Three months in, the numbers look thin and somebody senior has noticed. Work through the causes in this order before changing anything.
By Will Nash
9 November 2026
Month three is when the question arrives. Somebody looks at the download figures, notices they are smaller than expected, and asks the obvious thing in a meeting where it is awkward to answer.
Before changing anything, work out whether the number is bad, and if it is, which of four things caused it. The usual response is to change the format, which is almost never the problem.
First, decide whether the number is actually bad
A firm's show is not a consumer podcast and the figures should not be read like one. A hundred listeners who are your buyers is a good outcome. Three thousand who will never instruct you is a vanity number with a production cost attached.
Look at who rather than how many. If the people you built this for are listening, the show is working even if the total is small, and that is the version of the number to put in front of whoever asked.
If you cannot tell who is listening, that is the finding. A public feed will not tell you, which is worth knowing before you start rather than at month three.
Then diagnose, in this order
Does anyone know it exists? This is the cause about eight times in ten and the least interesting to fix. A show that was announced once and then mentioned nowhere is an unpromoted show, not a failing one. Count how many times it has actually been put in front of people since launch. It is usually one.
Can they find it? Search the show name and see what comes up. If people are hearing about it verbally and cannot locate it afterwards, you have a naming problem rather than an audience problem.
Do the first two minutes tell them what they are getting? Listen to your own opening as a stranger would. If it begins with music, then a welcome, then some housekeeping, the useful part starts at minute three and most people will not reach it.
Was anyone asking about the subject? The hardest one to accept. If nobody outside the firm has ever asked the question the episode answers, the audience was never there to be found.
Fix in order of cost
Free and immediate: have the people in your firm send episodes to individuals for whom they are relevant. Ten personal sends will do more than any amount of scheduled posting, and they also tell you whether the episode is actually useful, because people reply.
Cheap: rewrite the episode titles and descriptions so they name the question being answered. A great deal of firm audio is titled like an internal document rather than like something a person would choose.
Moderate: put episodes on the service pages they relate to, and put the transcript with them, so the material can be found by people searching for the subject rather than for your show.
Only after all that: reconsider the format or the cadence.
What not to do
Do not buy listeners, and be sceptical of anything that promises them. You will move a number and learn nothing.
Do not run paid advertising to promote a professional-services show. It is expensive, and reaching more strangers is rarely the problem.
Do not change the format every episode. A show that keeps reinventing itself never accumulates the small group of regular listeners that makes it worth continuing.
Give it long enough to be worth judging
An audience for this kind of show is built by accumulation rather than launch. Six to nine episodes is a fair point to judge it, provided somebody has genuinely been promoting it throughout. Three episodes and no promotion is not evidence of anything.
Say that at the start, to the person who will eventually ask the question. It is a far easier conversation in month one than in month three.
When to stop
Worth saying plainly. If the right people know about it, can find it, and are not listening, the subject is wrong or the firm has nothing distinctive to say about it. Both are real findings and neither is fixed by continuing.
Stopping cleanly after a stated run is respectable. A show that fades out without anyone deciding to end it is the worse outcome, and it makes the next project harder to get approved.
In short
Check who is listening before you judge how many. Then work through the four causes in order, starting with the dull likelihood that almost nobody has been told the show exists. Fix the cheap things first, give it six to nine episodes with real promotion behind it, and if the right people still are not listening, say so rather than letting it drift.
We are an AI-native podcast agency, and we are sometimes brought into a show somebody else started. The diagnosis is usually one of the four above. If you have a show that is not landing and you want a straight read on why, get in touch.
Related: Is Your Company Podcast Working?
