Should I delete old episodes from my feed?

Almost never for audience reasons, since old episodes cost you nothing to keep and some will keep earning downloads indefinitely. The cases for removing one are narrow: factually wrong content you cannot correct, a guest who has asked to be removed, or dated promotional claims you no longer stand behind. Storage and feed length are not good reasons.

How far back should my RSS feed go?

Feeds commonly carry a limited number of recent episodes while the rest stay reachable through your website and the directories. Check what your host's feed limit is set to, because a truncated feed means new subscribers see only recent episodes and your back catalog gets discovered only by people who go looking.

Does a large back catalog help a new listener?

It helps if there is an obvious entry point and hurts if there is not. Someone landing on episode 240 with no guidance often leaves. A pinned starter episode, a short listening guide, or a clearly marked season structure turns volume into an asset.

What is a back catalog in podcasting?

Your back catalog in podcasting is every episode you have already published that listeners can still download. It sits behind your newest release and, on most shows, quietly produces a meaningful share of monthly downloads without any promotion at all.

Most podcasters track the first week of a new episode and never look at what the other 200 are doing. That is where the interesting number is hiding.

Why the back catalog number matters

Split your monthly downloads into two buckets: episodes released this month, and everything older. The ratio between them tells you something a total cannot.

A show where almost all downloads come from new releases resets to zero every week. Growth depends entirely on promotion, and it stops the moment you do.

A show with a substantial back catalog share has built something that compounds. New listeners arrive, work backwards, and old episodes keep delivering months after release.

Pattern	What it means	What to do
Nearly all downloads from new episodes	Attention is release-driven only	Build searchable evergreen episodes
Back catalog share rising over time	New listeners are working backwards	Make the entry point obvious
Back catalog share falling while totals hold	Old episodes are aging out of relevance	Refresh or replace dated content
One old episode dominating	A single episode is doing discovery work	Find why, then build more like it

That last pattern is worth chasing down every time. An old episode pulling steady downloads is usually ranking for a search query or sitting on a recommendation surface, and knowing which one tells you what to make next.

What makes a back catalog keep earning

Topic choice does most of the work. An episode about a news event is finished within a fortnight. An episode answering a question people will still type into a search box in two years keeps working, which is the case for treating evergreen content as a deliberate part of the schedule rather than a happy accident.

Discoverability does the rest. Episode pages that answer a real query pull listeners from search long after release, and that is largely an on-page job covered in our podcast SEO guide. An episode nobody can find does not become evergreen no matter how good it is.

Old episodes also have to stay reachable before any of that applies. Apple documents what a feed and its episode items have to contain to be accepted and kept in the catalog in its podcast requirements, and a host that truncates your feed quietly shrinks what a new subscriber can browse.

Internal routing matters more than people expect. If your newest episode mentions the relevant older one, and your show notes link to it, you are sending your most engaged listeners backwards into the catalogue yourself.

Monetizing it

This is where the format of your ads decides what your archive is worth.

Baked-in ads are welded to the file. An episode from two years ago is still running a campaign that ended, for an offer that expired, possibly for a company that no longer exists. Those downloads earn nothing and occasionally embarrass you.

Dynamically inserted ads are stitched at request time, so every back catalog download can carry a current campaign. A show with a large archive and dynamic insertion can quote an advertiser monthly impressions across the whole catalogue, not just the impressions its next four episodes will generate.

The practical move is to mark ad slots in older episodes even if you are not filling them yet. The slots cost nothing to define and they turn a static archive into sellable inventory the day you need it.

How to measure it

Pull downloads by episode for a full month, then group by release date. Three numbers are enough.

The share of downloads from episodes older than 30 days is your headline back catalog figure. Track it monthly and watch the direction rather than the absolute value, because what counts as healthy varies enormously by format.

Downloads per episode by age shows you the decay curve. Most episodes drop steeply and then flatten. Where they flatten, and how high, is your evergreen floor.

Your top ten all-time episodes, refreshed quarterly, tells you which old work is still pulling. If the same episode keeps appearing, link to it from your newest show notes and record a follow-up on the same question.

Be careful reading a sudden back catalog spike. Re-uploading files, a feed change, or an app refreshing its cache can all produce archive downloads that no human requested. The IAB Podcast Measurement Technical Guidelines v2.2 are the rules most hosts and analytics vendors filter against, so reading them once tells you what your dashboard is already discarding and what it is not. Cache and re-upload artefacts are also worth ruling out when your download numbers move unexpectedly.

Related terms

Worth having alongside this: what a unique download counts, since back catalog analysis is only as good as the download definition under it.

Podder reports downloads per episode by date, so the split between new releases and back catalog takes a minute to pull. Start with the Analytics plan.