Every pricing cycle at most restaurant brands starts the same way: finance names a number — "we need 3%" — and someone spreads 3% across the menu. It feels fair, it's easy to execute, and it hits the check target on paper.
It is also, reliably, the most expensive way to take price. Not because 3% is too much, but because the same 3% on every item is wrong for almost every item.
Elasticity is not one number
Price elasticity — how much demand for an item moves when its price moves — varies enormously within a single menu. Destination items that customers come for will often hold volume through several increases. Add-ons, value anchors, and habitual impulse items can shed measurable volume over a fifty-cent move.
A flat increase treats those items identically. The result is predictable:
- On the sensitive items, the increase costs more traffic than the extra margin is worth. Regulars don't storm out; they come slightly less often, and the decline shows up weeks later, tangled in weather and seasonality where nobody attributes it to the price change.
- On the strong items, the increase leaves money on the table. An item that could carry 8% got 3% like everything else, and the difference is margin the brand simply chose not to collect.
The brand ends up with both problems at once: paying for the increase in traffic and under-monetizing its pricing power.
The same target, taken surgically
Here's the part most operators find counterintuitive: two pricing plans can hit the identical check target with completely different traffic outcomes.
A surgical plan starts from where the pricing power actually is. Items with measured room carry more than their share of the increase. Sensitive items carry little or nothing. Landmine items — the ones where history shows a small move measurably shifts behavior — sit the round out entirely. Increases respect psychological price thresholds, staying under round-number boundaries where demand breaks, and no single move crosses the line customers consciously notice.
Same 3% of check. A fraction of the customer impact.
The catch is that "where the pricing power actually is" is an empirical question. It can't be answered by intuition or industry benchmarks, because elasticity is item-specific and store-specific — the same item can have real headroom in one market and none in another. It has to be measured from the brand's own item-level POS history, reading how real customers in each store responded to real price changes.
How to tell which kind of increase you took last time
If you took price in the last year, three questions reveal whether it was surgical or flat:
- Did any items sit the round out? If every price moved, nobody was protecting landmines.
- Was the increase the same percentage in every store? Markets differ; a plan that ignores that was built on an average.
- Was the impact measured against a counterfactual baseline — what would have happened anyway — or against last year? If the answer is "sales stayed up," you know the increase plus everything else that happened. You don't know what the increase did.
None of these require new philosophy, just measurement the industry has historically skipped because the data was hard to assemble. That part is solvable — it's exactly what we built applied pricing to do, starting from item-and-store-level elasticity.
For the full method — pricing approaches compared, elasticity, zones, testing, and honest measurement — see our complete guide to restaurant menu pricing.