SEO reporting for people who do not care about rankings
A board does not want to hear about position four. It wants to know what the channel produced, what it cost, and whether to keep funding it. Here is a report that answers that.
Key takeaways
- Lead with pipeline, change, cost per acquisition and share of answer; put rankings and sessions in an appendix.
- Sessions can fall while qualified leads rise, so a traffic headline invites the wrong decision.
- Use impressions as the visibility measure, since they keep counting when clicks do not.
- Name what is blocked; the unshipped engineering fix is usually the real constraint and the one a board can clear.
An executive SEO report should answer four questions on one page: what did search produce in pipeline or revenue, how did that change, what did it cost per unit, and what will change next quarter. Rankings, sessions and domain metrics belong in an appendix, because none of them is a business outcome and all of them can move in the wrong direction while the business improves.
The four numbers on the front page
- Pipeline or revenue attributed to organic and AI-referred traffic, with the attribution model stated plainly so nobody argues about it later.
- Change against the previous period and the same period last year. Seasonality in most markets is large enough that a month-on-month figure alone is misleading.
- Cost per qualified lead or per acquisition, compared with your paid channels. This is the number that decides budget, and it is where search usually wins.
- Share of answer for your priority questions: how often you appear in AI Overviews and assistant responses. This is the leading indicator that predicts the others.
Why sessions belong in the appendix
Sessions have become actively misleading. With AI Overviews on roughly half of tracked queries and click-through compressed on those queries, a well-run programme can lose informational sessions while growing qualified leads. If the front page of your report shows a traffic decline, you will be asked to explain a success as though it were a failure, every month, until the budget is cut.
Report impressions instead where you need a visibility number. Impressions keep counting when clicks do not, so they remain an honest measure of whether you are being seen.
Being honest about attribution
Search is usually an assist rather than a last click, and AI-referred traffic makes this worse because the discovery happened inside a conversation you cannot see. State your model, apply it consistently, and resist the temptation to change it in a quarter where the numbers are unflattering. A consistently imperfect model is far more useful than a sequence of favourable ones.
What to say about the next quarter
Three specific commitments, each with an owner and a date, and a note of what is blocked. The most useful line in most SEO reports is the one naming the fix that engineering has not shipped, because that is usually the constraint on results and it is the one thing a board can actually resolve.
How to measure share of answer without a tool budget
Pick twenty questions that matter commercially, phrased the way a customer would ask them. Once a month, run each through the assistants your market uses and record three things: whether your brand is named, whether your site is cited as a source, and which competitor appears instead when you do not. Twenty questions takes under an hour and produces a trend line within a quarter.
Keep the question list frozen. The temptation to add questions you have started winning is strong and it destroys the comparison. If the list needs to change, start a second list and report both until the old one has enough history to retire.
The slide nobody asks for and everybody needs
Include one slide showing what search costs against what the equivalent paid traffic would cost at current rates. For most established programmes the ratio is stark, and it reframes the conversation from an expense to be trimmed into an asset being maintained. Do the arithmetic conservatively and show your working, because the number is usually surprising enough to be challenged.
If your headline metric can fall while the business improves, it is the wrong headline metric.
Sources