Radio Advertising: A Complete Guide

Dayparts, and what you are actually buying

Radio is sold by time of day, and the divisions are not arbitrary — they reflect genuinely different listening situations. The names are broadly standard, though exact hours vary by station.

Morning drive, roughly 6am to 10am. The largest audience of the day on most stations and the most expensive inventory. Listeners are in the car, alert, and often listening alone. Strong for immediacy and for offers that can be acted on later the same day. The trade-off is competition: everyone wants this daypart, so you pay for it.

Midday, roughly 10am to 3pm. Heavy workplace listening, longer sessions, and often the best value in the schedule. The audience is smaller than drive time but frequently more attentive and more able to act immediately — a phone or a browser is usually within reach.

Afternoon drive, roughly 3pm to 7pm. The second peak. Listeners are heading home, and the mood is different from the morning: less news-seeking, more unwinding. Good for retail and dining offers where the decision is imminent.

Evening, roughly 7pm to midnight. Much smaller audience, considerably cheaper, and skewed by format. Sports and talk hold evening audiences far better than music formats.

Overnight, midnight to 6am. The least expensive inventory on the station and the most frequently misused. It suits a narrow set of advertisers — genuinely national direct-response offers, or businesses whose customers work nights. For most advertisers, a schedule weighted overnight is cheap because it is weak, not because it is a bargain.

Weekend. A different audience pattern again, with listening spread across the day rather than concentrated at the commute. Useful for retail, events, home improvement and automotive.

The most common scheduling error is buying whatever mix a station offers rather than specifying the dayparts the offer actually needs. A rate that looks efficient because it includes a lot of evening and overnight inventory is not efficient.

Reach and frequency: the two numbers that decide the outcome

Almost every disappointing radio campaign can be traced to a misunderstanding of these two ideas, so they are worth setting out plainly.

Reach is how many different people hear your commercial at least once. Frequency is how many times the average person hears it. A fixed budget buys a trade-off between them: the same money can reach many people once, or fewer people several times.

Radio rewards frequency. A message heard once is background noise; a message heard repeatedly by the same listener over a period of weeks is what produces recall and response. This is why spreading a budget thin across many stations or many markets so reliably fails — it maximises the wrong variable.

The industry term for the point at which repetition starts working is effective frequency. There is no universal number for it, despite what you may be told: it depends on how familiar the brand already is, how complex the message is, how competitive the category is, and how strong the offer is. A well-known brand with a simple offer needs fewer exposures than an unfamiliar one asking for a considered purchase.

What follows practically: decide the minimum weekly weight the campaign needs in one market first, then work out how many markets the budget will support at that weight. Doing it the other way round — choosing markets first and dividing the budget among them — is how campaigns end up airing everywhere and landing nowhere.

Choosing a spot length

Sixty seconds gives room to establish a problem, present the offer and repeat the response mechanism. It suits anything that has to be explained before it can be acted on — services, considered purchases, unfamiliar brands, direct response with a phone number that needs repeating.

Thirty seconds is the workhorse. Enough for one clear idea, one offer and one call to action, provided the writing is disciplined. Most local campaigns live here.

Fifteen seconds carries one idea and one instruction. It works as a reminder inside a schedule that already has longer spots doing the explaining. It does not work as the only length for anything that needs understanding first. Availability is also limited — not every station sells fifteens, and some restrict them by daypart.

Note that pricing does not scale neatly with length. A sixty is rarely exactly twice a thirty, and a fifteen is often much closer to a thirty than the ratio suggests, because the station is giving up a break position either way. Ask for the lengths you are considering quoted side by side on the same schedule.

How radio audiences are measured

Understanding where station audience figures come from makes it much easier to judge whether a proposal is telling you something meaningful.

In the United States, radio audience estimates are produced primarily by Nielsen Audio, using two different methods depending on market size.

The Portable People Meter is used in larger markets. Panellists carry a pager-sized device that passively detects inaudible codes embedded in station broadcasts. Because it records exposure automatically rather than relying on memory, it captures short listening occasions the older method missed.

The paper diary is used in smaller markets. Respondents write down what they listened to across a seven-day period. It depends on recall and on people completing the task, which introduces well-understood limitations.

A change worth knowing about: in May 2026 Nielsen outlined a hybrid methodology for diary markets, adding a shorter survey instrument alongside the existing diary and using modelling to project full-week listening. Nielsen has said the diary itself remains in place, and that data from the new approach is expected to begin appearing in 2027. If you buy in diary markets, expect some discontinuity in the numbers when it lands, and treat year-over-year comparisons across that transition with care.

The terms you will encounter on a proposal:

• AQH (average quarter-hour) persons — the average number of people listening during a typical fifteen-minute span. This is the number station rates are effectively built on.
• Cume — the cumulative audience, meaning the number of different people who listened at least once during a period. Cume is a reach measure; AQH is an intensity measure. Sellers quote whichever flatters the station.
• TSL (time spent listening) — how long the average listener stays. High cume with low TSL means a lot of people passing through briefly.
• GRPs (gross rating points) — reach multiplied by frequency, the total weight of a schedule.
• CPP and CPM — cost per rating point and cost per thousand listeners. Efficiency measures, and both are meaningless if the listeners counted are not plausible customers.

The single most useful habit when reading audience data is to ask which measure is being quoted and why that one was chosen. A station with a large cume and weak AQH will lead with cume. That is not dishonest, but it is selective, and an independent buyer is free to ask for the other number.

Four things worth a follow-up question

None of these are dishonest. All of them are incomplete, and the follow-up is where the useful information lives.

"We reach 500,000 listeners a week." That is almost certainly cume — different people who tuned in at least once. Ask what the AQH figure is for the dayparts being proposed, and what the audience composition looks like against your customer.

"We'll produce your commercial for free." Genuinely free, and produced by people whose employer benefits from you spending on their air, using their announcers. Fine for a simple local offer. Ask who writes the copy and whether you can use the finished spot on other stations.

"We can throw in some added value." Ask what it is worth and how it is delivered. Promotional mentions and digital extras are frequently introduced specifically to avoid moving on rate. Some of it is worth real money; much of it is not.

"This rate is only available if you commit today." Sometimes true, in the sense that inventory genuinely moves. Ask what specifically changes tomorrow. A rate that evaporates under a single question was not a rate.

How radio fits with everything else you run

Radio is rarely the only thing an advertiser is doing, and it behaves differently from the channels it sits alongside. Three interactions matter.

Radio drives search. A listener cannot click a radio commercial. What they can do is remember a name and look it up later, which means a working radio campaign typically shows up first as a rise in branded search and direct traffic rather than as attributable clicks. If you are running paid search on your own brand terms, radio will push that volume up and your search reporting will take the credit.

Last-click attribution systematically undercounts it. This follows from the point above, and it is the single most expensive misunderstanding in audio. A campaign that is genuinely working can look flat in an analytics dashboard, because the visit that converted arrived through search or direct. Advertisers cancel effective schedules over this regularly. Agree before launch how radio will be judged — baselines, branded search, call volume, time-correlation against the airing log — rather than defaulting to a model that was never built to see it.

It compounds with what people already know about you. Radio works considerably harder for a brand that a listener has encountered elsewhere, because the commercial confirms something rather than introducing it. That argues for running audio alongside other visible activity rather than in isolation, and it is part of why a short standalone radio test on an unfamiliar brand tends to read as a failure.

None of this means radio cannot be measured. It means the measurement has to be designed for how audio actually works. See how a schedule is planned and tracked.

Before you launch: a checklist

Everything on this list should exist before the first spot airs. Retrofitting any of it is harder and less reliable.

1. A countable objective. Calls, orders, form fills, showroom visits. Not "awareness," unless you have decided how awareness will be observed.
2. A baseline. What your calls, traffic, branded search and sales looked like in the weeks before the campaign. Without it there is nothing to compare against.
3. Tracking numbers. Ideally one per station, at minimum one per market, forwarding to the same line.
4. A sayable response mechanism. A URL or phone number that a person can remember while driving. Clever is worthless here; memorable is everything.
5. Capacity to answer. Who picks up, during which hours, and what happens to calls outside them. A schedule that generates calls nobody answers has bought nothing.
6. A commercial written for the offer. Not a repurposed TV script, and not left to the station as an afterthought.
7. Agreed terms in writing. Placement, preemption, makegoods, cancellation rights and notice periods — settled before the buy, not argued afterwards.
8. A review point. A date when the schedule will be assessed and a shared understanding of what would trigger a change.

A short glossary

Avail — an availability request, or the inventory a station offers in response to one.

Affidavit — a station's certified record of when your commercials actually aired. Check it against what you ordered.

Clearance — on network and syndicated buys, the proportion of affiliate stations that actually carried your spot, and where.

Daypart — a defined block of the broadcast day, priced separately.

Flight — the period a campaign runs.

Makegood — replacement airtime offered when a scheduled spot did not run as ordered.

Preemption — the displacement of a scheduled spot, usually because a higher-paying advertiser took the position or news interrupted programming.

Rate card — a station's published pricing. A starting point rather than a settled price.

Remnant — unsold inventory offered at short notice and lower cost, with correspondingly less control. See remnant radio advertising.

Rotator — a spot scheduled to rotate through a daypart rather than at a fixed time.

Spot — an individual commercial announcement.

Traffic — a station's scheduling department, which decides where in the break your spot falls.

Next steps: what radio advertising costs, how a schedule is assembled, or who would be handling it.

Measurement methodology source: Inside Audio Marketing, 29 May 2026. Berk Marketing is not affiliated with Nielsen. Last reviewed August 2026.