By the time an author is asking this question, they have usually already tried something — a new keyword, a tweaked description, perhaps a price change — without a clear sense of whether it helped. That is a common and understandable place to be, but it usually means the effort has been scattered rather than sequenced. This guide sets out the fuller sequence: establishing what is actually happening before changing anything, then working through the product page methodically, one element at a time, before deciding whether more drastic steps like a relaunch are justified.
This is the cornerstone guide for post-publication optimisation on Amazon, and it draws together the more detailed guides elsewhere in this section — on covers, descriptions, keywords, categories, pricing, reviews, metadata and Author Central — into a single, ordered process. Treat this as the map; use the linked guides for the depth on each individual step.
Why order matters
Changing a cover, a description and a price in the same week might feel efficient, but it removes any ability to know which change, if any, actually mattered. It can also create a confusing, inconsistent impression if changes are not coordinated with intent. Working through a sequence, starting with diagnosis and moving one deliberate step at a time, takes longer but leaves you with an actual understanding of the book’s situation by the end, rather than a pile of edits and no clearer picture.
The optimisation sequence
The fourteen steps below are not all needed for every book, and not all of them will turn up something to change. They are, however, a sensible order to work through, because each step gives context that makes the next one easier to judge accurately.
- 1
1. Establish a baseline
Before changing anything, note current sales, page views (if visible), reviews and ranking as they stand today. Without this, no later comparison means much.
- 2
2. Diagnose discoverability
Work out whether the core issue looks like low traffic (few people finding the book) or low conversion (people finding it but not buying), since these point to very different next steps.
- 3
3. Review the product page as a whole
Look at cover, title, description, price and reviews together, as a reader would see them in sequence, rather than each in isolation.
- 4
4. Review the cover
Check genre signalling, thumbnail readability and finish against comparable, successful books in the same category.
- 5
5. Review the title and subtitle
Check whether they communicate genre, promise and, for non-fiction, the reader’s problem, clearly and without confusion.
- 6
6. Review the description
Check whether it reads as compelling jacket copy for the right reader, not a synopsis or an over-explained pitch.
- 7
7. Review keywords
Check whether backend keywords reflect how real readers would search, rather than aspirational or vague terms.
- 8
8. Review categories
Check whether current categories are accurate and specific, and whether a more precise option exists.
- 9
9. Review pricing and formats
Check whether price and available formats (ebook, paperback, hardback, audio) suit the genre and comparable titles.
- 10
10. Review reviews and trust signals
Check review count, recency and content honestly, and consider whether legitimate ARC or reader outreach is worth pursuing.
- 11
11. Review the author profile
Check that Author Central is complete, current and links the full catalogue, particularly if there is a backlist or series.
- 12
12. Review external platforms
Check consistency across other retailers or formats (such as IngramSpark editions) so the reader experience does not vary confusingly by platform.
- 13
13. Decide whether to relaunch
Only where several elements have genuinely improved together is a coordinated relaunch — a fresh push around a meaningfully better listing — usually worth considering.
- 14
14. Test changes deliberately
Where a specific change is uncertain, test it alone, record the date, and review the result after a fair period before moving to the next step.
What not to do
Do
- Work through the sequence in order, using the baseline as the reference point throughout.
- Change one element at a time and allow a fair period before judging it.
- Treat a relaunch as a deliberate, coordinated decision, not a routine step.
Avoid
- Change the cover, description, price and keywords in the same week.
- Skip the baseline and diagnosis steps because they feel like they are "not doing anything".
- Assume every book needs a full relaunch; most issues are solved by a smaller, targeted change.
A worked example
Example
Imagine a self-published historical fiction novel, six months after launch, with steady but low sales and a handful of reviews. A baseline shows page views are reasonable but the conversion from view to sale looks low. Working through the sequence, the author notices the description reads more like a plot summary than a pitch, and the categories are broad rather than specific to the subgenre. They rewrite the description alone, wait several weeks, then separately review and tighten the categories, recording each change and its date, rather than doing both at once.
When a relaunch is, and is not, justified
A relaunch — a coordinated push, potentially including a new cover, updated description and fresh review outreach, timed together — makes most sense when several elements have genuinely improved and there is a coherent, better product to bring back attention to. Relaunching around only a superficial change, such as a new price alone, tends to generate little genuine renewed interest, because the underlying reasons for weak performance have not actually shifted.
| Situation | Likely to be |
|---|---|
| Cover, description and categories all substantially improved together | Justified — there is a genuinely better product to reintroduce |
| Only the price has changed | Premature — unlikely to generate much fresh interest on its own |
| New edition or format has been added (e.g. audiobook) | Often justified — gives a genuine reason for renewed visibility |
| No specific improvement has actually been made yet | Premature — relaunch activity without a better product rarely lasts |
Post-publication optimisation plan
Private worksheet
Post-publication optimisation plan
A private, in-browser planning worksheet to work through the fourteen-step sequence for one book at a time. Use the notes space in each group to record what you found and what, if anything, you plan to change. Nothing here is scored, saved, or sent anywhere.
Your answers stay in this browser tab only. Nothing is scored, sent or saved to an account.
How this connects to the rest of this section
Each step above has a fuller guide elsewhere in the book sales and growth section: covers, titles, descriptions, keywords, categories, pricing, reviews, metadata timing and Author Central are each covered in more depth than this cornerstone guide can hold on its own. Use this sequence to decide where to look next, and the specific guide for the detail once you get there. For ongoing, longer-term work across a catalogue rather than a single optimisation pass, see Author Growth; where an author would rather this work were handled for them, Managed Author support may be the more relevant next step.
Common mistakes
- Changing everything in the same week. Coordinated activity feels productive but removes any ability to tell which change actually mattered.
- Skipping the baseline. Without recording current sales, views and reviews first, later comparisons have nothing solid to measure against.
- Relaunching around a superficial change. A relaunch built around a price change alone, without any real improvement to the page, rarely sustains renewed interest.
- Treating this as a one-off task. The sequence is worth revisiting when circumstances genuinely change, not something to do once and never return to.
- Assuming every book needs a full relaunch. Most books improve through a smaller, targeted change; a relaunch is for situations where several elements have genuinely moved together.
Frequently asked questions
Where should I start if I do not know what is wrong?
Start with a baseline and a basic diagnosis of whether the issue looks like low traffic or low conversion; that distinction shapes which of the later steps matters most.
Do I need to work through all fourteen steps for every book?
No. Not every step will turn up something to change. The sequence is a sensible order to check things, not a mandatory checklist for every title.
How do I know if a relaunch is justified?
Generally once several elements, such as cover, description and categories, have genuinely improved together, giving readers a coherent, better product to come back to.
How long should I wait between changes?
There is no official figure, but at least several weeks per change is a reasonable minimum, avoiding other simultaneous edits in the same window.
Is this the same as marketing my book?
It overlaps but is narrower: this guide focuses on the book’s own Amazon presentation and metadata. See our [broader book marketing guide](/resources/book-marketing/how-do-i-market-my-book) for audience-building activity beyond the product page itself.