comp structure that was never designed for more than one motion.

TL;DR: the design principles that survive product two

A multi product comp plan needs three properties to work: a single quota number reps can hold in their head, differentiated rates that signal strategic intent without creating arbitrage, and a clear accelerator that pays for the behaviour the business actually needs (usually attach rate, not raw ARR). Plans missing any of these three fail predictably.

Why single product plans break at product two

A single product plan rewards the simplest behaviour: close more of the one thing. When a second product launches, the rep faces a tradeoff the plan does not acknowledge. The new product is harder to sell (longer cycle, smaller average deal size, less familiar messaging), so the rational rep keeps selling product one. Leadership reads the attach rate, panics, and ships a SPIFF. The SPIFF works for a quarter, expires, and attach rate collapses back to the structural baseline. The cycle repeats every product launch.

The fix is not bigger SPIFFs. The fix is a comp structure that treats the multi product motion as the default, not the exception.

The three options for multi product comp

There are really only three viable comp architectures for a multi product B2B motion. Each implies a different go to market model, and picking the wrong one for your stage creates years of friction.

Option 1: One quota, blended rate

Reps carry a single ARR number. All products count equally toward quota. Commission rate is the same regardless of product mix. Simple to operate, easy for reps to understand, and the default for companies under 50 reps. The weakness is that it gives leadership no lever to push specific products, so the rep mix mirrors whatever is easiest to sell.

Option 2: One quota, differentiated rate

Reps still carry a single ARR number, but commission rates differ by product. Product two might pay 1.5x the rate of product one. This is the right architecture for most companies between 50 and 200 reps, because it preserves quota simplicity while giving leadership a clean lever for strategic intent. The trap is setting the differential too high, which creates arbitrage: reps push the high rate product onto buyers who don't need it, and the data is corrupted for at least two quarters.

Option 3: Split quota by product

Reps carry separate quotas for each product. This is the right architecture for mature multi product motions (usually 200+ reps) where the products genuinely have different buyers, cycles and motions. The cost is operational complexity and the risk that reps mentally treat the smaller quota as optional. Almost no company under 100 reps should be running this model, and many that are have over rotated.

The attach rate accelerator

The single most underused lever in multi product comp is the attach rate accelerator. Instead of paying a flat rate on product two, pay a higher rate on product two only when it is attached to a product one deal in the same quarter. This rewards the cross sell behaviour the business actually needs, protects against arbitrage, and aligns the rep incentive with the customer lifetime value math.

A reasonable starting calibration: standard product one rate at 100%, standalone product two rate at 110%, attached product two rate at 140%. The differential is large enough to change rep behaviour without large enough to distort the customer recommendation. Calibrate against your own data after one quarter.

What to do about renewals and expansion

Most companies bolt renewals onto the AE comp plan and call it done. That is fine for the first two years of a multi product motion. Past that, renewals start to drag AE focus away from new logo and expansion, especially when the renewal book is large. Two structural options work: spin renewals into a dedicated CSM or renewals function with their own quota, or pay AEs a materially lower rate on renewal ARR (typically 20 to 40% of new logo rate). The third option, paying renewals at the same rate as new logo, creates a comp budget problem that scales linearly with installed base and almost always breaks by year four.

Capacity, quota and the comp budget

Multi product comp design has to be solved jointly with capacity planning, not after it. The classic failure mode is designing the comp plan first, then layering on quota assumptions that the capacity model cannot support. See how to size sales capacity before locking the rate card. The comp plan and the capacity plan have to agree on three numbers: total ARR target, productive capacity per rep by tenure cohort, and the implied comp budget as a percentage of new ARR. Most healthy plans land between 18% and 26% of new ARR for direct sales comp.

Common mistakes that break multi product plans

• Designing the plan in isolation from product marketing. If the messaging stack does not support the cross sell motion, no comp plan will produce it. Align on the buying motion first.
• Setting the rate differential too high. Any differential above 2x creates arbitrage. Reps will sell the high rate product even when the buyer does not need it.
• Shipping comp changes mid quarter. Comp changes ripple for two quarters minimum. Ship at quarter boundary or do not ship.
• Communicating the plan with hedges. If the announcement contains phrases like "we'll see how this lands," the team reads ambiguity as leadership uncertainty and the plan loses force.
• Skipping the modeling for top, median and bottom reps. Every comp change should be modeled against last year's actual rep performance distribution. If the model produces material winners and losers, the plan is not ready.

How to test the plan before you ship it

The discipline most companies skip is parallel running the new comp plan against the prior quarter's actuals before announcing it. Take the last four quarters of rep level performance, apply the proposed plan, and look at three things: total comp spend as a percentage of ARR, distribution of comp across rep cohorts, and the behaviour change the plan implies if reps had been operating under it.

If the model produces total comp spend more than two points higher than current, the plan is not affordable without a corresponding lift in productive capacity. If the distribution materially compresses or stretches the top to bottom rep ratio, the plan will produce attrition in one direction. Both are fixable before launch; neither is fixable after.

The quarterly review that keeps the plan honest

A multi product comp plan needs a quarterly review with three questions on the table. Did attach rate move in the direction the plan intended. Did the comp budget land within the modeled range. Did any rep cohort experience an outcome the plan did not anticipate. If the answer to any of the three is no, adjust the next quarter's calibration, not the plan structure. Structural changes belong on an annual cycle. Anything more frequent erodes the trust the plan depends on.

Where to start this week

Pull the last four quarters of rep level data. Calculate the actual attach rate of product two on product one deals. If it is below 30%, your current comp plan is not producing the cross sell behaviour the business is asking for, regardless of what the messaging says. Model the attach accelerator described above against your existing data. The numbers usually make the decision obvious.

Comp design sits at the intersection of three pillars in the GTM Diagnostic: Sales Effectiveness, RevOps and Strategy. The full methodology shows how comp signals roll up into each pillar score, and which pillar gaps usually indicate a comp redesign is overdue.