The typical SEO report looks like this: a forty-row rankings table, a sessions chart pointing up, and a caption reading "positive dynamics". The owner looks at it for five seconds, finds no answer to the only question they have — how much money is in there — and closes the file. After a year of reporting like that, trust in the channel disappears even when the channel is working. Below: the five numbers that actually show results, why rankings and traffic are not among them, and how to build a report that needs no verbal explanation.

Why rankings and traffic prove nothing

Position stopped being a number. Results are personalised by location, history, device and language, so "third place" for your contractor and "ninth" for you are both real. Add AI Overviews, map packs, image blocks and "people also ask", which physically push organic results down: first place under three blocks earns fewer clicks than third place with none.

Traffic breaks differently. Sessions are easy to inflate with pages targeting informational queries that will never buy anything. We have seen projects double organic traffic with a flat lead count — the entire gain landed on articles answering questions asked by people who are not customers. On paper it is growth; in revenue it is zero.

The five metrics that show money

1. 1

   1. Organic leads, split by page type

   Not the total, but the breakdown: how many came from commercial pages, from the blog, from the homepage. That split immediately tells you which content to produce next, and it usually contradicts the assumption the project started with.

2. 2

   2. Share of non-branded traffic

   People searching your company name would have found you without SEO. Growth has to come from queries that do not contain your name. If the non-branded share is flat while total traffic rises, you are paying for the results of advertising or word of mouth.

3. 3

   3. Pages earning at least one visit

   This shows real coverage. A hundred-page site where eight pages produce traffic does not have a rankings problem — it has 92 pages search engines do not need. The number grows slowly and is almost impossible to inflate.

4. 4

   4. Cost per organic lead

   Channel spend for the period divided by organic leads. Its main value is an honest comparison against paid. In most niches organic becomes cheaper than paid traffic somewhere between month six and month twelve, and that crossover is the moment worth recording.

5. 5

   5. Assisted conversions

   Organic is often the first touch while the lead arrives two weeks later via direct. Looking only at last click makes SEO look weaker than it is. A model that counts every touch typically adds 20–40% to the channel’s credited weight.

What to watch while the money is still zero

For the first few months the revenue metrics read as zero, and that is normal — there is a two-to-three-month lag between work and leads. We mapped out what appears when in the piece on how long SEO takes. To avoid deciding blind, you need leading indicators: they react earlier and show whether the project is moving.

Vanity metrics and what to use instead

How to build this report

Technically all of it comes from Google Analytics 4 plus Search Console — no separate paid tooling required. We covered the setup order in detail in the GA4 and Search Console guide, and the console itself in its own walkthrough. Here is the minimum without which none of the five metrics can be calculated.

1. 1Conversions as events. Form submit, phone tap, Telegram chat start — each as its own event, marked as a key event. Without this there is nothing to count "organic leads" from.
2. 2GA4 linked to Search Console. This puts queries next to on-site behaviour. Without it you see either queries with no conversions or conversions with no queries.
3. 3A non-branded segment. A filter excluding your company name and domain in every spelling, including misspellings and transliterations.
4. 4Page groups. Tag commercial pages, blog and homepage — this split is what makes the first metric useful.
5. 5A 28-day window. Week-to-week organic noise exceeds real change, so weekly conclusions are almost always wrong.

Frequently asked questions about SEO metrics

In short

• Rankings are personalised and traffic is easy to inflate with pages that never convert — neither proves the channel pays off.
• Five numbers are enough: organic leads by page type, non-branded share, pages earning traffic, cost per organic lead, assisted conversions.
• Site-wide average position averages queries that are not comparable and means nothing — delete it from the report first.
• There will be no revenue signal for the first two to three months; watch impressions, unique queries and indexation speed instead.
• All of it is measurable with free GA4 and Search Console, provided conversion events and a non-branded segment are configured.
• A working report fits on one screen. If it needs narrating out loud, it is a bad report.