A 30-second AM/FM radio spot delivers 94% of the ad recall of a 60-second spot, at half the airtime (MARU/Vision Critical, May 2017). Length turns out to be one of the weaker levers in radio creative: two-minute ad breaks keep 99% of the lead-in audience while six-minute breaks keep 85% (Cumulus Media Westwood One analysis, 2025), and a UK audio ad that crams 195 words into a minute scores 32.9% on Creative Standout versus 39.6% at 147 words (Colourtext, 2023). Meanwhile radio’s benchmark weighted return is $2.00 per dollar, second only to social, even though just 46% of marketers rate the medium as effective (Nielsen, Audio Today 2026). We aggregated data from Nielsen, the Radio Advertising Bureau, Radiocentre, the Journal of Advertising Research, TAM AdEx, SoundOut, System1, dentsu and a dozen other sources; for the revenue side of the business, see our radio industry statistics for 2026.
TL;DR
- A :30 radio spot earns 94% of the ad recall and 80% of the purchase intent of a :60 (MARU/Vision Critical, 2017).
- 67% of radio ad volume ran in the 20 to 40 second band in 2025; only 4% ran 40 to 60 seconds (TAM AdEx, 2026).
- Radio breaks kept more than 93% of the lead-in audience across 17,896,325 breaks (Arbitron, Media Monitors and Coleman Research, 2011).
- Two-minute breaks retain 99% of the lead-in audience; six-minute breaks retain 85% (Cumulus Media Westwood One, 2025).
- Mechanical ad avoidance benchmarked at just 3%, about one-tenth of older estimates (Journal of Advertising Research, 2020).
- Avoidance peaks at the fourth ad in longer breaks (Journal of Advertising Research, 2024).
- UK audio ads average 169 words per minute; cutting 10 words lifts Creative Standout by 1% (Colourtext, 2023).
- Sonic logos with the brand name are 9 times more effective at attribution than pure musical cues (SoundOut Index, 2025).
- 41% of audio ads produced correct brand recall versus a 38% norm for mostly video ads (dentsu and Lumen, 2023).
- Radio returns a weighted $2.00 per dollar, second to social at $2.22 (Nielsen, Audio Today 2026).
- Broadcast radio ROI runs 22% above the all-media average over two years, 4th of 11 media (WPP Media, High Gain Audio 2025).
1. How Long Radio Spots Run: The Inventory Mix
Where volume data is public, the 60-second spot is now a niche unit and the market has settled on a short standard: about two-thirds of radio ad volume sits in the 20 to 40 second band, and fewer than one spot in twenty runs between 40 and 60 seconds, according to TAM AdEx data reported by exchange4media in March 2026. Those figures describe India’s radio market, the only large market with a recent public breakdown we could verify; no U.S. or UK monitoring firm publishes an equivalent split.
The interesting movement is at the long end. Spots over 60 seconds doubled their share, though from a base of just 1%, so the long-form revival is still a rounding error in total inventory. Format follows length too: in the Radio Recall Research database republished by the RAB, most 30-second spots were straight announcer reads, while 60-second spots leaned much more on dialogue scenes, which matters for recall (see section 2).
| Metric | Value | Source |
|---|---|---|
| Share of radio ad volume in spots of 20 to 40 seconds (2024 vs 2025) | 68% vs 67% | TAM AdEx, reported by exchange4media 2026 |
| Share of radio ad volume in spots under 20 seconds (2024 and 2025) | 27% in both years | TAM AdEx, reported by exchange4media 2026 |
| Share of radio ad volume in spots of 40 to 60 seconds (2024 and 2025) | 4% in both years | TAM AdEx, reported by exchange4media 2026 |
| Share of radio ad volume in spots longer than 60 seconds (2024 vs 2025) | 1% vs 2% | TAM AdEx, reported by exchange4media 2026 |
| Year-on-year change in the frequency of radio ads longer than 60 seconds | +9% | TAM AdEx, reported by exchange4media 2026 |
| Split between ads shorter and longer than 60 seconds at Red FM | About 90:10 | Red FM, broadcaster’s own figure, 2026 |
| Share of ads shorter than 60 seconds at Music Broadcast | 99% | Music Broadcast, broadcaster’s own figure, 2026 |
| Share of 30-second vs 60-second spots using a straight announcer or sung-jingle format | 64% vs 44% | Radio Recall Research, republished by the RAB (undated) |
Context: the TAM AdEx bands (20 to 40 seconds, under 20 seconds) do not map exactly onto the :15, :30 and :60 units sold in the U.S., so treat them as a directional picture of the global mix, not a U.S. rate card.
2. Recall and Brand Lift by Spot Length
The evidence points in one direction: doubling the length of a radio spot does not double what listeners remember. A :30 delivers 94% of the ad recall and 80% of the purchase intent of a :60 in the MARU/Vision Critical study cited in the Radio Connects Creative Best Practices for Audio deck. Recall barely moves, while the persuasion metric loses a fifth, which suggests the extra 30 seconds buys argument, not memory.
Older controlled work agrees on brand recall but splits on message detail. In a 2006 Saint Joseph’s University experiment with 85 listeners aged 18 to 24, brand recall did not differ significantly by length, yet recall of the ad’s message was significantly higher for 60-second spots. The practical read: a :30 is enough to plant a brand name, while a :60 earns its cost only when the message itself has to be remembered. Most recent available data for controlled tests: MARU/Vision Critical 2017 and Saint Joseph’s University 2006.
| Metric | Value | Source |
|---|---|---|
| Ad recall of a 30-second AM/FM spot as a share of a 60-second spot | 94% | MARU/Vision Critical, May 2017 (cited by Radio Connects 2023) |
| Purchase intent of a 30-second AM/FM spot as a share of a 60-second spot | 80% | MARU/Vision Critical, May 2017 (cited by Radio Connects 2023) |
| Average recall score, 60-second vs 30-second spots | 19.0 vs 15.2 | Radio Recall Research, republished by the RAB (undated) |
| Brand recall, 60-second vs 30-second spots (n = 85, ages 18 to 24) | Not significantly different | David Allan, Saint Joseph’s University, 2006 |
| Message recall, 60-second vs 30-second spots (same 85 listeners) | Significantly greater for 60-second spots | David Allan, Saint Joseph’s University, 2006 |
| Frequent radio listeners who say they hear the full ad when it comes on | 61% | Audacy and Veritonic, January 2022 |
| Respondents reacting very similarly to 30-second and 60-second ads (thinking about how to use the product) | More than two-thirds | Audacy and Veritonic, January 2022 |
| Survey responses per campaign in a 15-second vs 30-second audio test (the 30-second spot led on recall, the 15-second spot on purchase intent) | 27,000+ | SiriusXM Media, vendor’s own case study (undated) |
Outlier: Audacy and Veritonic’s 2022 test found the 30-second spot led on most indicators overall, but retail ads tended to succeed at just 15 seconds, so the optimal length depends on the category.
3. Ad-Break Length, Pod Position, and Audience Retention
The biggest myth in radio buying is that listeners flee when the spots start. Commercial-minute data says otherwise: breaks kept more than 93% of the lead-in audience (the audience tuned in just before the break) across 17,896,325 breaks at 866 stations, according to the Arbitron, Media Monitors and Coleman Research study summarized by Radio Ink. These ratings come from the portable people meter (PPM), a wearable device that picks up inaudible codes embedded in station audio.
What does cost audience is the size of the pod (the cluster of ads in one break). Retention falls from 99% in two-minute breaks to 85% in six-minute ones, per the Cumulus Media Westwood One analysis reported by Radio World in July 2025. Academic work matches the shape: a 2024 Journal of Advertising Research study found mechanical avoidance (physically switching station or turning the radio off) follows an inverse U, peaking at the fourth ad in longer breaks. The lever for advertisers is placement in short pods rather than spot length.
| Metric | Value | Source |
|---|---|---|
| Commercial breaks analysed in the largest U.S. PPM break study | 17,896,325 breaks at 866 stations in 48 PPM markets | Arbitron, Media Monitors and Coleman Research, 2011 |
| Share of the lead-in audience retained during commercial breaks | More than 93% | Arbitron, Media Monitors and Coleman Research, 2011 |
| Lead-in audience retained, spoken-word vs music formats | 99% vs 88% | Arbitron, Media Monitors and Coleman Research, 2011 |
| Lead-in audience retained, two-minute vs six-minute breaks | 99% vs 85% | Cumulus Media Westwood One analysis, 2025 |
| Mechanical avoidance benchmark, Canadian PPM data | 3% across conditions, about one-tenth of earlier estimates of up to one-third | Michelon et al., Journal of Advertising Research 2020 |
| Ad position at which mechanical avoidance peaks in longer breaks | 4th ad | Michelon et al., Journal of Advertising Research 2024 |
| Mechanical avoidance, music stations vs talk stations | 4% vs 0% | Michelon, University of Adelaide doctoral thesis 2019 |
| Share of PPM listening occasions lasting three or four minutes | 23% | Nielsen, 2025 |
Context: the 23% figure explains why break length is back on the agenda. Since the PPM listening qualifier moved to three minutes in 2025, short listening occasions now count, which rewards stations that schedule more frequent, shorter breaks. Most recent available data for the large U.S. retention study: 2011; the 2020 Canadian benchmark is the newest peer-reviewed measurement.
4. Copy Density, Message Count, and Recall
If length is a weak lever, word count is a strong one. UK audio ads average 169 words per minute, and Colourtext’s analysis of transcribed AM/FM campaigns found that every 10 words cut raises Creative Standout (Colourtext’s ad-testing score for how much an ad stands out) by 1%, as detailed in Westwood One’s November 2023 summary. The penalty for dense copy shows up in web traffic, not just in test scores.
This reframes the :30 versus :60 debate. Word rate, not duration, is the variable that moved response in this dataset, and adding messages compounds the problem: the first of four messages in one ad keeps only 43% of the recall of a single-message ad, per Kantar Millward Brown research cited by Radio Ink in April 2026. The creative itself carries the result: Circana’s analysis of about 450 campaigns credits creative with 49% of sales lift.
| Metric | Value | Source |
|---|---|---|
| Average words per minute in UK audio commercials | 169 | Colourtext word-density analysis, 2023 |
| Average Creative Standout at 147 vs 195 words per minute | 39.6% vs 32.9% | Colourtext word-density analysis, 2023 |
| Average web response rate at 147 vs 195 words per minute | 1.8% vs 1.3% | Colourtext word-density analysis, 2023 |
| Change in Creative Standout for every 10 words removed | +1% | Colourtext word-density analysis, 2023 |
| Web visitors lost when 24 extra words are added to a 30-second ad on a 33-million-reach campaign | About 160,000 (a 27% drop in response) | Colourtext word-density analysis, 2023 |
| Recall of the first message in a four-message ad, as a share of a single-message ad | 43% | Kantar Millward Brown, cited 2026 |
| Share of sales lift attributed to creative quality (about 450 campaigns) | 49% | Circana, cited 2026 |
Caveat: this link between word count and response rests on 22 transcribed AM/FM campaigns that aired between 2020 and 2022, a small base, so the effect sizes are indicative.
5. Sonic Branding, Music, Voice, and Emotion
A sonic logo only works if listeners can name the brand behind it, and most cannot. Consumers who claim to recognize a sonic logo are right only 43% of the time, and without the brand name spoken in it accuracy collapses to 18%, according to the SoundOut Index 2025, built on more than 70,000 consumers and 174 brands. The fix is cheap: logos that say the brand name are 9 times more effective at attribution than pure musical cues.
Consistency beats novelty. In Radiocentre’s Big Audio Datamine meta-analysis of 1,002 campaigns, eight of the ten creative attributes most linked to higher ad awareness were recurring audio elements such as music, voices, straplines and sonic devices. Emotion matters as much as identity: System1 and Radiocentre’s Listen Up! study (October 2023) found music adds 0.7 to an ad’s Star Rating (System1’s 1 to 5 predicted-growth score), while System1 and Commercial Radio and Audio found half of Australian radio ads score just 1 Star. On voice, a 2024 listener study found calm delivery was strongly preferred over energetic delivery for the same script, which is relevant for brands now testing AI voice generators for spot production and for teams following AI voiceover adoption statistics.
| Metric | Value | Source |
|---|---|---|
| Brand attribution for the top three U.S. sonic logos (Arby’s, AutoZone, Liberty Mutual) | Over 90% | SoundOut Index 2025 |
| Attribution effectiveness of sonic logos with the brand name vs pure musical cues | 9 times higher | SoundOut Index 2025 |
| Recognition accuracy among consumers who claim to recognize a sonic logo, overall vs logos without a brand name | 43% vs 18% | SoundOut Index 2025 |
| Effectiveness of audio branding vs visual branding; share of advertisers using audio branding | 3 times on average; 8% | Ipsos, The Power of You 2020 |
| Top-ten creative attributes linked to ad-awareness uplift that are consistent audio elements | 8 of 10 | Radiocentre, Big Audio Datamine 2022 |
| Average Star Rating gain when music is included in a radio ad | +0.7 | System1 and Radiocentre, Listen Up! 2023 |
| Average brand fluency of Australian audio campaigns (listeners naming the brand after one hearing); share of radio ads scoring 1 Star | 68%; 50% | System1 and Commercial Radio and Audio, February 2025 (56 campaigns, 8,000+ listeners) |
| Acoustic parameters compared between likable and unlikable ad voices; preferred style | 88 eGeMAPS parameters; calm delivery preferred | Pajupuu, Altrov and Pajupuu, Estonian Papers in Applied Linguistics 2024 |
Context: System1 classes humour as one of several right-brain features (with characters, story and music) that raise emotional response, but none of the studies we reviewed published a humour-only recall figure, so we did not report one. The voice-style study used one gender-neutral mock advertisement read in two styles; its listener sample size was not stated in the abstract.
6. Radio ROI: What the Econometric Studies Say
Radio’s measured return and its reputation have drifted apart. The Global Compass benchmark puts radio’s weighted ROI at $2.00 per dollar, second only to social at $2.22, yet only 46% of marketers in the 2025 annual marketing survey rated radio as effective, the lowest of any channel, as reported in the Nielsen Audio Today 2026 report and by Radio Ink in March 2026. The gap is a pricing opportunity for buyers who trust models over perception.
UK econometrics tell the same story in profit terms. The High Gain Audio study, built by WPP Media and Radiocentre on Thinkbox’s Profit Ability 2 dataset, found broadcast radio beats the all-media average both short-term and over two years, and Radio Today’s September 2025 report lists the per-pound figures below. Peter Field’s IPA Databank work adds a market-share angle covering 2000 to 2022.
| Metric | Value | Source |
|---|---|---|
| Weighted ROI per dollar spent, radio vs social media | $2.00 vs $2.22 | Nielsen Global Compass Benchmark, in Audio Today 2026 |
| Marketers who rate radio as an effective channel (lowest of all channels) | 46% | Nielsen Annual Marketing Report 2025 |
| Short-term profit ROI per pound (up to 3 months), broadcast radio vs all-media average | £2.30 vs £1.87 | WPP Media and Radiocentre, High Gain Audio 2025 |
| Full-term profit ROI per pound (3 months to 2 years), broadcast radio vs all-media average | £5.00 vs £4.11 | WPP Media and Radiocentre, High Gain Audio 2025 |
| Broadcast radio two-year ROI vs all-media average; rank among 11 media | +22%; 4th | WPP Media, High Gain Audio 2025 |
| Return on marketing investment, IPA case studies using AM/FM radio vs not | +23% | Peter Field, IPA Databank 2000 to 2022 analysis |
| Average market share, brands using AM/FM radio vs not | 32% vs 25% | Peter Field, IPA Databank 2000 to 2022 analysis |
| Increase in profit effects, IPA case studies using AM/FM radio vs not | +42% | Peter Field, IPA Databank 2000 to 2022 analysis |
Context: the Field figures compare more than 1,200 case studies entered for the IPA Effectiveness Awards, while the $2.00 benchmark comes from marketing mix models; the methods differ, so the numbers are not directly comparable. The Field results are summarized in Westwood One’s March 2025 business-case summary, a radio sales house’s own compilation.
7. Audio vs TV and Digital: Recall, Attention, and the Host-Read Contrast
Audio does not lose the recall contest to screens. 41% of audio ads generated correct brand recall, against a 38% norm for dentsu’s mostly video database and 36% for display, in the dentsu and Lumen attention study reported by Ad Age in August 2023. What separates the media is price: per thousand attentive impressions, AM/FM radio cost 40 cents against $3.00 for TV.
Creative quality also travels across media. System1’s testing gives radio and TV ads the same 2.4 average Star Rating, and UK campaign tracking shows a 49.4% ad-awareness uplift among commercial radio listeners versus non-listeners. For contrast with podcasts, where the host usually reads the ad, Nielsen found more podcast listeners than AM/FM listeners say a host-delivered ad gets more of their attention than a pre-produced one; how radio personalities are now repurposed into podcasts is tracked in our radio-to-podcast simulcast statistics.
| Metric | Value | Source |
|---|---|---|
| Audio ads producing correct brand recall vs dentsu norm (mostly video) vs display | 41% vs 38% vs 36% | dentsu and Lumen audio attention study, 2023 |
| Attentive seconds per 1,000 impressions, audio vs dentsu norms | 10,126 vs 6,501 | dentsu and Lumen audio attention study, 2023 |
| Cost per thousand attentive impressions, AM/FM radio vs TV vs online display | $0.40 vs $3.00 vs $9.70 | dentsu and Lumen audio attention study, 2023 |
| AM/FM radio attentiveness as a share of linear TV attentiveness | 85% | Adelaide attention metric, reported August 2024 |
| Average Star Rating, radio vs TV ads; average Fluency, radio vs TV | 2.4 vs 2.4; 82 vs 83 | System1 and Radiocentre, Listen Up! 2023 |
| Ad awareness uplift, commercial radio listeners vs non-listeners (841 campaigns) | +49.4% | Radiocentre, Big Audio Datamine 2022 |
| Increase in unaided ad recall among in-market consumers, audio ad formats combined (3,500+ U.S. adults) | +22 points | Omnicom Media Intelligence and iHeartMedia, April 2026 |
| Listeners saying host-delivered ads get more attention than pre-produced ads, podcast vs AM/FM listeners (n = 1,000 U.S. adults) | 56% vs 44% | Nielsen consumer study, March 2022 |
Caveat: the attention figures were commissioned or publicized by audio sellers (Cumulus Media Westwood One summarized the Adelaide comparison), and the Omnicom and iHeartMedia study was co-sponsored by iHeartMedia, a radio broadcaster. Most recent available data for the dentsu study: 2023.
Summary: Radio Advertising Spot Length and Recall by the Numbers
| Metric | Value | Source |
|---|---|---|
| Ad recall of a 30-second AM/FM spot as a share of a 60-second spot | 94% | MARU/Vision Critical, 2017 |
| Purchase intent of a 30-second AM/FM spot as a share of a 60-second spot | 80% | MARU/Vision Critical, 2017 |
| Share of radio ad volume in spots of 20 to 40 seconds, 2025 | 67% | TAM AdEx, 2026 |
| Share of radio ad volume in spots longer than 60 seconds, 2025 | 2% | TAM AdEx, 2026 |
| Brand recall, 60-second vs 30-second spots (n = 85) | Not significantly different | Saint Joseph’s University, 2006 |
| Share of the lead-in audience retained during commercial breaks | More than 93% | Arbitron, Media Monitors and Coleman Research, 2011 |
| Lead-in audience retained, two-minute vs six-minute breaks | 99% vs 85% | Cumulus Media Westwood One, 2025 |
| Mechanical avoidance benchmark | 3% | Journal of Advertising Research, 2020 |
| Ad position at which avoidance peaks in longer breaks | 4th ad | Journal of Advertising Research, 2024 |
| Average words per minute in UK audio commercials | 169 | Colourtext, 2023 |
| Recall of the first message in a four-message ad vs a single-message ad | 43% | Kantar Millward Brown, cited 2026 |
| Sonic logos with the brand name vs pure musical cues, attribution effectiveness | 9 times higher | SoundOut Index 2025 |
| Recognition accuracy for sonic logos without a brand name | 18% | SoundOut Index 2025 |
| Audio ads producing correct brand recall vs dentsu norm | 41% vs 38% | dentsu and Lumen, 2023 |
| Weighted ROI per dollar, radio vs social media | $2.00 vs $2.22 | Nielsen Audio Today 2026 |
| Cost per thousand attentive impressions, AM/FM radio vs TV | $0.40 vs $3.00 | dentsu and Lumen, 2023 |
| Broadcast radio two-year ROI vs all-media average | +22% | WPP Media, High Gain Audio 2025 |
Methodology and Sources
- TAM AdEx radio ad-duration data, as reported by exchange4media, March 25, 2026.
- MARU/Vision Critical (May 2017), cited in Radio Connects, Creative Best Practices for Audio, January 2023.
- Radio Recall Research, Characteristics of Radio Commercials and Their Recall Effectiveness, republished by the Radio Advertising Bureau.
- David Allan, Saint Joseph’s University, 30-second vs 60-second radio ad study, May 4, 2006.
- Audacy and Veritonic, ad length study, January 6, 2022; SiriusXM Media, :15 vs :30 case study (vendor’s own).
- Arbitron, Media Monitors and Coleman Research commercial-break studies (2005 and 2011), summarized in Radio Ink, May 3, 2020.
- Cumulus Media Westwood One break-length analysis and Nielsen listening-occasion data, reported by Radio World, July 27, 2025.
- Michelon, Bellman, Faulkner, Cohen and Bruwer, A New Benchmark for Mechanical Avoidance of Radio Advertising, Journal of Advertising Research 2020 and Why Shorter Advertisement Breaks Reduce Radio Advertisement Avoidance, Journal of Advertising Research 2024; Michelon, University of Adelaide doctoral thesis, 2019.
- Colourtext word-density analysis (Jason Brownlee), via Cumulus Media Westwood One, November 13, 2023; Kantar Millward Brown and Circana, via Radio Ink, April 28, 2026.
- SoundOut, SoundOut Index 2025, October 13, 2025.
- Radiocentre, Big Audio Datamine, June 13, 2022 (includes the Ipsos 2020 audio-branding figures).
- System1 and Radiocentre, Listen Up! Emotion’s Defining Role in Audio Advertising Effectiveness, October 16, 2023; System1 and Commercial Radio and Audio, Australian audio research, February 11, 2025.
- Pajupuu, Altrov and Pajupuu, Preference of voice and performance style in radio advertising, 2024.
- Nielsen, Audio Today 2026: How America Listens, with ROI and marketer-perception figures confirmed in Radio Ink, March 17, 2026.
- WPP Media and Radiocentre, High Gain Audio, with per-pound figures from Radio Today, September 23, 2025 and rankings from Cumulus Media Westwood One, November 10, 2025.
- Peter Field IPA Databank analysis (2000 to 2022 case studies), via Cumulus Media Westwood One, March 17, 2025.
- dentsu and Lumen audio attention study, via Ad Age, August 1, 2023 and Cumulus Media Westwood One, August 7, 2023; Adelaide comparison via Cumulus Media Westwood One, August 12, 2024.
- Omnicom Media Intelligence and iHeartMedia, Turning Volume into Value, via Radio Online, April 28, 2026; Nielsen host-read attention study (March 2022), via Cumulus Media Westwood One, July 25, 2022.
- Data watch: (1) No public U.S. or UK breakdown of radio inventory by :15, :30 and :60 was found; section 1 relies on India’s TAM AdEx bands, which do not match U.S. units. (2) The 17,896,325-break retention study dates from 2011 (Arbitron era); Radio World’s 2025 article attributes it to Nielsen, Media Monitors and Coleman Research, so the 99% vs 85% break-length figures rest on that older dataset. (3) The 43% four-message recall figure is attributed to Kantar Millward Brown by Westwood One and Radio Ink, but to the Nielsen Podcast Brand Effect Study 2020 in the Radio Connects deck. (4) The Profit Ability 2 base is described as 1.8 billion pounds and 141 brands by Radiocentre but 2.2 billion dollars, 142 brands and 10 media channels by Westwood One; Radiocentre’s budget advice moved from 20% of media spend to audio (September 2025) to 25% in the full report. (5) The Radio Recall Research page carries no date and the 2006 and 2017 length tests are old; several attention and ROI studies were commissioned or publicized by radio sellers (Westwood One, Audacy, SiriusXM Media, iHeartMedia, Radiocentre, the RAB), so treat them as the industry’s own research. (6) No humour-specific recall statistic was found in a primary source.
Last updated: November 8, 2026. We update this roundup quarterly, and the next refresh is expected when Nielsen publishes Audio Today 2027 and Radiocentre releases its next Radiogauge or High Gain Audio update.