Most show analyses fail before they start. Not because the data is wrong or the instincts are bad, but because the person doing the analysis already knows what they want to find. They go in looking for confirmation, and they get it. That’s not analysis. That’s a more expensive version of the same thinking that created the problem.
Week 1 is built around a different premise. You’re not here to defend your show. You’re here to see it — clearly, without the usual filters of habit, history, or ego. That requires a specific kind of discipline: the ability to sit with what you find without immediately reaching for solutions.
That discipline is easier to maintain when you start with the right question. Not “what’s wrong with my show?” but something sharper: if we were launching this show from scratch today, how would we attack it? That reframe puts you on the outside looking in. It’s the only vantage point where honest analysis is actually possible.
What you’re building this week: A one-page diagnosis focused on three strengths, three problems, a prioritized list of what matters most, and one clear hypothesis about the single change with the greatest leverage. That document will serve as your guide for the next three weeks. Everything else is data-gathering in service of it.
This week is only about diagnosis. Not solutions. Not fixes. The entire job is to look at the data, look at the audience, and name what’s true — even when it’s inconvenient. Especially then.
Get organized before you start. Then don’t stop to fix anything. That urge will be strong, and it’s the enemy of clear thinking right now.
Before Starting Week 1:
- Watch the Week 1 Introduction
- Read the Week 1 Introduction document: Stop Guessing: Get Clarity. Find it here!
- Download the Ratings and Research Observations Template, Ratings Tracking Tool, and the One Page Diagnosis documents.
Keep an open mind seeking answers. Nothing is sacred.
But don’t tear things down yet. That comes later!
Apply Zero-based thinking in the process. Key question:
If we were launching a new show today, what would we do? How would we attack our own show?
Expected Time for Research: You’ll need to be organized, but once you have the source materials in place, this should take 90 minutes or less. Don’t overthink it, and don’t stop to fix anything. This is all about discovery.
______
Step 1: Ratings Analysis
Download the Ratings and Research Observations Template before you begin and keep it open as you work through this step. This is where your findings go, and you’ll be referencing it again in Step 2.
NOTE: If ratings data isn’t available for your show, station, or market — or if you simply don’t have access to it — skip to Step 2 and invest that time there instead. The research step will surface most of the same patterns.
Purpose:
Ratings data gives you a longitudinal picture of your show’s performance — not a snapshot, but a trend. The goal here isn’t to find the worst numbers and label them the problem. It’s to find patterns: where the show is gaining traction, where it’s stalled, and where it’s quietly losing ground without anyone making a fuss about it.
Most people approach this analysis backward. They start with the lowest number, treat it as the story, and stop there. What you’re looking for is movement — direction and pace across time. That’s the information that actually tells you something
What to Analyze:
Pull at least a full year of data, and more if you have it. Single-period ratings are easily distorted by seasonal shifts, one-time events, or methodological quirks. You need enough data to distinguish consistent patterns from noise.
- Trend direction over time. Is the line moving up, down, or sideways across AQH and TSL? Not just the numbers themselves, but the direction and pace. A slow, steady decline is different from a sharp drop followed by stabilization.
- Cume movement vs. share. These tell different stories. Cume tells you whether you’re reaching people at all. AQH and TSL tell you whether they stay once you have them. A show can have a healthy cume and an eroding TSL, and that’s a very specific problem with a very specific set of causes.
- Quarter-hour patterns. Map your strongest and weakest quarter hours against what the show is actually doing in those windows. Is there a content explanation for the patterns? Are there lifestyle factors (commute windows, school drop-off, workplace routines) driving the shape of the curve?
- Day-part and day-of-week trends. Look across the full week, not just the average. Some shows perform meaningfully differently on Fridays than on Mondays, and that difference is worth understanding.
- Your show’s cume vs. your station’s cume. Calculate this as a percentage. It’s a measure of your show’s relative strength — whether your audience is broader or narrower than the station’s overall reach, and by how much. This number tends to reveal opportunities that other metrics conceal.
Guidance:
Stay at 30,000 feet. Ratings analysis is where smart people get lost in the weeds, because there’s a lot of data and it’s easy to mistake complexity for insight. What you’re looking for is big-picture trends, not anomalies. If you find yourself drilling into a narrow demographic breakdown in a single quarter hour, you’ve gone too deep.
Demographic detail is rarely the story at this stage. Demo swings in narrow slices are almost always noise. The signal lives in broad, sustained trends across the full picture.
If everything looks flat, and this is more common than most people admit, that’s still a finding. Sustained flatness can mean your show has reached the ceiling of its current format. Or it can mean you’re executing consistently inside a stable but limited range. Either way, name it. Don’t paper over it with optimism.
Deliverable:
Use the Ratings and Research Observations Template to document your three most significant findings from this step, and only three. If your list is longer, edit it down. The discipline of choosing the three that matter most is part of what this Playbook is teaching, and this is a good place to start practicing it.
Each finding must be specific, measurable, and connected to a pattern rather than a moment. Test each one against this sentence: “We consistently see…” If you can’t complete it with something the data actually supports, keep looking. A conclusion that only holds for one rating period is just a coincidence.
______
Step 2: Research Your Audience
Ratings data tells you what happened. Audience research tells you why. More importantly, it tells you what your listeners actually experience versus what your show believes it’s delivering — and those two things are often startlingly different.
A team might believe chemistry is its greatest strength. Their audience might describe them as “the morning show where they talk about their weekends.” One of those is a brand asset. The other is a warning. The only way to know which is which is to look at the evidence without editing it first.
This step requires discipline. You’re not gathering support for a position. You’re pressure-testing whether your show’s self-image holds up when the audience gets a vote.
NOTE: If possible, launch a quick listener survey before completing this step — even a short 3-question survey can surface insights you won’t find anywhere else. It takes a day to set up and can change everything you think you know. That said, don’t let the absence of a survey stop your progress. Work with what you have.
Purpose:
To surface the gap between how your show sees itself and how your audience actually experiences it. The findings from this step get added to the Ratings and Research Observations form you downloaded in Step 1.
Where To Look:
Pull from as many of these sources as you have access to. You’re looking for patterns, not individual data points — one angry email is noise, but the same complaint appearing in emails, social comments, and survey responses is a finding.
- Existing audience research. Perceptual research, format studies — anything that has a sample size and methodology behind it. Weight these heavily. Know how the sample was recruited and screened before drawing conclusions.
- Listener surveys and comment cards. Useful for directional trends, less useful for specific quotes. A short quarterly survey of even 50-100 responses will reveal patterns that years of gut instinct won’t.
- Social media comments and direct messages. Read for patterns, not volume. Don’t take these literally — a vocal minority can distort your reading. But when the same theme appears repeatedly across different voices, pay attention.
- Emails and call-screener notes. These represent your highest-engagement listeners — the ones motivated enough to reach out. Weight them accordingly, but remember they’re not your average listener. They’re your most passionate ones, in both directions.
- Digital behavior. On-demand listening, podcast downloads, engagement on specific content — these are behavioral data. They tell you what listeners actually do, not just what they say they like.
- Sales and promotion feedback. Account executives and promotions staff hear things talent never does. Ask them directly: what do clients and listeners say about the show when you’re not pitching them?
Rules:
No cherry-picking. This step is only useful if it’s honest. The goal is to identify the three or four things your audience consistently experiences (both good and bad). Don’t use it to build a case for what you already believe.
Negative feedback counts double. If your audience doesn’t complain, that’s not necessarily a good sign. It may mean they’ve stopped caring enough to bother. Treat consistent criticism as the most valuable data in this step.
Separate facts from perceptions. “Our show is funny” is a belief. “Listeners describe us as the show that makes their commute go faster” is a finding. The deliverable from this step should be filled with findings, not beliefs.
What To Look For:
You’re trying to answer five questions about how your audience actually experiences the show:
- What do they say the show is for? How do they describe it to someone who’s never heard it?
- What do they consistently praise, and does that match what the show intentionally produces?
- What do they say is confusing, repetitive, or not worth their time?
- What listening occasions do they associate with the show? Morning drive, background, active listening?
- What’s the one thing they’d tell a friend about the show right now? Include both good and bad.
Deliverable:
Add your three most significant audience insights to the Ratings and Research Observations form. Focus on what the audience consistently says, not just occasionally. If a finding can’t be connected to at least two separate sources, leave it out for now.
Look especially for places where the audience’s experience diverges from the show’s self-image. That gap is where the most important work in Weeks 2 and 3 will happen.
______
Step 3: One-Page Show Diagnosis
This is where the work from Steps 1 and 2 comes together. Everything you’ve gathered gets distilled into a single document. Not a report. Not a summary. A diagnosis.
The discipline this step demands is different from the first two. Steps 1 and 2 were about gathering information without judgment. Step 3 is about making decisions. You’re editing everything you found down to the only things that really matter, and the document is only useful because you were willing to cut everything else.
Download the One-Page Diagnosis template from the Resources section below if you haven’t already. It’s a guide, not a form. Your first draft will almost certainly run longer than a single page, and that’s exactly right. Write everything out first, then edit until only what’s true and important remains.
NOTE: This is a good step to do with at least one other person. Maybe you can enlist the help of a co-host, a trusted colleague, your PD. The goal is to see the forest, not just the trees, and someone outside your daily perspective will catch the conclusions you’ve been quietly avoiding.
Purpose:
To force clarity and prioritize what needs attention. At the end of this step, you won’t have a list of observations and a few decisions: your three strongest assets, your three most critical problems, your three highest-leverage priorities, and the one hypothesis that will drive the next three weeks of work. Everything else gets cut.
The Format:
The document has four components. Each one requires a decision, not just a description.
- Three strengths. The things the show does well that listeners actually experience — not what the team believes it does well. Be specific enough to defend. “We’re relatable” isn’t a strength. “Our highest quarter-hours consistently come during storytelling segments, and our audience survey describes us as the show that makes the drive feel shorter” is a strength. If you can’t connect it to data from Steps 1 or 2, keep working.
- Three problems. Name them clearly and precisely. Vague problems can’t be fixed. “Our show lacks energy” is a feeling. “TSL drops consistently in the second half-hour, correlating with the feature block we’ve been protecting for three years,” is a problem. If your list is longer than three, prioritize and cut. Choosing is the skill.
- Three priorities. The three most urgent things to address right now to advance the show. These should connect directly to your problems, not new ideas or wish-list items.
- One hypothesis. The most important line on the page: “If we fix X, Y should happen.” One thing. The single change with the greatest leverage. If you can’t get to one, you haven’t finished thinking yet. Go back to your data. The answer is in there. Keep editing until you are ready and willing to commit.
Rules:
Every statement must be defensible. Don’t fill the categories with answers that feel right. Fill them with answers that the data from Steps 1 and 2 supports. If you can’t point to something specific that backs up what you’ve written, it doesn’t belong on the page yet.
“We’ve always done it this way” is not a strength. Neither is “our audience expects it.” Both are descriptions of inertia, and inertia doesn’t show up as a positive in ratings trends. Go back to what the data actually shows.
If any point can’t be explained simply and clearly, it’s not finished. Complexity in a diagnosis is almost always a sign that you haven’t made the decision yet. Keep tightening until every line is crisp enough to read aloud to your team without qualification or apology.
Once this document is complete, it will serve as your guide for the rest of the Playbook. Weeks 2, 3, and 4 will all refer back to it. The clearer and more honest it is now, the more useful it will be then.
______
Before Moving On To Week 2:
Additional Resources:
Each of the three steps you just completed has a more in-depth version in the Insiders Radio Network library. These aren’t optional reading—they’re the full expansion of what you did this week, written for people who want to go beyond a 90-minute audit.
If a particular step surfaced more questions than it answered, which is a good sign, these resources will sharpen your thinking before you move to Week 2. If everything felt clear, they’re worth reading anyway. The difference between a show that improves once and a show that keeps improving is usually found in how thoroughly someone did the diagnostic work. In the links below, you’ll find:
Stop Guessing: Get Clarity. Save this detailed guide to Week 1 to reference throughout your Playbook, and use it when analyzing your show in the future as well.
Key Ratings Metrics To Master. The full breakdown of what to measure, what to ignore, and how to read trends across a station’s competitive set.
Conduct A Yearly Free Research Project. A step-by-step guide to building a real audience research process without a research budget. Everything you need to run a credible listener survey from scratch.
Update Your Show Without Freaking Out Your Audience. Once your diagnosis is complete, this is how you start making strategic changes without disrupting listeners who already like what you do.