SEO measurement
Most SEO reporting measures activity because activity is easy to count. These are the metrics that actually indicate progress, in the order they move, and the ones that quietly flatter a programme that is not working.
Measure in the order things move
Search performance moves in a sequence: pages get indexed, impressions rise, average position improves, click-through rate follows, and only then do clicks and revenue change. Reporting the last stage first makes good work look like failure for the first three months.
| Stage | Metric | Typical response |
|---|---|---|
| Indexing | Indexed pages, coverage errors | Days |
| Discovery | Impressions by query | 2 to 6 weeks |
| Competitiveness | Average position by query | 4 to 12 weeks |
| Attractiveness | Click-through rate by page | Immediate after a snippet change |
| Outcome | Organic clicks, conversions, revenue | 3 to 9 months |
What to read in Search Console
- Queries gaining impressions but stuck around positions 8 to 20: your highest-return improvement list.
- Pages with high impressions and low click-through rate: a title and description problem, not a content problem.
- New queries appearing over time: evidence that topical coverage is broadening.
- Page indexing report: watch for growth in 'crawled, currently not indexed', which usually signals thin or duplicated pages.
- Compare 28-day periods rather than week to week; weekly noise will make you undo work that was fine.
Metrics that flatter
- Total keywords ranking: inflated by hundreds of irrelevant terms at position 90.
- Domain authority scores: third-party estimates, not signals any search engine uses.
- Traffic without segmentation: brand searches rising after a marketing campaign is not SEO progress.
- Links reported but never verified: a promised placement is not a link until someone fetches the page and confirms it.
- Average position across the whole site: an average of unrelated queries is a number about nothing.
Measuring links honestly
A link report should be checkable. For each placement, record the live URL, the anchor text, the rel attribute, the date verified, and the evidence. Re-check on a schedule, because links are removed during redesigns and content clean-ups more often than anyone expects.
This is the standard LinkRobin holds itself to, and the aggregate results are published openly in the Link Index.
Reporting without inventing attribution
- State the model you used. Last-click and assisted conversions tell different stories and both are legitimate.
- Separate brand and non-brand queries in every traffic chart.
- Report what changed, what moved, and what did not — the last one is what makes the rest believable.
- Include the confidence level: seasonality, algorithm updates and competitor changes all affect the period you are reporting.
Frequently asked questions
How long before SEO shows results?
Impressions and average position usually move within weeks of publishing, because that only requires indexing and initial evaluation. Clicks and revenue lag by months, because they need a position good enough to be clicked. Anyone promising ranking movement on a fixed schedule is describing a purchase, not an outcome.
What is the single most useful SEO metric?
Impressions by query in Search Console. It is first-party, unsampled, and it moves before clicks do — so it tells you whether the work is landing while there is still time to adjust.
Are rank tracking tools accurate?
They are directionally useful and absolutely precise about nothing. Results are personalised and localised, so a tracked position is one sample of many possible results. Use Search Console's average position for reality and rank trackers for daily movement.
How do you attribute revenue to SEO?
Honestly and with stated assumptions. Organic search is usually one touch in a multi-touch journey, so report assisted conversions and last-click separately, and never present a single number as the whole truth.