65% of podcast listeners say they feel more connected to a brand after hearing it on a show. That statistic gets quoted in a lot of pitch decks. What rarely follows it is the harder truth: that connection only happens when the show was built for the listener in the first place. Most branded podcasts are built for the brand.
They are built around what the marketing team wants to say, what the executive sponsor wants to be known for, what the PR team thinks sounds good. The audience is assumed. Their needs are gestured at. And then the show launches into a market that has no particular reason to care.
The audio empathy map is a framework for fixing that before you record a single word.
Why "We Know Our Audience" Almost Never Holds in Audio
Every marketing team has personas. Most of them were built for digital advertising, email nurture, or event programming — mediums where you control the environment and compete for attention in a single, defined moment. Audio is a different context entirely.
A listener on a morning run has maybe 40% of their cognitive bandwidth available. Someone cooking dinner while the kids fight over something in the next room has less. A commuter on a packed train is distracted, possibly self-conscious about volume, and probably going to skip anything that requires too much front-loaded effort to understand. Demographics do not capture any of this. A buyer persona does not tell you what emotional state someone is in when they press play.
The listening environment is not a background detail — it is a design constraint. If your show is built for peak attention and most of your listeners encounter it during low-attention moments, your format has already failed before the host says hello. Segment length, episode structure, how quickly you pay off the premise — all of it changes depending on where and how your audience actually listens. The research that informs your show has to account for this, or it does not actually account for your audience.
This is why the audio empathy map is a different instrument than a marketing persona. It maps the experience of listening, not the demographics of a buyer.
What the Audio Empathy Map Actually Captures
The framework is an adaptation of the UX empathy map, recalibrated for the specific conditions of audio consumption. The dimensions worth mapping are five.
Listening context and environment. Where is your audience when they press play? At a desk with studio headphones, during focused work? On a commute? At the gym? Each environment implies a different cognitive state, a different tolerance for complexity, a different emotional posture walking into the episode.
Emotional state at the moment of play. This one is underestimated. A listener who queues up a podcast during a frustrating workday wants something different than one who listens on a Sunday morning walk. Tone, pacing, and host energy all land differently depending on what emotional frequency the listener is already tuned to.
Current knowledge level and vocabulary. Does your audience know your industry's terminology? Are they deep practitioners or curious newcomers? The answer determines whether your show builds confidence or creates alienation — usually within the first five minutes.
What they're hoping to feel or know by the end. This is the clearest editorial test. If you can articulate the specific thing a listener should walk away with — not as a mission statement, but as an actual human outcome — every episode structure decision becomes cleaner.
What would break trust or make them skip. This dimension is rarely mapped, and it's arguably the most valuable. What would make your target listener roll their eyes, reach for their phone, or feel talked down to? Where are the landmines? Knowing this before you produce a single episode is enormously more useful than learning it from a one-star review after 40 episodes.
This is not a one-time spreadsheet exercise. The map is a filter — something you hold editorial decisions up against to check whether you're still standing inside the listener's experience or whether you've wandered back into the brand's.
Where Real Listener Intelligence Actually Comes From
Surveys are the most common method and frequently the least useful. People answer surveys in a moment of relative calm and reflection. They do not answer them while running, distracted, slightly impatient, or in the specific conditions under which they actually listen to podcasts.
Social listening inside the communities where your target audience actually talks is more reliable. Not brand communities — the unfiltered spaces where professionals in your space process their frustrations, debates, and curiosities. Reddit threads, LinkedIn comments, Slack communities, Discord servers. The language people use when they aren't talking to a brand is the language your show needs to speak.
Direct conversations before a show launches are even better. Not focus groups. Actual conversations with people who represent your intended listener, asking about their current information diet, what they skip, what makes them feel like something was worth their time. When building Infernal Communication for Staffbase, the process involved speaking directly to internal communications professionals before scripting a single episode — understanding their real frustrations, the specific kinds of content that already felt patronizing to them, and what genuine value would feel like coming through their headphones. The result was a show that felt made for its audience rather than aimed at it.
The same principle shaped The Sound Bath for Lush. Rather than assuming editorial direction from inside the brand, the team turned to Lush's actual social audience — the people who already had a relationship with the brand — to understand what topics and angles would resonate with their values and curiosities. Audience input as editorial input, not as post-launch feedback.
For shows that already exist, listener reviews are an underused intelligence source — and a specific kind of intelligence. Pay attention to the language people use, not just the sentiment. We'll come back to why that distinction matters.
Turning the Map Into Editorial Decisions
The empathy map earns its keep not as a research document but as a production filter. The question is not "who is our audience" but "given what we know about our listener, what does this episode need to do differently?"
Listening environment should directly shape pacing and segment length. If your audience is primarily mobile — running, commuting, doing chores — long unbroken monologues are a structural mismatch. Short, punchy segments with clear sonic transitions give low-attention listeners a better chance of staying with you through the whole episode. If your audience skews toward desk listening, you have more room to develop complexity.
Emotional state should inform host energy and tone. A show targeting stressed practitioners mid-week needs a different register than one aimed at curious learners on a weekend. This is not about dumbing up or down — it's about meeting the listener in the room they're actually in.
Knowledge level should determine vocabulary and how much context you build in. The Nice Genes! podcast produced for Genome BC offers a clear illustration. The target audience — young science fans in the early stages of engaging with genomics — needed a show that started at the foundations before moving into complexity. Episodes were structured to build understanding progressively, not to impress the already-initiated. The empathy map made that decision legible as a strategy rather than a creative preference.
Desired takeaway shapes episode structure more than almost any other variable. If your listener should feel equipped to do something by the end of an episode, the structure needs to be functional — it needs to build toward that outcome, not just orbit a theme. If they should feel inspired, the emotional arc matters more than the information sequence. Being clear on the intended listener outcome before you write a script prevents the most common branded podcast failure: episodes that cover a lot of ground and deliver nothing in particular.
For a deeper read on why audio hits differently at the cognitive level — and what that means for how you structure episodes — this piece on how audio gets into your brain differently covers the neuroscience worth building your production decisions around.
The Language Test: Does Your Show Sound Like It Was Made for Them?
Once a show is live, listener language is the fastest diagnostic signal available. It tells you whether the empathy map is doing its job.
Scrape your reviews and social mentions and look at where the loyalty is landing. If you see "love her" or "he's such a good interviewer," the equity lives in the host's personality. The audience is following a person, not a concept. That's a particular kind of risk — one that becomes structural vulnerability if the host leaves or if the brand needs to build something more durable than a personality relationship.
If you see "love the show" or "the way this series builds on itself" or "these stories always make me think," loyalty has transferred to the concept. The audience is in a relationship with the show, not just the person delivering it. That's what you're building toward.
Beyond reviews, the deeper test is vocabulary. Does the show speak the language your audience uses, or the language your brand uses? These are not always the same language, and the gap between them is audible. Your branded podcast may literally be speaking a language your customers don't speak — and the empathy map is the instrument that catches this before it becomes a retention problem.
The audio empathy map does not guarantee a great show. No framework does. But it does guarantee that the editorial decisions you make are grounded in how your listener actually experiences audio — not how your brand wants to be heard. Those are different problems with different solutions, and confusing them is where most branded podcasts go wrong before they even start.
If you're building a show from scratch or auditing one that isn't performing, start with the map. Every other production decision follows from it.
