Most authors arrive at this question having already tried a few things: perhaps a new cover, a price drop, a burst of social posts. If none of it clearly helped, the frustration is understandable, because without a method, it is genuinely hard to tell what the actual problem is, let alone what to try next. This guide sets out a deliberate, step-by-step process for narrowing it down, rather than another list of things to try at random.
This is more procedural than the general overview in [why isn’t my book selling](/resources/book-sales-growth), and it is designed to be worked through roughly in order, since each step depends a little on ruling out the one before it. It also feeds directly into a private [investigation plan worksheet](#book-sales-investigation-plan) at the end, and pairs with the more data-focused companion guide on [diagnosing poor book sales](/resources/book-sales-growth/how-do-i-diagnose-poor-book-sales).
Why the order matters
Checking positioning before checking whether anyone sees the book at all risks fixing something that was never the bottleneck. Working from visibility inward, towards conversion, positioning, pricing, trust signals, audience and consistency, means each step is checked once the ones before it are reasonably ruled out.
The eight-step investigation
- 1
Step 1: Check whether enough relevant people see it
Look at whatever visibility signals are available: retailer category rankings, any ad impressions if campaigns are running, website traffic, and social reach. If almost no relevant readers are seeing the book anywhere, later steps about conversion are premature; the first job is discoverability, covered in why is nobody finding my self-published book.
- 2
Step 2: Check retailer or product-page conversion
If people are seeing the book but a reasonable share are not buying, look at the page itself: does the cover read clearly at thumbnail size, does the description explain the book quickly, is the sample chapter engaging, and is the price visible and sensible for the format.
- 3
Step 3: Check positioning
Ask whether the book is being presented to the right readers in the first place. A well-made product page for the wrong audience will still convert poorly, however good the cover and description are individually.
- 4
Step 4: Check pricing and formats
Compare price against genuinely similar books in the same category and format. Consider whether the formats a reader might want (ebook, paperback, hardback, audio) are actually available, since a missing format quietly closes off a segment of readers.
- 5
Step 5: Check reviews and trust signals
A very low review count, or reviews that flag a specific, recurring issue, can affect a hesitant reader’s decision to buy. This is about honest signals of trust, not chasing a review count as a target in itself.
- 6
Step 6: Check owned audience
Ask whether there is any group of people, a mailing list, a social following, a community, who can be reached directly when there is something to say. Without one, every piece of marketing activity starts from zero each time.
- 7
Step 7: Check marketing consistency
Look at whether marketing activity has been sustained or sporadic, and whether the message has been consistent across channels. Inconsistent or one-off activity is hard to distinguish from "marketing that does not work", when the real issue may be that it barely happened.
- 8
Step 8: Change one meaningful variable at a time
Once a likely weak point is identified, change that one thing, give it a reasonable period to show any effect, and observe before changing anything else. This is the step most often skipped, and skipping it is why so many authors end up unsure what actually helped or hurt.
Why changing everything at once breaks diagnosis
It is tempting, once several possible weak points have been identified, to fix them all in the same week: a new cover, a lower price, a rewritten blurb, new keywords, and a fresh ad campaign, all launched together. If sales then improve, there is no way to know which change did it, or whether it was the combination, and no way to repeat the success on a future book with confidence.
If sales do not improve, the situation is worse: several genuinely good changes may have been made, but a remaining, unaddressed problem (perhaps positioning, perhaps visibility) could be masking their effect entirely, leading the author to wrongly conclude that none of the changes mattered.
Example
Imagine an author who redesigns the cover, drops the price by a third, and starts an ad campaign in the same fortnight. Sales rise slightly. Was it the cover, the price, the ads, or the fact that the ad campaign simply bought more visibility regardless of the page behind it? Without changing one thing at a time, there is no way to answer that, and no reliable lesson for the next book.
Book sales investigation plan
Private worksheet
Book sales investigation plan
Use this as a private working log while moving through the eight steps above. Note what was found at each stage and what, if anything, is being changed, one step at a time.
Your answers stay in this browser tab only. Nothing is scored, sent or saved to an account.
What if nothing obviously stands out
Sometimes every step in the investigation comes back "seems fine", and sales are still modest. That is a real, if unsatisfying, possible outcome: a competently presented, discoverable, fairly priced book can still sell modestly, for reasons covered in why do some self-published books not sell, including market size and competition, which are not failures of execution.
Common mistakes
- Skipping straight to conversion or pricing. If barely anyone sees the book, checking the page’s persuasiveness first is premature.
- Changing the cover, price and keywords in the same week. Simultaneous changes make it impossible to know what actually helped or hurt.
- Treating every step as equally likely. Working through the steps in order, rather than jumping to a favourite theory, avoids wasted effort.
- Giving a change no time to show a pattern. Judging a single change after a day or two rarely reveals anything beyond ordinary daily noise.
- Assuming a clean result means the book has no problems at all. A book can pass every diagnostic step and still sell modestly, for reasons unrelated to execution.
Frequently asked questions
Do I have to complete all eight steps in order?
It is designed to work best in sequence, since later steps depend on ruling out earlier ones, but a step that is clearly fine can be moved through quickly rather than laboured over.
How long should I wait after making one change?
There is no fixed period; it depends on the book’s usual sales volume. The aim is to wait long enough to see a pattern rather than react to normal day-to-day variation.
What if I have already changed several things at once?
It is not possible to retroactively separate their effects. The most useful move now is to stop making further changes and observe the current state for a period before adjusting anything else.
Is this the same as the future Book Growth Check?
No. This is a manual, self-guided investigation you can work through now using this article and its worksheet.
What if the investigation does not turn anything up?
That is a genuine possible outcome, and it points towards the broader explanations covered in [why do some self-published books not sell](/resources/book-sales-growth/why-do-some-self-published-books-not-sell), rather than a hidden execution fault.