Podcast SEO: how to get your episodes found in search

Podcast SEO is the work of making an episode readable by search engines, because search engines still mostly read text, not audio. That means the episode title, the description, the transcript, the chapter markers, the video page on YouTube and the post on your own site. When those exist and match what people type, episodes get found months after they ship.

It works, but not the way blog SEO works. A podcast episode rarely ranks for a broad head term. It ranks for the specific question your guest answered in minute 34. That is why transcripts matter so much: they turn one hour of speech into a few thousand words of indexable text covering dozens of long tail phrases you would never have thought to target.

The honest ceiling: SEO will not fix a show nobody wants to hear. It compounds a show people already like. Treat search as distribution for episodes that are already good, not as a substitute for making them good.

If you are still setting the show up, the production side matters first. Clean audio and decent light on faces keep people watching, and watch time feeds YouTube ranking. My podcast lighting guide covers the cheap version of that.

Write the title as the thing a person would type, then add the guest name after it. "How to price freelance work with Ana Reyes" beats "Ana Reyes on pricing" and destroys "Episode 112". The searchable phrase goes first because titles get truncated in podcast apps, in YouTube search and in Google results.

A format I keep coming back to: [outcome or question] + [with guest name]. Around 55 to 60 characters for the front half. Keep the episode number out of the title entirely, or push it to the very end where nobody has to read it.

Common mistakes I see weekly. Clever titles that hide the topic. The show name repeated inside every episode title, which wastes the characters that matter. Emoji in the first three words. And titles written for the guest's ego rather than the listener's question, which happens a lot when the booking was hard won.

One more thing: your title should match the way you pitched the episode when you invited the guest. If you plan bookings around topics rather than names, the titles write themselves. That is half of what booking podcast guests is really about.

Tags do almost nothing now, and show notes plus a full transcript do almost everything. Tags were a 2014 signal. Text on a page you own is a 2026 signal.

Good show notes are not a summary paragraph. They are: a two sentence hook, three to six bullet takeaways written as full sentences, timestamps for every real topic shift, links to everything mentioned, and the guest's bio with a link. That structure gets pulled into AI answers and featured snippets because each bullet reads as a standalone claim.

The transcript goes on the same page, below the notes, in real HTML rather than a downloadable PDF. Speaker labels, paragraph breaks every few sentences, and no timestamps mid paragraph. A raw machine transcript with no punctuation is worse than none because it reads like spam. Budget twenty minutes to clean names, jargon and the first thirty seconds, which is what most people skim.

We do show notes and chapters as a $35 add-on per episode for exactly this reason. It is the cheapest line item in podcasting and the one hosts skip most often. If you would rather do it yourself, do it yourself, just do it every single week.

Chapters do two jobs: they hold people in the video, and they hand search engines a labelled map of your episode. YouTube surfaces chapter titles as key moments, which means one episode can appear in search for six different phrases instead of one.

Name chapters like search queries, not like inside jokes. "Why retainers beat hourly" not "The rant". Six to twelve chapters for a sixty minute episode is about right. Anything under sixty seconds is noise.

Clips are the other half. Vertical cuts do not rank in Google, but they feed the platforms that send new listeners back to the full episode, and they keep your channel alive between weekly drops. We cut vertical clips at $25 each, and most shows we cut run three to five per episode. Pick the moments where the guest says something slightly risky, not the moments where they explain the basics.

One caution: do not let clip volume replace the main episode page. Clips get attention, the episode page gets the compounding search traffic. You want both, and the clips should always caption the full episode title so people can find their way home.

YouTube ranks on watch behaviour first, so packaging carries more weight there than keywords do. The thumbnail and the first eight seconds decide whether the algorithm ever tests your episode on a wider audience.

Practical list. Thumbnail with a readable three or four word phrase and a face at a legible size. Description with your searchable summary in the first two lines, since the rest is collapsed. Chapters pasted as timestamps starting at 00:00, or YouTube will not build the chapter bar. A pinned comment with the one link you actually want clicked.

The cold open matters more than any tag. Lead with fifteen seconds of the sharpest thing said in the hour, then your intro, then the conversation. Long static intros before any content are the single most common reason a video podcast bleeds viewers in the first minute.

On packaging specifically, our sister studio ClickTheory does thumbnails at 8 for $250, and I would rather send you there than watch you fight Photoshop at midnight. If you want the reasoning behind what makes a podcast thumbnail work, the podcast thumbnail guide breaks it down.

Here is the version that survives contact with a real schedule. Before the record, pick the topic and draft the title. After the record, send the file to your editor. While it is being cut, write the show notes from your own recording notes rather than waiting for the transcript.

When the cut comes back, publish audio and video the same day, paste chapters into both, drop the transcript on your site page, and schedule the clips across the week. That whole publishing block is about ninety minutes if the edit and the notes are already done for you, and closer to five hours if you are doing everything yourself.

That gap is the honest argument for hiring out the edit. Our Starter plan is $199 per episode for a single camera at 1080p with one revision round, and Studio is $299 per episode for two to four cameras at 4K with unlimited revisions. Both are per episode up to sixty minutes, with shorter shows cheaper and ninety minute shows a bit more. A 4-episode pack saves 15% and monthly saves 25%. First cut is back in 48 hours, revisions the next working day, and rush 24-hour delivery is $75 when a launch date moves.

If you are weighing an outside editor against hiring in house, I wrote the maths out in editing service versus an in-house editor.

The expensive mistakes are boring ones, and almost all of them are one-time fixes that keep paying.

No episode page on your own domain. If your entire archive lives inside Spotify and Apple, you own no search real estate at all. One page per episode on a domain you control is the foundation.

Duplicate descriptions across every episode. Copying the same show blurb into all 90 episodes gives search engines 90 near identical pages and it flattens all of them.

Blocking or hiding the transcript behind a tab that loads with JavaScript after a click. If it is not in the HTML, treat it as invisible.

Inconsistent naming between platforms, so the same episode is called three different things on your site, YouTube and Apple. Pick one title and use it everywhere.

And the slow one: publishing irregularly. Search rewards archives that keep growing. Twenty episodes a year at a steady pace beats forty in a burst followed by silence. Our whole process is built around that rhythm, which you can read in how it works, and the plan maths lives on pricing.