Get a quote
Tips & guides

How To Grow a Podcast Audience That Stays

Every guide to this question answers the same half of it. Post clips, go on other shows, optimise the listing, ask for shares. All of it is about getting more people to press play, and none of it asks whether the people already pressing play are still there four minutes later.

Sasi K
Sasi K
6 Nov 2026 · 12 min read
How To Grow a Podcast Audience That Stays

That omission is strange, because the platform tells you. Apple's own creator analytics separate the number of people who started an episode from the number who got meaningfully through it, and the distance between those two figures is a number worth watching that almost nobody reports. Close it and every listener you add afterwards is worth more. Ignore it and you are pouring into a bucket with a hole in it.

Your analytics already separate people who started from people who stayed

Most hosts look at one number and treat it as the audience. Apple publishes definitions for what each of its figures actually counts, and they are not interchangeable.

The definitions below are quoted from Apple's creator documentation (Apple Podcasts for Creators, listener analytics).

Metric What Apple says it counts What it tells you
Followers The total number of people following your show Intent to hear more. The closest thing to a subscription, and it does not mean anybody listened
Listeners, at episode level The number of people that played or watched more than zero seconds of the episode Curiosity. One second counts. This is the number most often mistaken for an audience
Plays Total number of plays for an episode where the duration is greater than zero seconds Activity, including repeats and accidental taps
Engaged Listeners Total number of listeners who heard or watched at least 20 minutes or 40% of an episode The real audience. People who stayed long enough for the episode to have done anything
Average Consumption A percent average indicating how much of the episode users played per device Where people leave. Apple notes that values above 100% mean somebody listened more than once

Read those side by side and the shape of the problem appears. A listener is somebody who played more than zero seconds. An engaged listener had to reach twenty minutes or forty per cent. Those are very different bars, and the gap between the two counts is the number worth watching month to month.

Growth without retention is arithmetic that does not work

Consider what adding listeners actually does when the gap is wide.

A show where most people who press play leave before the engagement threshold is not failing to be discovered. It is failing to be worth staying for, and every promotional push sends more people into the same experience. The promotion is not wasted exactly, but it is paying full price for a result it will not keep.

The order matters, and it is the opposite of how most shows work. Fix what happens after the play button, then spend on getting more people to press it. Doing it the other way round means paying twice: once to acquire somebody, and again to re-acquire them after they left.

There is a second reason to work this way. Retention problems are production problems, which means they are fixable with decisions you control. Reach problems depend on other people's attention, other people's platforms and other people's willingness to share. One of those is in your hands.

Five production decisions that lose people in the first two minutes

If average consumption drops early, the cause is usually in the opening rather than in the subject. The five below are all decided before anyone presses record, which is what makes them fixable rather than a matter of talent.

What happens Why it costs listeners The fix
The preamble Housekeeping, apologies for the gap since last episode, a long welcome. The listener came for the subject and is being asked to wait Start on the subject. Put housekeeping at the end, where only the people who stayed will hear it, which is the correct audience for it anyway
The guest introduction that runs long Reading a biography for ninety seconds before the guest speaks. Nobody has decided yet whether they care Fifteen seconds: name, role, the specific thing they know. Earn the biography by making them interesting first
Audio that is tiring rather than bad Room echo, uneven levels between host and guest, a guest quieter than the host. It is rarely unlistenable, it is just effortful A treated room and separate channels per voice, so levels can be balanced after the fact rather than argued about during
No reason to keep listening The episode states what it is about and then meanders. There is no question the listener is waiting to have answered Pose the question in the first minute and answer it late. The gap between those two points is the reason to stay
Inconsistent length Twelve minutes one week, seventy the next. The listener cannot decide whether now is a good time to start Pick a band and hold it, so pressing play is a low risk decision

The audio point is the one most often treated as a luxury and it is the one listeners feel without being able to name. What a room actually does to a recording is set out in why choosing the right podcast recording studio matters, and the specification side is in the video podcast equipment guide.

Episode two is a second retention problem, and a different one

Retention inside an episode is one problem. Retention across episodes is a different one, and it is the one that decides whether a show has an audience or a series of one off visitors.

  • The format has to be recognisable. Somebody who liked episode one is deciding whether episode two is the same kind of thing. If the structure, length and look all changed, they are being asked to make the decision again from scratch.
  • The publishing rhythm has to be predictable enough to be worth following. A show that appears when there is time does not build an expectation, and following it has no payoff.
  • The next episode has to exist. Somebody who finishes an episode and finds nothing after it has no reason to follow. A back catalogue, even a short one, converts a visitor into a listener in a way a single good episode cannot.

That last point is an argument for recording more than you publish, and for launching with several episodes rather than one. It is also the one that costs least to act on, because it is a scheduling decision rather than a production one.

The recording rhythm that makes a buffer possible is set out in producing a series of expert interview videos, and the guest pipeline that keeps it fed is in how to get guests for a business podcast.

Fix things in the order the numbers point to

Open the analytics before deciding what to work on. The pattern tells you which problem you have, and they need different work.

What the numbers show What it means What to work on
Plays are low and consumption is high Few people find it, and the ones who do stay to the end A reach problem. The show is working. Go and promote it
Plays are healthy and consumption drops early People arrive and leave quickly An opening problem. Work through the five production decisions above
Consumption holds but followers stay flat People finish an episode and do not come back A series problem. Format consistency, publishing rhythm, and something to listen to next
Everything is low Too early to diagnose, or too few episodes to judge Keep publishing to a fixed rhythm. Apple notes it needs at least five unique listeners in the period before showing data at all
Consumption above 100 per cent Somebody listened more than once, which Apple states explicitly Worth knowing which episode, because whatever it does is worth repeating

Resist working on all of it at once. One change per month, with the metric watched afterwards, tells you what actually moved. Four changes at once tells you nothing.

Be careful reading a small sample as a trend. A single episode with an unusual guest, an unusual length or an unusual release date will move a monthly figure on its own, and chasing that movement produces changes that undo good decisions. Wait for several episodes pointing the same way before acting. A single episode is an anecdote with a percentage attached to it, which is a persuasive kind of noise.

Then, and only then, go and get more listeners

Once people who arrive are staying, acquisition is worth paying for, because what you acquire now sticks. That is a separate body of work with its own channels, its own rules and its own measurement.

It is covered in full in how to promote a business podcast, which deals with the channels a business already owns, the Australian email rules that apply to promoting an episode, what platform tactics actually do, and using the sales process as a distribution channel. This article is deliberately the half that comes before it.

One caution about the goal itself. For a business podcast, audience size is a means rather than the objective, and a small audience of the right people is worth more than a large one of the wrong people. Retention work is the exception that serves both: a show people finish is better for a buyer and better for a stranger, so it is worth doing whichever audience you are after.

Konnect Studios fixes the retention problems that are physical

Three of the five opening problems above are production decisions you make in a room. The room either helps or it does not.

Konnect Studios is a video and podcast studio in the Parramatta CBD at Suite 02, Level 1, 111 Phillip Street, a six minute walk from Parramatta station, with parking at the building. Shows can book a Sydney studio session built for listenable audio from $199 an hour.

The retention problem What the published session provides
Room echo that makes listening effortful Recording rooms acoustically treated and isolated to minus 40 dB
A guest quieter than the host, unfixable afterwards Up to four Shure SM7dB microphones, each recording to its own channel, so voices can be balanced separately in the edit
Inconsistent look and sound across episodes Two sets included in every booking, both fully lit, with two Sony FX30 cameras in 4K framed and colour matched before you start
The host operating equipment instead of hosting An operator sets up before you arrive and looks after audio, cameras and lighting for the whole session. No technical experience needed
No buffer, so the next episode does not exist Sessions in 1, 2, 4 and 8 hour blocks, open 24 hours a day, 7 days a week, so several episodes can be recorded in one visit
Episodes that sit unedited and never publish Craft at $799 per 2 hour session returns a full edit of the episode, 5 reels and 1 teaser, with trimming, colour grading and subtitles

Rates, inclusions, equipment, sets and booking blocks read from the Konnect Studios podcast studio hire page on 6 November 2026.

What happens in the session itself, including the consent step that applies the moment somebody outside the business is on camera, is in how to record a professional video interview in Sydney, and what to send a guest before they arrive so the opening is not spent settling them in is in how to prepare a podcast guest for a professional recording. For work outside a standard session the Sydney video production service and the content creation service cover brand films and short form runs.

Three questions before spending anything on growth

  1. What is the gap between your listeners and your engaged listeners? If you have not looked, look before deciding anything. It is already in the dashboard and it costs nothing to look at.
  2. Where does average consumption drop? If it falls in the first two minutes, the problem is the opening rather than the subject, and no amount of promotion fixes an opening.
  3. If somebody finished an episode today, what would they do next? If the honest answer is nothing, the retention problem is the back catalogue rather than the episode.

For shows not yet launched, the decisions that come first are in how to start a podcast in Australia, the market conditions are in whether it is too late to start a podcast, and whether the show should be filmed at all is in audio podcasts compared with video podcasts for business. Where the show belongs to a company rather than a person, the approval layer is in what a corporate podcast setup needs in Sydney.

Questions people ask about growing a podcast audience

How do you grow a podcast audience?

In two stages, and most shows attempt them in the wrong order. First make the show worth staying for, because adding listeners to a show people leave early costs money and keeps nothing. Then work on reach. The first stage is production work you control, the second depends on other people's attention.

What is the difference between a listener and an engaged listener?

Apple defines an episode listener as somebody who played or watched more than zero seconds, and an engaged listener as somebody who heard or watched at least 20 minutes or 40% of an episode. One second of curiosity counts towards the first number and not the second. The distance between the two is the most useful figure in the dashboard, and it is the one most hosts never look at.

What is a good average consumption percentage?

This article does not publish a benchmark, because no credible one was found that would survive scrutiny, and an invented figure would be worse than none. Use your own trend instead: the number that matters is whether consumption on your show is rising or falling across episodes, which is a comparison you can make honestly.

Why do listeners drop off at the start of an episode?

Usually the opening rather than the subject. Housekeeping before content, a guest introduction that runs a minute and a half before the guest speaks, room echo and uneven levels that make listening effortful, or no stated question the listener is waiting to have answered. All four are decided before recording starts.

Does audio quality affect how long people listen?

Audio is easy to deprioritise because it is not usually the thing anybody names when they stop listening. The failure mode worth guarding against is not unlistenable audio, it is tiring audio: room echo, a guest quieter than the host, levels that move. A treated room and a separate channel per voice remove most of it, and both are easier to hire than to build.

How many episodes should exist before promoting a show?

Enough that somebody who finishes one has something to do next. A single good episode converts nobody into a follower, because following it offers no payoff. Launching with several, and keeping a buffer recorded, is a scheduling decision rather than a production one, which makes it cheaper to act on than most of the alternatives.

Should a business podcast chase a bigger audience at all?

Only as a means. For a business show the objective is reaching a specific set of people who can buy, so a small relevant audience beats a large irrelevant one. Retention work is worth doing regardless, because a show people finish serves both goals, which is not true of most reach tactics.

How often should you check podcast analytics?

Monthly, and change one thing at a time. Four changes in a month tells you nothing about which one moved the number. Apple also notes that a show needs at least five unique listeners playing content in the selected period before data appears at all, so very new shows should keep publishing rather than keep refreshing.

Related articles
Contact us