Modèles
Monthly SEO Report Template
A monthly report should answer three questions: what changed, why, and what we are doing next. Most answer none of them and show a lot of charts.
Lancez un audit DomainLens frais et utilisez le rapport comme liste de priorités.
The structure
The "what we shipped" section is the one that makes the report honest. A month with no changes should say so, and a report that cannot list changes is reporting weather rather than work.
| Section | Answers | Length |
|---|---|---|
| Summary | What changed and whether it matters | Three sentences |
| What we shipped | What was actually done last month | A list |
| Results | What moved, and whether we can attribute it | One page |
| Next | What we are doing this month, and why | A short list |
| Appendix | The full numbers | As long as needed |
Metrics worth reporting
Year-on-year comparison matters more than month-on-month for anything seasonal, which is most things. A February that looks poor against January frequently looks fine against the previous February.
- Clicks and impressions from Search Console, compared to the same period last year rather than last month.
- Indexed page count against intended page count, which catches bloat and exclusion early.
- Core Web Vitals field status by URL group.
- Position bands — how many queries in the top three, four to ten, eleven to twenty — rather than average position.
- Conversions or leads from organic, if you can attribute them at all.
Metrics that waste a page
Average position is the worst offender because it looks meaningful. It moves when you gain a position 80 ranking for an irrelevant term, and it falls when you win visibility on a competitive one. See Search Console vs third-party tools for which numbers are facts and which are models.
| Metric | Why it misleads |
|---|---|
| Average position | Averages across queries that are not comparable |
| Domain authority | A vendor score Google does not use |
| Keyword count "ranking" | Includes position 90 for terms nobody searches |
| Bounce rate | A single-page visit can be a satisfied one |
| Lighthouse score | A simulation, not a ranking input |
| Total backlinks | Volume without quality says nothing |
Write for the person who will not read it
The realistic audience for a monthly report is someone who reads the first paragraph and skims the rest. That is not a failure of attention — it is a rational response to a recurring document, and the report should be built for it rather than against it.
It follows that the summary carries the whole thing. Three sentences saying what changed, whether it matters, and what happens next. Everything after that exists for the person who does want detail, and for you in six months when you need to remember what was done.
The corollary is that a chart is rarely the right way to say something important. If a number matters, write it in a sentence with the comparison that makes it meaningful. Charts are for showing shape over time, not for carrying the message.
Attribution, honestly
Most months you cannot prove what caused a change, and saying so is more credible than a confident story. Algorithm updates, seasonality, competitor activity and your own work all move the numbers at once.
What you can do is state what you shipped, when, and whether the affected pages moved differently from the rest of the site. That is evidence rather than proof, and it is the honest maximum.
- How long should a monthly SEO report be?
- One page plus an appendix. If the summary and next steps do not fit on a page, the report is describing activity rather than decisions.
- Should I report average position?
- No. Use position bands instead — how many queries sit in the top three, four to ten, and so on. Averages hide everything that matters.
- What if nothing changed this month?
- Say so, in one line, and move to what is planned. Padding an empty month is what teaches people to stop reading the report.