Most SEO reports are a data dump with a logo on top. The client opens it once, scrolls to the chart that looks good, and never reads the rest. A useful report fits on one page and answers three questions: what changed, what it means, and what happens next.
That is the format I use for the Fix Sprint report, and it is the format I would want if I were paying someone. This is the structure, the numbers that earn a place, and the ones that should be cut.
The one-page structure
- The headline. One sentence: the most important thing that happened this month. Not "traffic is up". Something specific, like "the three rewritten service pages doubled their clicks but the blog lost a third of its impressions".
- Three numbers. Impressions, clicks and one business number, with last month's figure beside each. Three numbers can be read in five seconds.
- What changed. The work that shipped, with dates. A list, not a paragraph.
- What the data says. Two or three observations with the evidence. "The local page moved from position 14 to 9 for the city query, and the map clicks increased." Not "local SEO is working".
- What we are doing next. The next two or three actions, with an owner and a date. No vague promises.
- What is blocked. Anything waiting on the client, the developer or a decision. This section protects the relationship.
The numbers that earn a place
Impressions are the best leading indicator of whether the content is being served for the right queries. Clicks are the best indicator of whether the titles and descriptions are working. Average position is useful only as a trend, never as a single month's score.
A business number matters because SEO is not the goal. Leads, sales, booked calls, downloads. If the site has no conversion tracking, fix that before the next report. A report that cannot connect to a business outcome is a report nobody will renew.
The numbers to cut
- Domain authority. A third-party score that does not explain what to do.
- Keyword rankings for phrases nobody searches. The long tail is useful for finding opportunities, not for monthly reporting.
- Total pages indexed. It changes when it wants to and nobody makes a decision from it.
- Backlink totals. Unless someone is actively building links, the number is noise.
- Every metric a tool can produce. The report is not a dashboard export.
Write the interpretation, not the number
"Impressions fell 12%" is not an insight. "Impressions fell 12% because Google stopped showing the blog posts for a query where they ranked on page three, and the newer posts have not taken those impressions yet" is an insight. It tells the client what happened and what to do.
If you cannot explain a movement, say so. "I do not know yet, here is what I am checking next" is a better sentence than a made-up cause. Clients forgive uncertainty. They do not forgive being managed with confident nonsense.
The report and the work
A good report is written while the work is happening, not reconstructed from a dashboard at the end of the month. Keep a running note of what shipped, what changed and what was blocked. The report is mostly editing that note.
Before the meeting, run the site through the website auditor so the basics are current. Check that the pages you claim are live actually return 200, and that the title changes you reported are on the page. I have seen reports list a fix that was never deployed. That is the fastest way to lose trust.
The report skeleton
Copy this layout and fill it in. It fits one page and it takes about forty minutes once the data is in front of you.
- One-line headline. The most important thing that happened this month, written as a sentence rather than a label.
- Three numbers. Impressions, clicks and one business metric, each with the previous month and a percentage change.
- Shipped. A list of the work that went live, with dates and URLs.
- Observed. Two or three findings with the evidence beside them.
- Next. Two or three actions with an owner and a date.
- Blocked. Anything waiting on someone else, with the specific request.
Write the "Observed" section before the numbers. If you cannot explain the movement, the report is not ready, and the next action should be a question rather than a claim. That is a better report than a confident graph with no cause behind it.
Send the page as text in the email as well as a PDF. Clients read email on phones, and a one-page summary in the body gets read in the minutes after it arrives. The PDF is for the file, not for the reader.
If a month had no meaningful change, say that. Not every month produces a story, and a report that invents one teaches the client that the numbers are decorative. A quiet month with honest notes is a better argument for renewal than a fabricated trend.
The report is the product as much as the work is. If you want the done-for-you version, the service page has the scope and prices. If you are building the report yourself, the tools hub has the checkers and the experiments post has the honest version of how tests actually go. Keep the report to one page. The client will read it, and that is the whole point.