An accelerator is a higher commission rate that applies once a salesperson passes target. A plan paying 8% up to quota and 12% above it is accelerating at 1.5 times the base rate. Accelerators exist to keep strong reps selling after they have already secured their on-target earnings for the period.
Three weeks from the end of a quarter, a rep who has already cleared their number is looking at a deal that could sign this week or next. On a flat plan the rational answer is next quarter, where the same deal counts against a fresh target and helps them clear that one too. Sandbagging is not a character flaw in that situation; it is the plan giving correct instructions. An accelerator changes the instruction by making the dollars above quota worth more than the dollars below it.
Multipliers somewhere between 1.25 and 2 times the base rate are the common range above target, with the steeper end used where quotas are set hard and few reps clear them. The trade-off is worth stating plainly, because it rarely gets said out loud in plan design: an accelerator spends its money on the reps who need motivating least, and it is a bet on the quality of your own quota setting. If the quotas came in soft, the accelerator pays premium rates for ordinary work, and it does so at scale, because soft quotas put a lot of people into the accelerated band at once. The cost overrun in a compensation plan usually shows up here rather than in the base rate.
Plans often stack more than one accelerated band: the base rate up to 100%, a higher rate from 100 to 125, higher again above that. Each step adds pull at the point where the previous step has run out of it. A smaller number of plans run the idea backwards at very high attainment, cutting the rate above some ceiling to limit exposure on a deal nobody planned for. Reps read a decelerator as a cap wearing a different word, and they are broadly right to.
A rep carries a $250,000 quarterly quota. The plan pays 8% of revenue up to quota and 12% on everything above it. They finish the quarter at $340,000, which is 136% attainment.
The same production on a flat 8% plan pays $27,200, so the accelerator is worth $3,600 and all of it is earned in the last 36 points of attainment. Seen from the rep's side, the choice to keep selling after quota returned $10,800 on $90,000 of extra work rather than the $7,200 a flat plan would have paid.
Compensation budgets are usually built from on-target earnings multiplied by headcount, which prices every rep at exactly 100% attainment. That is the one attainment level at which an accelerator costs nothing. Model the same plan against last year's actual distribution instead, with its cluster of reps between 110% and 150%, and the variable spend comes out materially higher. The fix is not to remove the accelerator. It is to fund it deliberately, by running the payout curve against a realistic spread of attainment before the plan is signed off, so the overspend is a decision rather than a surprise in month eleven.
The long form lives in the guides: Commission structures.
Commish pays accelerator the way your plan describes it, shows the arithmetic on every line, and traces each payment back to the deal that earned it.