The Attention Landscape
The Host-Read Premium: What Podcast Advertising's Trust Effect Actually Proves
Podcast listeners genuinely trust the hosts they choose, and that trust shows up in the data for host-read ads, but no evidence supports the claim that podcast audio beats television, radio, direct mail, or out-of-home advertising.
Part of The Attention Landscape in the Insights library.
Abstract
The US podcast ad market grew 26.4 percent year over year in 2024 to more than $2.4 billion, and host-read ads, the kind where the podcaster reads the sponsor message in their own voice rather than a pre-recorded announcer spot, account for a real and substantial share of that spending. Peer-reviewed research consistently finds that the trust listeners place in a host, what researchers call a parasocial relationship, is associated with higher engagement and recall for host-read ads specifically compared to other podcast ad formats. What the evidence does not support is a stronger claim you may have heard elsewhere: that host-read audio beats every other ad format by a wide, consistent margin. No comparative data exists against television, radio, direct mail, or out-of-home advertising, and at least two peer-reviewed studies found the host-read advantage is inconsistent rather than universal.
Podcast advertising has moved from a scrappy, ad-hoc niche to a genuine media category with its own economics. Industry-reported revenue reached more than $2.4 billion in 2024, up 26.4 percent from the year before, according to the IAB and PwC's annual internet advertising revenue accounting, which is the established, high-confidence figure at the center of this study. That growth is riding on a real and expanding audience: Edison Research's Infinite Dial 2026 survey found that 80 percent of Americans age 12 and older, roughly 230 million people, have listened to a podcast at some point, 58 percent listened in the last month, and 45 percent listened in the last week. A separate Nielsen release found US podcast audiences grew 40 percent over the three years leading up to mid-2022, with roughly half of listeners describing themselves as having started within the prior two years. Inside that growth, the format mix matters: IAB data (traced through a 2018 revenue study, so treat the exact figure as dated rather than current) put host-read ads at 66.9 percent of ad units and, per a separate FY2021 IAB report, 55 percent of ad revenue. Both of those host-read figures reached us only through secondary industry aggregation rather than a freshly retrieved IAB report, so we are treating them as directionally real but not up to date measurements of where that share stands today.
The behavioral case for host-read advertising rests on a specific psychological mechanism: listeners form what researchers call a parasocial relationship with a podcast host, a one-sided but genuinely felt sense of familiarity and trust built over repeated listening. Brinson and Lemon's peer-reviewed work in the Journal of Marketing Communications found that this trust is associated with greater engagement and brand recall, and that the effect is measurably larger for host-read ads than for other ad formats within the same podcast. Moe's two-part 'Podvertising' research found listeners rate host-read ads more favorably and skip them less often than pre-recorded spots, with the effect strongest in comedy and true-crime genres. A study of college-aged listeners by Brooks, Bichard, and Craig in Advertising & Society Quarterly found a clear preference for host-read over pre-recorded ads among that audience, plus a related finding that ads matching the show's own subject matter landed better regardless of who read them. Three independent research teams, three different listener populations, one converging signal: the trust effect is real.
For a business owner, the practical question is not whether podcast trust is real, it clearly is, but where it sits relative to everything else competing for a buyer's attention, and how confidently that can be measured today. Podcast advertising currently occupies a well-documented pocket of a much larger attention terrain: we know its size relative to itself (host-read versus non-host-read), we have one frequently cited but only secondarily sourced comparison to digital display advertising, and we have essentially nothing comparing it to television, radio, direct mail, or out-of-home spend. We also have no dedicated data on small or local business use of podcast sponsorship specifically, so the intuition that a local plumber or med spa could buy trust cheaply through a niche local podcast is plausible, built on the same host-trust research, but currently unproven as its own line item. And critically, there is no equivalent yet of a full census, for audio, of whether a business shows up when someone asks an AI engine like ChatGPT, Perplexity, or Google's AI Overviews about local options. That gap is itself useful information. It tells us this corner of the attention terrain still needs a proper map before anyone should reallocate serious budget into it on faith.
The data, in one read
A market that has grown up fast
US podcast advertising revenue crossed $2.4 billion in 2024, up 26.4 percent from the year before, according to the IAB and PwC's Internet Advertising Revenue Report. That is not a rounding-error category anymore, and it did not get there by accident. Edison Research's Infinite Dial 2026 survey, running continuously since 1998 and fielded again in January 2026, found that 80 percent of Americans age 12 and older, about 230 million people, have listened to a podcast at some point, 58 percent listened in the past month, and 45 percent listened in the past week. A separate Nielsen release from 2022 found US podcast audiences had grown 40 percent over the prior three years, with about half of listeners describing themselves as newer to the format. Put simply, the audience is large, still growing, and increasingly habitual, which is exactly the kind of surface that draws advertising dollars.
US podcast advertising revenue crossed $2.4 billion in 2024, up 26.4 percent from the year before.
What makes a host-read ad different
Most podcast ads fall into one of two buckets: a pre-recorded, announcer-voiced spot dropped into the episode, or a host-read ad where the podcaster reads the sponsor's message in their own voice, often with their own commentary woven in. The distinction matters because listeners do not experience the two the same way. Industry figures traced through IAB's Full-Year 2018 Podcast Ad Revenue Study put host-read ads at 66.9 percent of all ad units running in US podcasts, and a separate IAB report on FY2021 revenue put host-read ads at 55 percent of total ad revenue. We are treating both figures cautiously: neither was retrieved directly from an IAB report during this research, both reached us through industry blog aggregation, and the 66.9 percent figure in particular is drawn from 2018 data, not a current measurement. The direction, that host-read is a large and commercially significant share of the format, is credible. The exact current percentage is not something we can stand behind as fresh.
The trust mechanism, and why it is not marketing folklore
The reason host-read ads command a premium in the first place has a name in the academic literature: the parasocial relationship, a one-sided sense of closeness a listener develops with a host they hear regularly, often for hours a week, sometimes for years. Brinson and Lemon's peer-reviewed study in the Journal of Marketing Communications found that this trust is associated with greater engagement and brand recall, and that the effect shows up more strongly for host-read ads than for other formats in the same show. Moe's two-part 'Podvertising' research, published across the Journal of Radio & Audio Media and the Journal of Economics and Behavioral Studies, found listeners rate host-read ads more favorably and skip them less, with the pattern most pronounced in comedy and true-crime podcasts specifically. A study of college-aged listeners by Brooks, Bichard, and Craig in Advertising & Society Quarterly found a clear preference for host-read over pre-recorded ads among that audience, plus a related finding that ads matching the show's own subject matter landed better regardless of who read them. Three independent research teams, three different listener populations, one converging signal: the trust effect is real.
The trust effect is real, but it is a specific, moderate signal, not the sweeping conversion advantage sometimes claimed for it.
The Nielsen number everyone repeats, and why we are treating it carefully
You have probably seen the claim that podcast ads generate four times the brand recall of digital display and lift purchase intent by 10 percent. That figure traces to Nielsen's 2019 research and shows up consistently in peer-reviewed citation, most directly in Brinson, Lemon, Bender, and Graham's 2023 paper in the Journal of Consumer Marketing. We were not able to independently retrieve Nielsen's original report, its publication URL has since been deprecated, and the figure survives only through consistent academic secondary citation rather than a document we could open and check ourselves. That does not make the number false. Academic researchers citing the same Nielsen figure across multiple papers is a meaningful form of corroboration. But it is a different tier of confidence than a figure we pulled and verified from the source ourselves, and we are labeling it accordingly: a real, credible, widely repeated figure that has not been independently reverified at the primary-source level.
Where the simple story breaks down
If the research stopped with Brinson, Moe, and Brooks, the case for host-read advertising would read as close to settled. It does not stop there. Fischer's 2019 master's thesis at Texas State University ran a controlled audio experiment with 113 college students and found that perceived host credibility did correlate with more favorable brand attitude and higher purchase intent, both statistically significant. But perceived host trustworthiness and attractiveness did not significantly predict brand recall at all, and perceived host expertise predicted unaided recall in the opposite direction from what researchers expected, a genuinely counter-intuitive result. Separately, Wang's 2024 article in the Journal of Radio Studies, verifiable only at the abstract level because the full text sits behind a paywall, found results that diverged from the general assumption that host-read ads always raise perceived credibility, specifically on the measure of advertising receptivity. Neither study overturns the broader trust literature. Both are a genuine caution against treating 'host-read wins' as a uniform law rather than a tendency that depends on which outcome you measure, which listener you ask, and which host is doing the reading.
The trust literature is real, but at least two peer-reviewed studies found the host-read advantage is inconsistent, not universal.
What we don't know: the comparison that was never made
Here is the gap that matters most if you are deciding where to put a marketing budget. Every comparison in the available research measures podcast advertising against itself (host-read versus pre-recorded) or, in the single Nielsen-derived figure, against digital display. Nothing in the retrievable literature compares host-read podcast advertising to television, radio, direct mail, or out-of-home advertising on recall, purchase intent, or conversion. That is a genuinely different question from 'is podcast advertising good,' and it is the specific question a business owner weighing channels actually needs answered. Until that comparison exists, the accurate claim is a scoped one: host-read outperforms non-host-read podcast ads on several behavioral measures, and outperforms digital display specifically according to one widely cited but not independently reverified figure. It has not been shown to out-convert every format measured, because most formats have not been measured against it at all.
The small-business question nobody has answered yet
There is a reasonable-sounding argument that local and small business advertisers should love podcast sponsorship: it is relationship-driven, relatively low-cost per listener, and rides on exactly the kind of host trust the research keeps finding. We looked for data to back that argument specifically, across academic research and industry reporting, and found none. No dedicated dataset quantifies SMB or local-business share of podcast ad spend, and no case study in the literature isolates a small or local advertiser's recall or conversion numbers from a host-read placement. The argument remains plausible, it is built on the same parasocial-trust research already discussed. But it is currently an inference, not a finding, and we are labeling it that way.
The frequency reality behind the trust story
One more figure worth knowing if you are considering a podcast sponsorship: the typical US podcast episode carries a mode of seven ads, ranging from one to twelve, per research out of the University of Hamburg cited in the Journal of Communication Technology. Ad-skipping also appears to be less common among podcast listeners than on other digital ad formats, per Sounds Profitable's 2024 'Ad Nauseam' survey of 1,000 weekly listeners, though the exact skip-rate percentages sat behind that report's download gate and could not be independently verified. Put together, this tells you two things: your host-read ad is one of several competing for the same listener's attention within a single episode, and the audience is, by these researchers' account, meaningfully less prone to skipping than on other digital surfaces. That is a genuinely useful operational detail for anyone evaluating a sponsorship slot, separate from the trust question entirely.
The evidence, in numbers
Key findings, dated and sourced
-
US podcast advertising revenue grew 26.4% year over year (2023 to 2024), surpassing $2.4 billion.
established IAB (Interactive Advertising Bureau) / PwC, IAB/PwC Internet Advertising Revenue Report: Full-Year 2024 Results (2025-04-17)
-
66.9% of ads running in US podcasts are host-read rather than pre-recorded announcer or programmatic spots. This figure traces to IAB's Full-Year 2018 Podcast Ad Revenue Study, not a current dataset, and was accessed only through secondary industry aggregation, not a direct IAB report.
contested IAB (Interactive Advertising Bureau), IAB, Full-Year 2018 IAB Podcast Ad Revenue Study (accessed via secondary citation) (2018-2019 (recirculated by later secondary sources))
-
More than half (55%) of total US podcast ad revenue comes specifically from host-read ads, per IAB's FY2021 podcast ad revenue report and 2022-2024 growth projections, accessed only via secondary industry aggregation.
contested IAB (Interactive Advertising Bureau), IAB, FY 2021 Podcast Ad Revenue and 2022-2024 Growth Projections
-
Podcast advertising generates roughly four times better brand recall than digital display ads and lifts purchase intent by an average of 10%, per Nielsen 2019 data as cited in peer-reviewed literature. The primary Nielsen report could not be independently retrieved and is known only through consistent secondary academic citation.
emerging Nielsen, via Journal of Consumer Marketing (peer-reviewed), Nielsen (2019), cited in Brinson, Lemon, Bender & Graham, 'Consumer response to podcast advertising: the interactive role of persuasion knowledge and parasocial relationships,' Journal of Consumer Marketing 40(7) (2023)
-
US podcast audiences grew 40% over the three years prior to mid-2022, with 51% of listeners reporting they began listening within the last two years.
established Nielsen, 'U.S. podcast listenership continues to grow, and audiences are resuming many pre-pandemic spending behaviors' (2022-08-05)
-
80% of Americans age 12+ (about 230 million people) have ever listened to a podcast; 58% (167 million) listened in the last month; 45% (130 million) listened in the last week.
established Edison Research, The Infinite Dial 2026 (annual US digital-media-consumption survey running since 1998; n=2,050, fielded January 2026) (2026-01)
-
Parasocial relationships that listeners form with podcast hosts are associated with greater trust, engagement, and brand recall, with these effects measurably larger for host-read ads than for other podcast ad formats.
emerging Journal of Marketing Communications (peer-reviewed), Brinson & Lemon, 'Investigating the effects of host trust, credibility, and authenticity in podcast advertising,' Journal of Marketing Communications (2022)
-
Host affinity positively influences listener perceptions of podcast ads, and host-read ads are skipped less often than pre-recorded spots, with the effect most pronounced for comedy and true-crime podcast genres.
emerging Journal of Radio & Audio Media; Journal of Economics and Behavioral Studies (peer-reviewed), Moe, 'Podvertising' and 'Podvertising II,' Journal of Radio & Audio Media / Journal of Economics and Behavioral Studies (2021-2022)
-
A survey of college-aged, primarily Gen Z podcast listeners found they generally preferred host-read ads over prerecorded/announcer ads, and that thematic congruence between the advertised brand and the podcast's subject matter improved perceptions of the ad.
emerging Advertising & Society Quarterly (peer-reviewed), Brooks, Bichard & Craig, 'Advercasting: Young Adult Preferences for Advertising Placement in Podcasts,' Advertising & Society Quarterly (2022)
-
In a controlled audio experiment (n=113 college students), overall perceived host source-credibility correlated positively with brand attitude and purchase intent; however, perceived host trustworthiness and attractiveness did not significantly predict brand recall, and perceived host expertise significantly predicted unaided recall in the opposite direction expected.
contested Texas State University (peer-reviewed thesis, primary experimental data), Fischer, V.K., 'Unaided and Aided Brand Recall in Podcast Advertising: An Experiment in the Role of Source Credibility's Impact on Brand Message Efficacy,' Master's Thesis, Texas State University (2019-05)
-
Contrary to the general belief that listeners prefer host-read ads and that host-read raises perceived credibility, this study's results diverged from that expectation on the specific measure of advertising receptivity, complicating the 'host-read always wins' narrative. Only the abstract and citation fragment were verifiable; full text was paywalled.
contested Journal of Radio Studies (peer-reviewed, abstract-level verification only), Wang, R., 'The Context Effect,' Journal of Radio Studies, vol. 31 (HeinOnline) (2024)
-
Ad-skipping is less prevalent among podcast listeners than on other digital ad vehicles, based on a survey of 1,000 weekly podcast listeners; exact skip-rate percentages sat behind the report's download gate and could not be verified.
established Sounds Profitable (podcast industry research firm), Webster, T., 'Ad Nauseam' (2024-06-12)
-
The typical (modal) US podcast episode carries seven ads, with a range of one to twelve ads per episode.
emerging University of Hamburg / Journal of Communication Technology (peer-reviewed secondary citation), Wolf, F.V., 'Identifying Advertisements in Podcasts,' University of Hamburg dissertation, cited in Fisher, N.S., JoCTEC 7(3) (2025)
Learning outcomes
What this study teaches
- Podcast advertising is a genuinely growing, well-documented category, up 26.4 percent to more than $2.4 billion in 2024, riding an audience where 80 percent of Americans have listened at some point and 45 percent listen weekly.
- The trust listeners place in a podcast host is real and shows up across multiple peer-reviewed studies as higher engagement and recall for host-read ads specifically, but it is a moderate, format-internal effect, not proof that podcast audio beats every other channel.
- Treat the '4x brand recall, +10% purchase intent versus display' figure as credible but secondary-cited: it comes from Nielsen 2019 research that could not be independently retrieved and is known only through academic citation.
- There is no dedicated evidence yet that local or small-business podcast sponsorship is a distinct high-trust channel; it is a reasonable inference from the broader trust research, not a measured finding, so treat it as a hypothesis to test at small scale rather than a proven play.
- Before shifting real budget into podcast sponsorship on the strength of trust research alone, get a read on where your own buyers' attention actually sits today, podcasts included, rather than assuming a general finding applies to your specific market.
Honest limits
What this does not yet settle
- No verified, direct data compares host-read podcast ads to television, out-of-home, direct mail, or other non-digital ad formats on recall or conversion. The only cross-format comparison with a real (if secondary-cited) source is podcast versus digital display, from the 2019 Nielsen figure. A claim that host-read audio 'out-converts every ad format measured' is broader than the evidence supports; the accurate scope is 'outperforms non-host-read podcast ads, and, per one widely secondary-cited figure, digital display.'
- No dated data was found specifically on small or local-business podcast sponsorship spend, SMB share of the podcast ad market, or a dedicated SMB podcast-advertising case study with recall or conversion numbers, despite targeted searches across academic databases, IAB reporting, and industry sources. The 'local podcast sponsorship as a high-trust SMB surface' framing is supported only by adjacent evidence (general host-trust research), not by SMB-specific podcast-ad data.
- The primary Nielsen 2019 report behind the widely cited 4x brand recall and +10% purchase intent figures could not be directly retrieved; its publication URL has since been deprecated. The figure is known only through consistent secondary academic citation, so its methodology and sample size could not be independently verified.
- Exact quantified ad-skip-rate percentages from Sounds Profitable's 2024 'Ad Nauseam' report sat behind a report-download gate; only the qualitative topline finding and survey methodology (n=1,000 weekly listeners) were retrievable.
- No large-scale, randomized experimental study, as opposed to correlational or self-report survey research, was found that isolates host-read as a causal driver of purchase conversion at scale. The strongest causal-design evidence located (Fischer, 2019) is a small (n=113), single-university, single-exposure college sample with mixed and in one case counter-intuitive results.
- No measurement yet exists connecting podcast host-read advertising to visibility inside AI answer engines (ChatGPT, Perplexity, Google AI Overviews, and similar). That connection should not be asserted until such a measurement exists.
This is a synthesis of dated, attributed evidence, not a census. The AI-answer layer in particular has no independent, Nielsen-grade measurement yet, so readings of it are directional and named as a frontier, never presented as settled.
Straight answers
Frequently asked questions
Does host-read podcast advertising outperform TV, radio, or direct mail?
No evidence answers that question either way. Every comparison in the retrievable research measures podcast ads against other podcast ads (host-read versus pre-recorded) or, in one case, against digital display. Nothing in the literature compares host-read podcast advertising to television, radio, direct mail, or out-of-home spend on recall, purchase intent, or conversion, so claims that podcast audio beats those formats go beyond what has actually been measured.
What is a host-read podcast ad and why does it work better than a pre-recorded spot?
A host-read ad is one the podcaster reads in their own voice, often with their own commentary, rather than a pre-recorded announcer spot. Researchers attribute its effectiveness to a parasocial relationship, the one-sided sense of trust and familiarity a listener builds with a host over repeated listening. Brinson and Lemon's peer-reviewed study in the Journal of Marketing Communications, plus Moe's Podvertising research and a study by Brooks, Bichard, and Craig, all found this trust translates into higher engagement, recall, and preference for host-read ads specifically.
How large is the US podcast advertising market?
US podcast advertising revenue crossed 2.4 billion dollars in 2024, up 26.4 percent from the year before, according to the IAB and PwC's Internet Advertising Revenue Report. That growth sits on top of a large and expanding audience: Edison Research's Infinite Dial 2026 survey found 80 percent of Americans age 12 and older have listened to a podcast, with 45 percent listening in the past week.
Is the claim that podcasts get 4 times the brand recall of digital display accurate?
The figure is real and widely repeated in peer-reviewed citation, most directly in a 2023 Journal of Consumer Marketing paper by Brinson, Lemon, Bender, and Graham, but it traces back to Nielsen 2019 research that could not be independently retrieved because the original report's publication URL has since been deprecated. The number survives only through consistent secondary academic citation rather than a source document that could be checked directly, so it is treated as credible but not independently reverified.
Should a small local business sponsor a podcast to build trust cheaply?
That argument is plausible but currently unproven. No dedicated dataset quantifies small or local business share of podcast ad spend, and no case study in the literature isolates a small or local advertiser's recall or conversion numbers from a host-read placement. The idea rests on the same general host-trust research covered in this study, which makes it a reasonable hypothesis to test at small scale rather than a measured finding a business can act on with confidence.
Provenance
References
- IAB/PwC, Internet Advertising Revenue Report: Full-Year 2024 Results (2025) https://www.iab.com/insights/internet-advertising-revenue-report-full-year-2024/
- IAB, Full-Year 2018 Podcast Ad Revenue Study (2019) https://www.iab.com/wp-content/uploads/2019/06/Full-Year-2018-IAB-Podcast-Ad-Rev-Study_6.03.19_vFinal.pdf
- IAB, FY 2021 Podcast Ad Revenue and 2022-2024 Growth Projections (2022) https://www.iab.com/wp-content/uploads/2022/05/IAB-FY-2021-Podcast-Ad-Revenue-and-2022-2024-Growth-Projections_FINAL.pdf
- Brinson, Lemon, Bender & Graham, 'Consumer response to podcast advertising: the interactive role of persuasion knowledge and parasocial relationships,' Journal of Consumer Marketing 40(7) (2023), citing Nielsen (2019) https://doi.org/10.1108/jcm-01-2023-5819
- Nielsen, 'U.S. podcast listenership continues to grow, and audiences are resuming many pre-pandemic spending behaviors' (2022) https://www.nielsen.com/insights/2022/u-s-podcast-listenership-continues-to-grow-and-audiences-are-resuming-many-pre-pandemic-spending-behaviors/
- Edison Research, The Infinite Dial 2026 (2026) https://www.edisonresearch.com/the-infinite-dial-2026/
- Brinson & Lemon, 'Investigating the effects of host trust, credibility, and authenticity in podcast advertising,' Journal of Marketing Communications (2022) https://doi.org/10.1080/13527266.2022.2054017
- Moe, 'Podvertising' and 'Podvertising II,' Journal of Radio & Audio Media / Journal of Economics and Behavioral Studies (2021-2022) https://doi.org/10.1080/19376529.2021.1960840
- Brooks, Bichard & Craig, 'Advercasting: Young Adult Preferences for Advertising Placement in Podcasts,' Advertising & Society Quarterly (2022) https://doi.org/10.1353/asr.2022.0032
- Fischer, V.K., 'Unaided and Aided Brand Recall in Podcast Advertising,' Master's Thesis, Texas State University (2019) https://digital.library.txst.edu/bitstreams/0221aeda-7ae5-40f4-967c-de21f18ffa9b/download
- Wang, R., 'The Context Effect,' Journal of Radio Studies, vol. 31 (2024) https://access.heinonline.org/hol-cgi-bin/get_pdf.cgi?handle=hein.journals/jradstud31§ion=26
- Webster, T., 'Ad Nauseam,' Sounds Profitable (2024) https://soundsprofitable.com/research/ad-nauseam/
- Wolf, F.V., 'Identifying Advertisements in Podcasts,' University of Hamburg dissertation, cited in Fisher, N.S., JoCTEC 7(3) (2025) https://www.joctec.net/article/158132.pdf
Every measured figure is dated to its capture and tagged with an evidence tier. Every cited work is real and locatable. Where an engine could not be captured this round, it is named as uncaptured, not estimated. Small-sample readings are labelled as directional.