For two decades, audio sellers have heard the same objection in some form: “We can’t measure it the way we measure digital.” It has done more damage to radio’s share of ad budgets than any single competitor. Reach and frequency told advertisers how many people heard an ad. They never told advertisers what happened next.
The cost is visible in every allocation meeting: Ad-supported audio commands roughly a fifth of the time Americans spend with media, yet captures under 5% of ad budgets. When buyers are asked what would move more money into audio, the most common answer is not more reach; it is better proof.
That old objection is losing its force. Not because radio suddenly became digital, but because the tools now exist to close the loop between exposure and outcome, and AI is the reason those tools work at the speed and scale advertisers expect.
In plain terms, advertisers’ objections weren’t that radio doesn’t work, but that they couldn’t prove it worked in a way that satisfied a CFO or a media mix model. That’s a measurement problem, not a performance problem, and it’s an important distinction because it points to a fixable issue rather than a structural one.
Television faced a version of the same challenge and solved it by integrating with cross-channel attribution platforms. Audio is now doing the same thing, and AI is what makes the integration practical rather than theoretical.
So, what changed?
Three developments are doing the heavy lifting. First, first-party listener data has matured. Streaming platforms and connected devices generate behavioral signals that didn’t exist when radio measurement meant diary panels and rough demographic buckets. Advertisers can now see listening patterns, frequency and device-level signals that map far more closely to how digital and CTV campaigns are measured.
Second, connected-car analytics have become one of the most valuable data sources in audio. Cars are no longer passive listening environments. They’re connected endpoints that pass anonymized signals back into measurement systems, giving audio a data layer it never had in the broadcast-only era.
Third, AI-driven attribution modeling has caught up to the data. Media mix modeling and incremental lift analysis used to require months of data science work and large budgets. AI has compressed that timeline. Attribution that once took a quarter to produce can now be modeled in days, enabling audio sellers to deliver results-led conversations rather than promises. Put together, these shifts mean audio can finally answer the question advertisers have been asking for 20 years: What did this actually do for my business?
The shift is most evident in two areas: Competitive intelligence and ad creation. Real-time competitive intelligence tools now let sellers and advertisers see category spend, creative trends and share of voice as they happen, not months later in a retrospective report. That changes the sales conversation from “trust us” to “here’s what’s happening in your category right now, and here’s where you’re exposed.” It’s a sharper, faster, more defensible pitch than anything reach numbers alone could support.
Automated ad creation is doing something similar on the production side. AI-assisted tools can generate and test multiple creative variations quickly, enabling campaigns to be optimized in-flight rather than locked in before launch. A campaign that adapts its message based on early performance data produces cleaner, more convincing results than one that runs unchanged for eight weeks.
Neither capability is radio specific. That’s the real story: Audio isn’t inventing new measurement science. It’s catching up to a standard other channels set, and AI compressed the catch-up into a few years instead of a decade.
Measurement’s blind spot
There’s a deeper issue beneath the attribution gap that isn’t discussed openly enough: Most of the industry’s measurement effort is still built around in-car listening, because that’s where the legacy ratings infrastructure lives. Car listening matters, and the data investment there is real. Still, it has become the easy default, the place research and case studies keep returning to because the measurement pipes already exist.
Meanwhile, listening has spread far beyond the dashboard to smart speakers, phones, smart TVs and connected devices throughout the home and the workday. Traditional receivers, chiefly in the car dashboard, still account for a large share of listening. However, growth is everywhere else, and a disproportionate share of industry research still treats the car as the center of the listening universe. The audience is dispersed but the measurement discipline didn’t follow.
This isn’t a small gap. It’s a structural blind spot with a direct business consequence. Streaming platforms didn’t take audio budget by out-measuring radio in the car. They took it by measuring the listener everywhere, across every device, with the kind of identity-level data advertisers now expect as a baseline. Audio lost share because the industry hasn’t applied the same measurement discipline to in-home and mobile listening as it did for the car decades ago.
Closing the attribution gap only works if it’s applied to where people actually listen now, not just where the infrastructure happens to be strongest. AI-driven attribution and first-party data give the industry the tools to measure the full listening footprint. The harder part is choosing to point those tools at the whole picture rather than defaulting to the dashboard.
What this means for advertisers
Advertisers don’t need perfect attribution. They need credible evidence, and that’s a lower bar than the industry has often assumed. A campaign that shows incremental lift, even directionally, against a stated business outcome will out-compete one that only reports impressions and reach, regardless of channel.
There’s a window here that performance marketers are paid to find. Audio now offers the attention and trust of broadcast, the addressability of digital and pricing that hasn’t caught up to its provable value. Channels rarely stay mispriced once proof arrives. Advertisers who apply digital-grade scrutiny to their audio partners today, asking what identity data underpins the targeting and who validates the attribution methodology, will buy attributable outcomes at pre-attribution prices.
This is where the conversation needs to shift. Audio sellers who keep leading with reach are competing on the metric advertisers trust the least. Sellers who lead with attribution and business outcomes are competing on the metric advertisers actually use to make renewal decisions. The channel hasn’t changed; the currency advertisers accept has.
The measurement problem is improved, not finished. Standardization remains the biggest gap, as attribution methodologies still vary across platforms and vendors, making it hard for advertisers to compare audio performance with other channels using a consistent framework. Privacy constraints add another layer, limiting how granular cross-device attribution can get even as AI does more of the modeling work. These aren’t reasons to slow down but to be precise about what audio can prove today versus what it will prove in eighteen months. Advertisers respect precision; they lose trust in overclaiming.
The most important change isn’t technological. It’s psychological. For years, audio sold confidence about reach and hoped advertisers would extend that confidence to performance. That hope was always the weak link in the pitch.
AI-enabled attribution removes the need for that leap of faith. Sellers can now show evidence rather than asking for trust, which is a fundamentally stronger position in any budget conversation. The channel that proves its contribution, even imperfectly, will out-earn the channel that only promises it.
Audio’s next growth wave won’t come from better inventory or more reach. It will come from better proof, applied to where listening actually happens. The tools to deliver that proof finally exist. The opportunity now is for the industry to use them with the same discipline advertisers expect from every other channel they fund.
The author is V.P. of digital sales and strategy at StreamGuys.
This story originally appeared in the RedTech special publication Radio Futures: The Intelligent Signal. You can read or download that edition for free here.
You can access all past RedTech publications, also for free, here.
These stories might interest you
RedTech Magazine July/August 2026 drives growth through connections
Steering strategy with real-time audience data
Guest Commentary: Authenticity vs. AI — Embracing radio’s real competitive advantage