Consumable and Edison Research at SSRS told the IAB Podcast Upfront in New York on September 16, 2026 that unaided awareness of a home security brand rose from 39% to 68% after 2,158 US adults each experienced one version of the same audio-led advertisement. The 29-point gain is pooled across three formats - sound alone, sound with a still image and sound with video - so the published materials do not show how much of it the audio delivered by itself.

In Short

A research firm played the same ad for a home security company to about 2,000 American adults, with some only hearing it, some seeing a picture alongside it and some watching a video with it, and far more of them could name the brand afterwards. People who listen to or watch podcasts every month liked the ad more and warmed to the brand more than everyone else. The catch is that the results for all three versions were added together, so the study cannot yet tell you whether the sound or the picture did the work.

One advertisement, three versions

The research was presented under the title "Seen and Heard: What Audio Ads Deliver When Sound Has the Full Screen" by Jared Lapin, chief strategy officer at Consumable, and Megan Lazovick, vice president at Edison Research at SSRS. Ahead of the event, the Podcast Upfront agenda listed a session from Edison Research at SSRS and audio-targeting company Consumable on what happens to advertising effectiveness when audio-led advertising gains a visual dimension. It was the fifth edition of the trade body's podcast marketplace, held at Convene 360 Madison Avenue, and Consumable also sponsored the day's closing networking reception. Consumable shared early figures with press on September 14 under an embargo that lifted after the presentation, then distributed a two-page topline summary and the seven-slide deck on September 25.

The design is simple. According to the topline summary, 2,158 US adults were surveyed in August 2026. Each experienced one version of the same advertisement. "Only the visual layer changed," the summary states: one group heard audio alone, a second heard the audio accompanied by an image, and a third heard it with video. The deck adds that results were weighted to reflect the US adult population. The advertiser was, in the deck's words, "a real, familiar brand in a category everyone thinks about: home security." Neither document names it.

Edison, which announced its acquisition by SSRS in October 2025, is best known in advertising for survey programs such as the Infinite Dial. The 2026 edition of that study, built on 2,050 interviews conducted in January, put monthly podcast consumption at a record 58% of Americans aged 12 and older. The Consumable study is of similar size, but it covers adults only and a single advertisement.

Awareness up 29 points, favorability up 4

The headline number concerns unaided brand awareness, the share of people who can name a brand without being shown a list. According to Consumable and Edison, it stood at 39% before exposure and 68% after, a rise of 29 percentage points, or roughly 74% in relative terms. Brand favorability moved far less, from 66% to 70%.

Four measures of the ad experience accompany those figures. Across the full sample, 67% said they liked the ads, 78% found them easy to understand, 69% agreed they were informative and 62% agreed the ads kept their attention. Read the other way, 38% of respondents did not agree that an advertisement they had been asked to watch or hear held their attention, and 32% still could not name the brand unprompted after it had just been played to them.

The favorability baseline raises a question of its own. At 66% before any exposure, it sits well above the 39% who could name the brand unaided, which implies the brand was named when favorability was asked. The materials do not describe the wording of either question, or whether favorability was measured across the whole sample or only among people who recognized the brand.

The format question the data leaves open

The deck opens with a direct question: "Does sound need a screen?" Its headline results do not answer it. The awareness and favorability gains, along with the four ad-experience figures, are labeled "Full sample, across every ad format tested." The topline summary says the same thing in different words, attributing the response to audio advertising "across all three formats tested."

That matters because of how the findings were framed before the event. In its pre-Upfront summary to press, Consumable asked what audio ads can deliver on their own and what happens when visuals are added, and presented the 29-point gain as a result of audio-led advertising. All three versions were audio-led, so the description is not inaccurate. Yet a number that blends sound-only, sound-plus-image and sound-plus-video responses cannot isolate what sound achieved. If the video version produced most of the gain, the pooled figure would still read 29 points. If the audio-only version matched the other two, it would read 29 points as well. The published materials contain no split by format for awareness, favorability, liking, comprehension or attention.

The only format-specific results concern the visual treatments, and they come solely from the two groups that saw them. Among participants exposed to audio with an image or with video, 64% agreed the audio and visuals worked well together and 66% agreed the visuals made the message easier to understand. Those are opinions about the visual layer, collected without any equivalent question to the audio-only group.

Other datasets have tilted toward pictures. Magellan AI's first quarterly measurement benchmark found a 2.49% response rate for podcasts distributed with video against 1.39% for RSS-only audio shows. Sounds Profitable's Podcast Atlas, as compiled in the Video Advertising Bureau's September report, found 15% of respondents made an immediate purchase after a video podcast ad, against 10% after an audio one. Those measure different things - observed response in one case, recalled behavior in the other - and neither tested the same creative with and without a screen. The Consumable and Edison design could, in principle, do exactly that.

Where the three documents diverge

Several discrepancies separate the pre-event summary, the topline and the deck. The pre-event summary stated that monthly podcast listeners outperformed the total sample on every ad response metric measured. The topline and deck show podcast-group figures for three measures only - liking, attention and favorability lift - and none for comprehension, informativeness or unaided awareness.

The same pre-event summary carried two figures that appear in neither of the documents presented at the Upfront. According to Consumable, 83% of the total sample said they see or hear ads across mobile apps, and 68% of monthly podcast listeners described reward-based mobile ads as acceptable, against 60% of the total sample.

Terminology drifts too. The press summary refers to monthly podcast listeners, while the topline and deck use monthly podcast consumers. The deck also reports that 67% "liked any of the ads," although each participant saw only one. Edison's Infinite Dial changed its question between 2024 and 2025 to count people who listened to or watched podcasts, as the VAB report noted when reproducing its figures. The Consumable materials do not define which behavior qualified a respondent for the podcast group.

Podcast consumers as the standout group

The topline summary is titled "The most responsive audience in audio: podcast consumers," and the matching slide in the deck carries the heading "Monthly podcast consumers love audio & ads." According to Consumable and Edison, 77% of monthly podcast consumers liked the ads, against 67% of the full sample. Some 71% agreed the ads kept their attention, against 62%. Brand favorability rose 7 points among them, against 4 points overall. The topline calls that "nearly twice the lift in brand favorability following a single exposure"; the ratio is 1.75.

Three features of that comparison shape how far it travels. The materials do not state how many of the 2,158 respondents fell into the podcast group, and they report no significance testing for any of the differences. Nor do they say whether the group differed from the rest of the sample on age, gender or income, characteristics that could account for part of its warmer response.

The third point is arithmetic. The comparison sets a subgroup against a total that contains it. Because monthly podcast consumers scored above the full-sample figures, respondents outside the group must have scored below them. The gap between podcast consumers and everyone else is therefore wider than the published comparison shows, by an amount the materials do not allow anyone to calculate.

From survey result to product pitch

The final pages of both documents move from research to Consumable's own product. PODiQ, which the topline describes as "Consumable's identity and intelligence layer for podcast audiences," is presented as a way to turn "podcast listening signals into addressable audiences that can be targeted, extended and measured across the mobile app ecosystem." Consumable described PODiQ in October 2024 as a people-based identity product for podcasts built on probabilistic identity graphs - an identity graph being a database that records which identifiers belong to the same person or household.

The study's contribution to that pitch is one figure: 93% of monthly podcast consumers said they already see or hear ads across mobile apps. The deck pairs it with a graphic linking the podcast consumer to gaming, weather, sports and news apps, and a firm conclusion: "PODiQ makes this audience addressable: identified and engaged with visual audio across the mobile app ecosystem, beyond the podcast episode. This study validates that premise directly."

That claim goes further than the data. The survey measured how people responded to an advertisement shown to them in a research setting, and what they said about their own exposure to app advertising. It did not test whether podcast consumers can be identified accurately outside podcast apps, how large the resulting addressable reach is, or whether an ad served to them inside a weather or gaming app performs as it did in the survey. A self-reported 93% - set against 83% for the total sample, according to Consumable's pre-event figures - shows that the audience uses ad-supported apps. It does not show that the audience can be found there.

Consumable is not alone in trying to carry podcast audiences into programmatic buying, and the routes differ sharply. Triton Digital made its Demos+ audience data selectable for podcast targeting in the United States, Australia and the Netherlands this month, attaching survey-based audience composition to episodes rather than to individuals and avoiding device identifiers. PODiQ runs the other way, attaching podcast behavior to people and following them out of the episode. Consumable Inc was among 19 companies that completed a first-time integration with mobile measurement company Kochava in the second quarter of 2026. The in-app audio segment it sells into is crowded: in July, Azerion added Audiomob's Moments format, which times audio ads to events inside mobile games, to its network.

What the study does not measure

The topline carries its own caveat in small print: "Findings are directional for the category tested." The qualification reaches beyond the category. The materials describe no unexposed control group, no margin of error and no interval between exposure and the follow-up questions. Brand measurement on digital platforms commonly rests on a holdout study, in which part of an audience is deliberately kept from seeing a campaign so that its responses stand in for what would have happened without the advertising. The Consumable and Edison figures are instead presented as before-and-after readings among people who were deliberately played the ad inside a survey, not people who met it in a feed, an app or an episode.

Under those conditions, a rise from 39% to 68% describes what one forced exposure did to recall among people who had just encountered the ad. It does not describe awareness in the market after a campaign. A single brand adds a further limit. A familiar home security name starting from 39% unaided awareness may behave differently from an unknown challenger or a category leader, and the materials say nothing about the creative's length or content, or where in the questionnaire the ad appeared.

Why it matters for audio buyers

Podcast budgets are growing faster than the evidence buyers use to defend them. United States podcast advertising revenue reached $2.9 billion in 2025, up 17.6%, according to IAB and PwC. Yet a July survey of 861 marketing and communications professionals found 44.9% of organizations using podcasts in their marketing, with 45.8% of non-hosting organizations citing doubts about effectiveness and only 16.8% of podcast marketers tracking formal return on investment.

Much of audio's effectiveness case has been built on performance attribution: promo codes, pixels and household matching. A counter-argument, aired on ExchangeWire's podcast in June, holds that audio has been pushed too hard toward performance credentials at the expense of its brand-awareness story. The Consumable and Edison data sits squarely on the brand side of that debate. Its most useful contribution, though, may be the one it has not yet published.

The screen question is where the stakes lie. The VAB report put the share of weekly podcast consumers who reach shows through video platforms at 88%, drawing on Cumulus Podcast Network and Signal Hill Insights data, and cited eMarketer forecasts giving podcasts 48% of US digital audio advertising by 2030. Buyers deciding whether to pay for pictures alongside the sound have very few controlled comparisons of the same creative with and without them. Consumable and Edison built exactly that comparison - and then reported it in aggregate.

What comes next

According to the topline, three further pieces of work follow: listener interviews on consumer response and creative implications, a full report containing "complete findings from this quantitative and qualitative research wave," and research comparing audio and visual audio with "other ad formats marketers buy." The deck states that the qualitative interviews are "starting now." It closes on three lines: "Audio works. Visuals add to it. Podcast listeners are the proof." Neither document gives a publication date for the full report, or says whether it will break the results out by format.

Timeline

Summary

Who: Consumable, an audio advertising and podcast identity company, and Edison Research at SSRS, presented by Jared Lapin, chief strategy officer at Consumable, and Megan Lazovick, vice president at Edison Research at SSRS. The findings concern advertisers, agencies and publishers buying and selling podcast and in-app audio inventory.

What: A survey of 2,158 US adults, each exposed to one version of the same home security advertisement as audio only, audio with an image or audio with video. Across all formats combined, unaided brand awareness rose from 39% to 68% and favorability from 66% to 70%. Monthly podcast consumers liked the ads more (77% against 67%), agreed more often that the ads held their attention (71% against 62%) and showed a 7-point favorability lift against 4 points overall. No results were published by format, and several figures in Consumable's pre-event summary do not appear in the presented materials.

When: Fieldwork took place in August 2026. The findings were presented on September 16, 2026, and Consumable distributed the topline summary and deck on September 25, 2026.

Where: The survey was weighted to the US adult population, and the research was presented at the IAB Podcast Upfront at Convene 360 Madison Avenue in New York.

Why: The study addresses whether audio advertising needs a visual layer at a time when most weekly podcast consumers reach shows through video platforms, and it underpins Consumable's pitch that podcast consumers can be targeted across mobile apps through its PODiQ product. Because results are pooled across formats, drawn from one brand and presented without a control group, the research leaves its own central question unanswered until format-level data is published.