Free check · 2 minutes · no sign-up

Can you actually measure your ads?

Every location has good weeks and bad weeks. If your ads add less than that swing, no tool can prove what they did. A few rough numbers, a straight answer, and the test that fits.

Start the checkHow measurement works
1. How is your business set up?

Example numbers are filled in. Replace them with yours — rough is fine.

Count cities or metros, not locations. Five locations in Dallas is one market.

Guests, new members, booked jobs, new patients — whatever you count as a customer.

4. How much do normal weeks bounce around?

Most weeks land within ±10%

Not sure about the cost per customer? Use roughly what a new customer costs you today. That number is exactly what a test would tell you.

Test length

Market twins · estimate

Yes — your markets can show this.

An 8-week read with 2 test markets and 4 in the comparison pool should show your ads’ effect: about 19% more customers, against a smallest provable lift of 10%.

Likely lift from your budgetSmallest lift a test can prove
Likely lift
19%
≈ 58 customers / week
Smallest provable
10%
≈ 31 customers / week
Markets test / pool
2 / 4
matched groups
How this estimate works

A test can only prove a lift bigger than the noise. We estimate that noise from how much your weeks bounce around plus counting noise (a ZIP code with two jobs a week is mostly noise), split your markets, ZIPs or weeks into a test group and a comparison group, and take the lift you could detect with 95% confidence and 80% power.

“Likely lift” assumes your whole test budget buys customers at the cost you entered. On/off tests get a wider margin because there is no comparison location to absorb weather, holidays and seasonality.

It’s a planning estimate from a few rough numbers. Countlift’s test designer does a similar calculation from your real sales history. Nothing you type here leaves your browser.

Three ways to measure

The right test depends on how your customers reach you.

The check picks one. Only market twins are self-serve today; we label the rest plainly.

In the product

Market twins

Locations in 3+ markets

Each market where spend changed is compared with a blend of your markets where it didn’t. This is what Countlift does today, from data you already have.

How MarketTwin™ works →

Early access

ZIP-level test

One location with a service area

Run ads in half your ZIP codes, hold back the other half, and compare. It works when you know each customer’s ZIP.

How ZIP tests work →

Free guide

On/off test

One walk-in location

Alternate ads-on and ads-off weeks and compare. Cheap and honest, but directional: there’s no comparison location.

Read the guide →

Run local media for clients? Check each one before you pitch a test. Countlift for agencies →

Passed the check? See it on your own data.

Free location report · No card · Totals only, never customer records