To measure radio advertising ROI for a multi-location business, run radio in flights in some metros, hold matched metros back, and count the extra customers your locations served against a weighted twin of the dark metros. Multiply those customers by what each one is worth, subtract the full radio cost, and divide by that cost. Codes and call-tracking numbers will undercount. The location comparison won’t.
Radio still reaches almost everyone: 92% of US adults each month, according to Nielsen’s Comparable Metrics report for Q2 2025 (Source: Nielsen, via Radio Ink, 2025). Reach was never radio’s problem. Proof is. Most multi-location brands renew radio on gut feel and the station rep’s ratings book, then cut it the first year the budget is tight, because nobody can say what it did. This guide shows how to get that number.
Why radio ROI is hard to pin down by location
- There is nothing to click. Listeners hear the spot in the car and act later, often days later, by searching your name or just turning in. That search will be claimed by your Google Ads account as its own conversion.
- Codes catch the few who bother. “Mention this ad” and radio-only offer codes need the customer to remember and the front desk to record it. Both happen far less often than the visit itself.
- Station signals ignore your map. Radio is sold by station, and a strong station’s signal can cover parts of several metros. Your markets and the station’s audience rarely line up neatly.
- The effect lags. For need-driven businesses like auto service, HVAC or dental, a spot heard today may matter when the brakes start squealing next month. Short reads miss it.
The radio ROI formula for multi-location brands
The formula itself is simple. Getting honest inputs is the work.
- Extra customers. Customers your locations served because of the radio, over and above what they’d have served anyway. This comes from the market comparison described below.
- Contribution per customer. Average ticket times gross margin, not revenue. A $380 repair order at a 45% margin is worth $171, not $380.
- Total radio cost. Media, plus production, talent fees for host-read spots, and agency commission. Leaving production out flatters radio by a few percent; leaving talent out can flatter it by much more.
ROI = (extra customers × contribution per customer − total radio cost) ÷ total radio cost. An ROI of zero means the radio paid for itself and nothing more. For more on turning that into a number finance will accept, see our guide to proving marketing ROI to a CFO.
Map station coverage to your locations first
Before you pick test markets, get the coverage map for every station on the buy. Then sort each location into one of three groups: inside the target metro and inside the signal (test), outside the target metro but still inside the signal (spill), and outside the signal (possible comparison).
Spill locations get dropped from both groups. They heard the ads, so they can’t be comparison, and they weren’t the target, so including them as test muddies the read. On a big-stick AM or FM station the spill group can be surprisingly large.
How to set up a radio test across your markets
- Pick three or more test metros. Pair similar metros on the last year of customer counts, then flip a coin for which gets radio. One test metro is an anecdote.
- Keep at least twice as many dark metros. The twin is built from these. More candidates means a better fit.
- Run one flight of six to eight weeks. That’s long enough for frequency to build and short enough to keep the rest of the calendar clean. Add a four-week post-period with no radio anywhere.
- Keep the schedule steady. Same weekly spot count and daypart mix through the flight. Changing it halfway leaves you with two small tests instead of one usable one.
- Check affidavits. Stations send proof of which spots aired. Pre-empted spots are common, and a flight that ran at 70% of plan should be read as a 70% flight.
- Size the test before booking. Measure how much your test metros wander against their twin week to week. If the lift you hope for is smaller than about 2.8 times that noise divided by the square root of the flight length, spend more in fewer metros.
If you already run radio everywhere, flip the design: go dark in a few matched metros for one flight and see whether those locations fall below their twin. It’s a harder conversation with the station rep. It’s also the fastest way to find out whether a long-running buy still earns its place.
Worked example: a 34-shop auto service chain
This example is illustrative. A 34-shop auto service chain (oil changes, brakes, tires) runs a six-week radio flight in three metros that hold 11 shops. Media is $8,000 a week per metro, so $144,000, plus $6,000 for production and talent: $150,000 in total. The other metros, minus spill shops, form the comparison. Each test shop writes about 260 repair orders a week, so the 11 test shops write 2,860.
| Line | Value | Working |
|---|---|---|
| Test shops, repair orders per week before | 2,860 | 11 × 260 |
| Change in test metros during the flight | +5.2% | Versus their own pre-period |
| Twin’s change over the same weeks | +1.0% | Weighted blend of dark metros |
| Lift during the flight | +4.2% | 5.2% − 1.0% |
| Extra repair orders, 6-week flight | 721 | 2,860 × 4.2% × 6 |
| Extra repair orders, 4 post weeks at +1.5% | 172 | 2,860 × 1.5% × 4 |
| Total extra repair orders | 893 | 721 + 172 |
| Cost per extra repair order | $168 | $150,000 ÷ 893 |
| Contribution per repair order | $171 | $380 average × 45% margin |
| First-visit ROI | +1.8% | (893 × $171 − $150,000) ÷ $150,000 |
On first visits alone, the radio roughly paid for itself. That’s where most radio post-mortems stop, and why many end with “it broke even, so let’s try something else.”
First-visit ROI versus 12-month ROI
Auto service is a repeat business. So is dental, fitness, HVAC maintenance and most restaurants. If some of the extra customers are new to you, their later visits belong in the ROI too, as long as you count them carefully.
Continuing the example: the chain’s system flags first-time customers, and the test shops’ new-customer count rose by enough to account for about 60% of the extra repair orders, or 536 new customers. From the chain’s own history, a new customer comes back about 1.4 more times in the following year. That’s 536 × 1.4 × $171 = $128,318 of extra contribution. Add the first-visit $152,703 and the 12-month value is $281,021, an ROI of about 87%.
Two cautions. Only count repeat visits from new customers. Extra visits from existing customers may be pulled forward from next month, and the post-period only partly catches that. And use your own return rate, measured from past cohorts, not the most optimistic one in the deck. If you can’t measure it, report the first-visit ROI and say so.
Do promo codes and call tracking measure radio ROI?
They give you a floor. A radio-only code redeemed 140 times tells you at least 140 people acted, and a unique tracking number in the spot catches some callers. Both are worth running because they’re cheap and they help compare creative. Neither gives you the total, because most people who act on a radio spot search your name or just show up.
Watch for the reverse problem too. Branded search and Google Business Profile calls usually rise in radio markets during a flight. Your search reports will credit those to search. When you add up channel ROIs, make sure the same customer isn’t counted by both.
Common mistakes when measuring radio ROI
- Using revenue instead of margin. A radio campaign with a 3:1 “return” on revenue can lose money at a 30% margin.
- Forgetting production and talent. Host-read endorsements can carry a talent fee on top of media. It’s part of the cost.
- Ignoring spill. Leaving spill locations in the comparison makes the twin look better than it should and understates radio.
- Reading the flight weeks only. In the example the post-period added 172 repair orders, almost a fifth of the total.
- Counting every repeat visit. Repeat value belongs to genuinely new customers only, using a return rate you’ve measured.
- Changing three things at once. New radio, a new offer and a price change in the same metros make an unreadable test.
- Trusting the ratings book as proof. Audience estimates tell you who could have heard the spot. They aren’t evidence anyone came in.
When radio ROI can’t be measured this way
If every location sits in one metro, there’s no dark metro to compare against. You can still alternate flights with dark periods and compare, but the season and everything else that happened in those weeks gets mixed in, so treat it as directional. If you have fewer than about 15 locations across your test metros, expect a wide range.
And if your radio is mostly sponsorships (the traffic report, the high-school football broadcast), the spend is often too small to move weekly counts by a measurable amount. That doesn’t make it worthless. It makes it a community line item, and it should be budgeted that way rather than defended with numbers it can’t produce.
Questions people ask about radio advertising ROI
How do you calculate radio advertising ROI?
Take the extra customers radio brought to your locations, multiply by the contribution each one is worth (ticket times margin), subtract the total radio cost including production and talent, and divide by that cost. The extra customers come from comparing locations in radio markets with a weighted twin of locations in markets without radio.
How long should a radio test run?
Six to eight weeks on air, plus a four-week post-period, after at least six months of clean history. Radio builds with frequency and its effect lingers, especially for need-driven services, so shorter flights and reads that stop when the spots stop tend to understate it.
Is radio advertising worth it for multi-location businesses?
It can be, especially for repeat businesses where new customers come back. The only way to know for your brand is a market test. Radio often looks break-even on first visits and profitable over a year, so decide which horizon you’re judging before the test starts.
Do radio promo codes measure ROI accurately?
No. They measure the minority of customers who remember the code and the staff who record it. Use codes as a floor and a creative comparison, then use a location-level market comparison for the real total.
What is a good ROI for radio advertising?
Anything above zero on the horizon you care about, after margin and full costs. There’s no reliable industry benchmark worth copying, because margins and repeat rates vary so much. Compare radio’s cost per extra customer with your other channels, measured the same way.
Radio deserves a fair hearing: a real test, honest costs, and a horizon that matches how your customers buy. Then keep it or cut it on the numbers.
Next step
Put your own locations through the comparison.
Your first location report is free at any size. Bring a CSV of weekly sales and spend, or connect your POS and ad accounts.
Written by The Countlift team at First Byte in The Woodlands, Texas. Questions or corrections? Email hello@countlift.com.