ROI on radio advertising: measure response and contribution

ROI on radio advertising compares matched contribution with the full cost of the radio campaign. Connect responses to realized sales, deduct the cost of delivering those sales, and then subtract placement, production, and tracking expenses. Label the result attribution-based unless separate evidence estimates what would have happened without the campaign.

Define the campaign before matching sales

Start with the station schedule and invoices. Keep the campaign ID, station, market, broadcast dates, daypart, placement cost, credits, and actual delivery evidence together. If a package includes broadcast spots and streaming placements, retain that split rather than assigning all outcomes to whichever format looks stronger.

Specify a response window and a later sales cutoff before comparing markets. A company with a long sales cycle needs time for a tracked inquiry to become an invoice. Keep pending opportunities outside realized revenue, and identify cohorts whose follow-up period remains incomplete.

Audience measures are useful alongside the financial result. The Radio Advertising Bureau's definitions distinguish reach, the people exposed, from frequency, their average exposure count. Neither is a count of paying customers, so neither belongs in the revenue cell.

The marketing ROI versus ROAS guide explains the general cost convention. Here the distinct work is matching radio responses to outcomes by station, market, and period.

Build a response-to-invoice trail

Bring the call or offer-code register, CRM outcome export, invoice and credit register, delivery-cost file, and campaign invoices. Useful matching fields include a response ID, campaign ID, market, response date, order ID, invoice date, net revenue, and variable cost.

One caller may contact the company repeatedly. Deduplicate the inquiry before counting responses, then match the eventual order once. If the buyer mentions multiple stations or channels, use the approved allocation rule and retain an unassigned category when the evidence cannot support a specific assignment.

The marketing attribution guide covers those allocation choices. A response recorded under a radio code is evidence of a tracked path, not proof that the buyer would otherwise have stayed away.

Work a market comparison

An illustrative US commercial-equipment service company reviews July 2026 radio advertising. It records responses during July, matches their completed sales through August 31, 2026, and includes related credits posted through September 30, 2026. The same rule applies to both markets, and each order is assigned once.

Revenue is net of discounts, credits, and sales tax. Variable cost covers the parts, labor, and travel needed for the matched jobs; it excludes campaign expenses. Production is allocated once across the markets, and tracking costs include the agreed campaign measurement charges.

Use this input layout with the header in row 1 and markets in rows 2 and 3:

Attribution-based radio ROI = (matched contribution before marketing − campaign cost) / campaign cost

Add these calculated columns and copy the row-2 formulas down to row 3:

I2 = SUM(F2:H2)
J2 = D2-E2
K2 = J2-I2
L2 = IF(I2=0,NA(),K2/I2)

Percentages are rounded to one decimal place. Total ROI is $11,500 divided by $18,500, not an average of the two market percentages:

I4 = SUM(I2:I3)
K4 = SUM(K2:K3)
L4 = IF(I4=0,NA(),K4/I4)

Check response quality and reconcile the dollars

The total response-to-order rate is 50 / 150 = 33.3%. Campaign cost per tracked response is $18,500 / 150 = $123.33, rounded to cents. These help explain the financial result, but a low response cost can coexist with weak profit if the orders have poor margins.

Order conversion = IF(B4=0,NA(),C4/B4)
Cost per response = IF(B4=0,NA(),I4/B4)
Reconciliation = D4-E4-I4-K4

The reconciliation must be zero: $75,000 − $45,000 − $18,500 − $11,500. Tie response totals to the deduplicated register, matched revenue to the selected invoices and credits, and campaign cost to the station and supplier invoices. Shared costs should reconcile across markets to the original total.

Where it goes wrong

• A call becomes revenue immediately. Unanswered calls, duplicate inquiries, and open opportunities need separate statuses until an actual outcome exists.
• Shared creative is counted repeatedly. Allocate the production invoice across markets; do not charge the whole amount to every station.
• Exposure becomes a sale. Audience estimates remain delivery evidence. They cannot replace the matched order register.
• Campaign timing becomes causal lift. A simultaneous sales increase can have other causes. NIST's randomized-design guidance describes random assignment of treatments to experimental units; an appropriate radio test requires its own design and sufficient evidence.
• One channel claims the entire order again. Use omnichannel ROI to reconcile overlapping channel claims before adding their returns together.

The advertising ROI by medium comparison addresses common cost and observation windows across channels. A radio result should enter that comparison with its attribution limitations attached.

Calculate radio returns from your own files

Bring a sample of your own station, response, revenue, and cost exports to Covirage. Covirage maps the columns; its tools compute and check the figures from the supplied evidence, and the external AI model explains the result with every figure traced to its rows. Create a free Covirage account and bring your campaign records after setting up your business.

See sales insights for the financial outcomes behind the response counts.