The Episode That Won't Die

00:00 Welcome back to How To Get Discovered. I'm Maya. And I'm Tom. Last week was Tom's episode, the loyalty one, and he got further than I'd like to admit, though he did concede that a shared answer, passed from one friend to another, is a discoverability tool that serves loyalty. Once. It happened once. Three times, by my count, and I framed all of them. This is episode 5, the episode that won't die. It's about why a handful of old episodes keep finding new listeners for years, while most of them fade in a fortnight, and what that should mean for the rest of your back catalogue.

00:39 And my job is to be suspicious of any story where an old episode mysteriously comes back to life. Ah, but this one never went anywhere. That's rather the point. Hence the title. I want to tell you a story and then spend half an hour fighting about what it means. Go on. Three years ago, a guest cancelled on me the night before recording. Proper panic. So I rang my accountant and asked whether he'd come on and talk about one question. Should a freelancer stay a sole trader or set up a limited company? Not to be rude, but that sounds incredibly boring.

01:18 Oh, it is. It's 40 minutes of a lovely man explaining corporation tax and dividends. And for about 18 months now, it's been the single biggest source of new listeners to my show. Out of how many episodes? Over 140. So your best recruiter is the one you threw together in a panic about tax. It beats the ones I spent weeks on. And I didn't do anything to it. No clever tools, no relaunch, nothing. People kept typing that exact question into Google, and the title and the show notes happened to say pretty much what they typed. What was it called? Soul Trader or Limited Company? With a question mark.

02:01 Because I was tired and couldn't think of anything clever. That's the least creative title I've ever heard. And yet, here we are. Which brings us to the obvious question. Fluke? That's what I want to fight about. Was it an accident, or could you do it deliberately? And if you could, what does that mean for everything else you've ever recorded? Before you get there, I want to poke at the example, because I reckon it's a fluke. You happen to record an episode about something people Google. Most podcasters don't. Their episodes are a guest's life story, or a take on this week's news, or two people chatting about whatever was on their minds.

02:42 None of that's a search query. Let me make your case better than you did. Generous. Soul Trader or Limited Company is a sentence a real person types at 11 at night with a spreadsheet open. My episode was shaped like an answer to it. Most episodes aren't shaped like answers. They're shaped like conversations. So of course mine got found and theirs don't. Yes, thank you for stealing my argument. And part of it's right. An episode that maps cleanly onto a question people keep asking has a far better chance with strangers than one that doesn't. Conceded. And the bit that isn't? The idea that you know in advance which episodes are which.

03:24 I had no clue. I had episodes I was sure would be my big ones, guests I'd chased for months. The tax one was filler, and I nearly didn't put it out. Not exactly the heroic origin story. They never are. It did fine at the time. Then it carried on doing fine, every single week for years, because people never stopped starting businesses and wondering whether to incorporate. The question just kept arriving. Still sounds like luck. One of your fillers happened to age well. Except it wasn't only that one. When I finally went through the download numbers my host gives me, episode by episode, there was a pattern.

04:05 Almost everything that answered a specific question people keep asking was still picking up the odd listener. Almost everything tied to a moment had flatlined. The news reaction, the guest who was briefly everywhere, the what this week's announcement means. So the survivors were answering durable questions, and the dead ones were pinned to a date. Pretty much. Which sounds obvious when I say it out loud, but it's not how I was choosing what to record. I was chasing the news. I was chasing whoever was hot that month. The episodes that ended up mattering were the ones I made when I had nothing better to do. Ouch. Yeah, that one's stung. Here's what I don't get, though.

04:49 Someone finds your tax episode because they've got a tax question. They get the answer. Why would they ever come back? Most of them don't. They get what they came for and off. they go which is fine but some of them like the accountant or like me and they look at what else is in the feed a few of those are still listening now i know because they tell me usually starting with i found you through the tax one so the search gets them in the door and then the show has to do the actual work always the tax episode can't make anyone stay it just gives them the chance that's my loyalty argument wearing your coat it fits us both then can i try to pin down the difference then because people throw evergreen around as if it's a genre go for it a disposable episode

05:40 is about when why something happened this week what someone said on tuesday an evergreen one is about how how do i price this how do i fix that should i do x or y the test is whether a stranger could ask the question on any random day in five years and your answer would still hold up that's better than my version i've been making both kinds for 10 years mostly the first kind if i'm honest i did a whole episode once on a platform changing its download counting felt enormously important at the time and now couldn't tell you which platform it was or i think which year

06:20 perfect example although i'd add one thing to the test your tax answer will go out of date eventually thresholds change rules change evergreen isn't forever true my accountant would be horrified if he knew how many people were still hearing his numbers from three years ago i've put a note at the top of the description telling people to check the current figures look at you maintaining your back catalogue under protest here's what bothers me though my tax episode got lucky twice it answered a durable question and its title said so out loud how many episodes in a back

07:00 catalogue answer a durable question and say absolutely nothing about it on the outside most of mine probably my titles are a disgrace as you've told the world give me one there's an interview i did with a woman who'd left an agency to go in-house at a big retailer it was meant to be a career chat but about 20 minutes in she does this brilliant 10 minutes on what actually changes when you stop having clients and start having a boss what you lose what you gain what nobody warns you about and the episode's called her name just her name so the most useful 10 minutes in it are

07:40 invisible someone weighing up that exact move searches agency to in-house what changes and nothing about your episode tells google or anyone else that the answer's sitting at minute 20. fine so i rename it you could but that's one you happen to remember the hard part of doing this deliberately isn't the renaming it's the finding nobody's going to re-listen to a few hundred hours of their own back catalogue with a notepad i'd honestly rather not no which is where tools are actually useful get the whole catalogue transcribed then let ai chapter and topic detection break each episode into labelled sections suddenly that 10 minutes has a name something like what changes when you go in-house instead of being lost in

08:30 the middle of an interview fine but nobody outside your feed is ever going to see that no so every episode needs its own searchable page with those chapters on it so the answer can be found by somebody who's never heard of you ideally they land on the page and can jump straight to minute 20. that's the bit my tax episode had by accident a page whose words matched the question i assume podherd does this yes on podherd back catalogue transcription ai chapters and topics and searchable public episode pages are on every tier including beeherd other tools do

09:09 versions of it too what i'd stress is that it's how you'd go looking not a promise of what you'll find good because i'd want to know what to expect if i actually tried it most chapters will turn out to be nothing a stranger would ever search for a few might be your in-house bit a durable question hiding inside a conversation and even those would take months to get picked up if they ever do it's still the only way i know of to look for them without spending a year with headphones on and how would you know which chapters are worth bothering with same test you gave earlier could a stranger ask that question on a random tuesday in five years

09:49 if yes it's a candidate if it only makes sense to someone who heard last week's episode leave leave it alone. And once you've found one? Then it's the dull stuff. A title that says what's in it. Notes that start with the question rather than the guest's name. That's all my tax episode ever had going for it. So you're not promising anyone a catalogue full of tax episodes. No. My point is that most people have no idea which ones they've got. I didn't. And the only one I found, I found by accident. I want to push back on the whole direction of this, though, before anyone listening goes home and starts making 40-minute tax explainers.

10:30 That's fair. Most episodes shouldn't be evergreen. A show that only ever answers durable questions is a reference book, and nobody falls in love with a reference book. The news episode, the rant, the one where you and a guest just get on brilliantly for an hour. That's often why people stay. I agree with that, actually. The episode of mine I'm proudest of is about a festival that went wrong. Completely of its moment. Nobody's searching for it now, and nobody should be. But the people who heard it that week still bring it up. Then it did exactly what it was for. Right. They're for this week's listeners, and they serve them well this week. Dying after a fortnight isn't a failure.

11:14 A newspaper isn't a failure because nobody reads Tuesdays on Friday. Where I'd quibble is that news episodes often have an evergreen bit buried in them. A guest goes off on a tangent about how they handled something, and that tangent outlives the news by years. Which is your in-house argument again? Yes, and I'm not apologising for it. I'm not saying make different episodes. I'm saying know what's in the ones you've already made. That I can live with. What I can't live with is people planning their whole schedule around search terms and wondering why the show's gone grey. Agreed. If anything, the tax episode proves your point. I didn't plan it.

11:57 I made something useful because I had to. There's one more angle I keep thinking about, and it's annoying because it's commercial. Go on. A sponsor of mine dropped me a while back. Not because they were unhappy. Their finance people had decided a podcast ad does all its work in the first few weeks, and anything older than 90 days was basically worth nothing. And for the average episode, they're probably right. Sure. But your tax episode still has whatever ads were in it three years ago playing to strangers every week. If I'd been able to show that one of my old episodes was still finding new listeners a year later, that would have been a very different conversation.

12:38 That's a concession. Don't frame it. Too late. It's going up next to the others. Then hang a caveat beside it. Someone who lands from a search, plays 10 minutes and leaves, isn't the same as a regular who's heard every episode. Nobody really knows what that ad's worth to them yet. Fair, but it's not zero. And right now, most deals assume it is. Noted. For the next time I'm negotiating. I'll take it. Next week's one we might actually agree on, which worries me slightly. It's mine. Stop writing bad show notes. The easy wins.

13:19 Which ties in nicely, because the only reason my tax episode got found was that its title and notes said what people were searching for. By accident. Nearly everyone writes them like a receipt. This week I talked to Brian. They should be written like a pitch, and I've got a lot of feelings about it. He has. They came up in the car on the way here. Thanks for listening to How to Get Discovered. See you next week. Bye.