A SPIFF you can't measure is spend and hope. Scope one selling behavior, put a reward on it, and track what it pays back.
A SPIFF goes out on Monday. Sell the most of this product by Friday, win the prize. The floor lights up, a leaderboard fills in, and by the weekend someone takes home a gift card. Then the numbers settle back to where they were, and nobody can say whether the week paid for itself.
That's how most sales incentives run. You pick a reward, add some urgency, and hope the extra selling covers the cost. Ask a month later what the SPIFF returned, and nobody actually tracked it.
A SPIFF can do better than that. The ones that work are scoped tightly enough that you can measure what came back, so you're spending and measuring instead of spending and hoping.
What a SPIFF is supposed to do
A SPIFF is a short-term reward for one specific selling behavior: move this product line, raise the attach rate, push units per ticket this week. It's different from commission, which pays a percentage of everything a worker sells no matter what moved.
That narrow aim is the whole point of a SPIFF, and it's also where most of them slip. The reward gets set and the contest gets announced, but the behavior never gets defined precisely enough to measure it afterward.
So you get a burst of activity and no clean way to separate what the SPIFF drove from what would have sold anyway. The prize is easy to see. The margin it bought stays fuzzy.
Why don't most SPIFFs get measured?
Because they're scoped too wide to measure. "Sell more this month" rewards a number that moves for a dozen reasons, and you can't pull the SPIFF's share out of it.
A SPIFF aimed at total sales competes with the season, the promotion running alongside it, foot traffic, the weather. When the month comes in strong, you can't tell whether the reward did it or the calendar did. When it comes in flat, you can't tell whether the reward failed or just held the line against a slow stretch.
Aim it at one behavior on one product and the math gets simple. You know the attach rate before, you know it during, and you know what you paid to move it. The tighter the scope, the cleaner the return you can read off it.
Scope one selling behavior
Pick a behavior you can already see in your numbers: attach, add-on, upsell, category mix, units per transaction. One behavior, one product or segment, one window.
"Attach a protection plan to every third laptop" is measurable: you can pull that rate from the POS before and after. "Sell more electronics" isn't, because the category drifts on its own.
- Attach: reward the warranty or protection plan added to a big-ticket item, tracked as attach rate on that item.
- Add-on: reward the accessory, protection plan, or loyalty signup a cashier rings up at the register, tracked as add-ons per ticket.
- Upsell: reward the move from a base tier to a higher one, tracked as the mix between them.
- Category push: reward units sold from one margin-rich line, tracked against that line's baseline.
The upsell nobody makes and the add-on nobody offers are quiet money you're already leaving on the floor. A scoped SPIFF puts a reward on exactly that behavior and lets you watch whether it actually moves.
The reward is the easy part
Setting the prize is the part everyone gets right. What decides whether the SPIFF was worth running is the return, and that's the part most programs never check.
The reward type barely changes the outcome. Cash, a gift card, or a day off all work as long as they're worth the effort to the person on the floor. What matters is whether you scoped the behavior tightly enough to measure what that reward bought.
Treat a SPIFF the way marketing treats an ad. You judge it by what it returned, then run the ones that pay again and cut the rest. Running incentives as a discipline means holding a floor reward to that same test.
How to design a SPIFF you can measure
Four steps turn a floor contest into a spend you can actually read.
- Scope one behavior. Pick a single selling action you can see in the POS, like attach rate on a product line, add-ons per ticket, or the upsell to a higher tier. One behavior, one segment, one window.
- Set the baseline first. Pull the rate for that behavior before the SPIFF starts. Without the before, you have nothing to measure the after against.
- Reward the behavior itself. Pay on the attach or the upsell, so every payout ties to the action you're moving rather than a sales number that wanders on its own.
- Measure what came back, then decide. Compare the behavior during the SPIFF to the baseline, weigh the added margin against what you paid out, and run again the ones that beat their cost while you drop the ones that don't.
Say you put $300 in rewards behind an add-on target on one product line for two weeks. The attachments it drives that wouldn't have happened otherwise add a couple thousand in margin, so a few hundred dollars buys a few thousand. And you can put a number on what that reward returned instead of guessing, week by week.
Plenty of SPIFFs are worth running, and always will be. The money just goes out with no way to read what it bought.
Scope one behavior, set the baseline, and measure what comes back, and a SPIFF stops being a burst of energy you hope paid off. You see what it bought: attach on the products that carry the margin, upsell at the counter, units per ticket you can point to, and a floor incentive you'll fund again because you know it works.
Frequently asked questions
What is a SPIFF in sales?
A SPIFF is a short-term reward for one specific selling behavior, like moving a product line or raising the attach rate within a set window. It sits on top of commission and targets a single action, rather than paying on total revenue the way commission does.
What's the difference between a SPIFF and a commission?
Commission pays a percentage of everything a worker rings, all the time. A SPIFF pays for one behavior over a short window, like attaching a warranty or upselling a tier this week. Commission is the base engine; a SPIFF is a surgical push you switch on to move a specific number.
Do sales SPIFFs actually work?
They work when they're scoped tightly enough to measure. A SPIFF aimed at "sell more" can't be separated from the season or a promotion running alongside it, so nobody can tell if it paid. Scope it to one behavior on one product, set a baseline, and you can see what the reward returned.
How do you structure a sales SPIFF?
Scope one behavior you can see in the POS, pull its baseline rate first, put the reward on that behavior rather than on total sales, then compare during to before. Weigh the added margin against what you paid out, and run again the ones that beat their cost.
How do you measure whether a SPIFF paid off?
Compare the targeted behavior during the SPIFF to its baseline, then weigh the added margin against the rewards you paid. If a few hundred dollars in prizes drove a few thousand in added attach margin, you can put a number on the return instead of guessing.
