What a channel audit should contain

Somebody has offered to look at your channel and tell you what is wrong with it. Reasonable question in return: what is going to arrive, and how would you know whether it was any good?

That question has an answer, and it is the same whether the audit is done by a studio, a freelancer or by you on a Sunday afternoon with a spreadsheet. The shape does not change. What changes is who has the time.

What is an audit for?

To turn a vague dissatisfaction into an ordered list of decisions.

Before one, the feeling is that the channel is not working and could be better in about nine ways at once. That state produces either paralysis or a burst of activity aimed at whichever fault was mentioned most recently, and neither is progress.

After a good one, you should have a small number of things to change, in an order, with a reason attached to each. That is the entire deliverable. Everything else in the document is evidence for it.

What should be in one?

Topics scored separately from execution

Two different questions get muddled constantly. Are you making the right things, and are the things you make any good? A channel can be excellent at both, weak at both, or, most usefully, strong at one and weak at the other, and the fix is completely different in each case.

So the topics get scored on their own. What you cover, how it maps to what people are looking for, whether the subjects repeat in a way anybody could follow. Then the sub-topics underneath, because a channel that reads as scattered is often three good series filmed in a random order.

The craft, split into its parts

Picture, sound, pacing, packaging, the first fifteen seconds, whether the video delivers what its title promised. Split, because a single overall mark is useless. Sound at four and picture at eight tells you where the afternoon goes. An average of six tells you nothing at all.

The good scores, explained

This is the part that usually goes missing and it is the part that changes behaviour. When something scores well, the document has to say why, in specific terms, because that is the thing you must not accidentally stop doing.

Reading that your intros work is pleasant and useless. Reading that your intros work because you start on the thing being made rather than on your own name is a rule you can apply to every video from now on, including ones nobody has audited.

An order of work

End with a short priority list, explaining impact and effort. Keep useful secondary findings in a separate backlog so detail does not obscure the first decisions.

Strengths belong in the audit

Because the next plan needs to preserve what already works.

Faults tell you to stop doing things. Strengths tell you what to do more of, which is a far cheaper instruction to follow. The person who learns that their explanations land and their openings do not can rebuild an opening in a fortnight. The person handed twelve faults rebuilds nothing, because there is no starting point in a list of complaints.

Check whether the observations refer to actual moments in your videos. Both praise and criticism need evidence; neither proves the auditor watched simply by being present.

How do you run a rough one on your own channel?

An afternoon, ten videos, no tools beyond a spreadsheet.

Use a simple scale and write one observation beside each score. Compare similar videos and revisit uncertain judgements rather than treating the first reaction as automatically right.

Then two passes over the finished sheet. First, find the columns where your marks are consistently low. That is your craft problem and it is the same problem in every video, which is good news, because it means one habit fixes ten videos. Second, find the rows where your marks are high and look at what those videos have in common. Subject, format, length, who you were talking to. That pattern is worth more than the fault list. People tend to skip it.

Finish by writing three sentences: what to keep doing, what to change next, what to stop. If you cannot get it down to three, you have not finished the audit, you have just made notes.

How can you tell an audit from a sales document?

Read the recommendations and ask who does them.

An audit should explain the recommendations, alternatives and likely effort. Some actions may be yours, some may need help. A supplier's ability to do the work is not itself evidence that the assessment is biased.

Look for specifics: a recurring sound issue, a particular opening, a gap in the subjects covered. Generic advice is weak evidence of a close review, even when your name appears at the top.

In this studio the audit scores main topics and sub-topics out of ten, and the good scores get the same explanation as the bad ones, which is also why we do not quote for work without looking at the channel first. The looking is free either way.

Run the spreadsheet version yourself before you accept anybody's offer, including this one. Twenty minutes into your own ten videos you will know which questions are worth asking, and an audit you can interrogate is worth several you cannot.

Sources and evidence

Our method. Scoring topics separately from execution, splitting the craft into its parts, explaining the good scores as carefully as the faults, and ending with a short ordered list rather than a long one, is how this studio structures an audit. It is a working practice, not a rule any external source sets.