SEO KPIs: the metrics worth reporting, the ones that mislead, and a scorecard that holds up

The SEO KPIs worth reporting to a board are the ones tied to money and demand: non brand organic revenue or qualified leads, non brand organic clicks, conversion rate by landing page group, and the share of your priority queries where you hold a first page position. Keep the working metrics such as indexed pages, crawl errors, publishing throughput and position distribution on a separate operational dashboard. The common mistake is reporting activity as if it were outcome, and the common fix is to split one document into two: a scorecard with four or five numbers, and a working report underneath it.

This page covers which metrics to report, which ones mislead, and how to build a scorecard that survives questioning.

Separate brand from non brand first

This is the single change that makes every other number honest. Brand search tracks awareness created by everything else the company does: advertising, press, sales, product. Non brand search is the demand SEO actually captured. A total organic number that mixes them will show growth whenever marketing runs a campaign, and will hide a decline in the part you are responsible for.

Build it in Search Console with a query filter that excludes your brand name and its common misspellings, then apply the same split in analytics by landing page group. Report non brand as the headline and keep brand visible alongside it as context.

The metrics worth reporting

KPI	The question it answers	Source	Cadence
Non brand organic revenue or qualified leads	What did the channel contribute to the business?	Analytics and CRM	Monthly and quarterly
Non brand organic clicks	Is captured demand growing?	Search Console	Monthly
Conversion rate by landing page group	Are we winning the right searches?	Analytics	Monthly
Share of priority queries on page one	Are we visible where we decided to compete?	Rank tracking	Monthly
Indexed share of priority pages	Can Google use what we published?	Search Console	Monthly
Pages earning clicks	How much of the site is doing work?	Search Console	Quarterly
New referring domains	Is authority still being built?	Link data provider	Quarterly

Two of those deserve more explanation. Share of priority queries on page one beats average rank because it is a fraction of a list you chose deliberately, so it cannot be improved by quietly adding easy terms. Define the list once, write it down, and change it only with a note in the report. Pages earning clicks, as a share of indexed pages, is the best single measure of content quality at scale: if the number of pages doubles and the share earning clicks halves, you published volume rather than value.

The metrics that mislead

• Total organic sessions. Rises on brand demand, seasonality and a single viral page. On its own it tells a board nothing about SEO.
• Average position. A blend across every query, device and location in the filter. Adding a page that ranks fifth for twenty new long tail terms can make your average worse while the business improves.
• Keyword rankings for a handful of head terms. Results are personalised and localised, so a single tracked position is one sample from a distribution. Report the distribution.
• Impressions. Useful for spotting a change in how often you appear, and a weak headline, because an impression at position 40 is counted the same as one at position 2.
• Domain authority scores. Third party estimates that are handy for comparing link profiles, and not numbers Google uses. Keep them out of a board pack.
• Bounce rate and time on page. On informational content, a fast answer looks identical to a failed visit.
• Published post count. This is throughput, an operational number. Reporting it as a KPI invites the question of what the posts earned, and that question should be answered in the same document.
• Core Web Vitals pass rate as a headline. Important engineering work, and it belongs on the operational dashboard rather than next to revenue.

Leading and lagging, in the same view

Organic revenue is a lagging indicator: it tells you about work finished months ago. A scorecard with only lagging numbers leaves a quarter where nothing looks like it is happening. Pair each lagging KPI with the leading indicator that precedes it.

• Priority pages published and indexed, which leads position.
• Queries moving into positions 4 to 10, which leads clicks.
• New referring domains to priority pages, which leads position on competitive terms.
• Internal links added to priority pages, which leads the same.
• Non brand clicks, which leads leads and revenue.

Showing the chain in that order is what turns a report into a forecast, because the board can see this quarter's inputs and understand which line they will move next quarter.

Building a scorecard a board will accept

• Five numbers on the first page. One of them is money. Each shows the current value, the target, and the change on the previous period.
• Define every metric in writing, including the filters and the date range. A metric that means something slightly different each quarter is the fastest way to lose the room.
• Use one source per metric. Search Console for search side numbers, analytics and the CRM for business side numbers. Never blend the two in one line.
• Quarter against quarter, not month against month. Monthly comparisons are mostly seasonality. Add year on year where you have the history.
• Annotate the chart. Releases, migrations, campaigns, outages and core updates. An unexplained dip gets remembered; an annotated one gets understood.
• Set targets as ranges with a date, derived from your own trailing growth rate and the opportunity in your priority list.
• Attach the operational report behind the scorecard: indexing, crawl errors, speed, publishing, links. The detail earns the credibility, and it stays on page two.

One warning worth repeating to any board. Google's own guidance on hiring an SEO says plainly that no one can guarantee a number one ranking on Google, and advises caution with anyone who claims otherwise. A credible forecast is expressed as a range of outcomes against a dated target, and anyone promising a specific position is selling something else.

This page covers choosing and reporting SEO KPIs. For the definitions of the Search Console metrics behind several of them, see a guide to Search Console. For attribution modelling across channels, that is a measurement project of its own and the KPIs above are designed to be readable without it.

Questions buyers ask about SEO KPIs

What is the single most important SEO KPI?

Non brand organic revenue, or non brand organic qualified leads where the sale happens offline. It is the only number that answers the question a board is actually asking.

How do I report SEO when the sales cycle is six months long?

Report pipeline created by organic alongside closed revenue, and show both against the leading indicators. A six month cycle makes the leading indicators the useful part of the report for most of the year.

Is traffic a vanity metric?

Total traffic is. Non brand organic clicks, segmented by landing page group, is a real demand measure. The difference is the segmentation.

How often should SEO KPIs be reported?

Operational metrics weekly for the team, the scorecard monthly for marketing leadership, and a quarterly view for the board. Changing the cadence to chase a weak month is how reports lose their authority.

Should share of voice be a KPI?

It works well as a secondary metric when it is calculated from a fixed, documented query list and the same competitor set every quarter. Recalculating the list each time makes the trend meaningless, so freeze it and note any change.

What target should I set in year one?

Set the first two quarters on leading indicators, priority pages live and indexed, and queries entering positions 4 to 10. Set revenue targets from quarter three, based on the conversion rates your own pages are already producing.

A line most scorecards do not have yet

A scorecard with brand and non brand organic, leading and lagging measures, still has no line for the part of the buying journey that happens inside an assistant. A buyer asks, gets a shortlist, and arrives later by name, which shows up as brand traffic with no obvious cause.

Add a small recurring measure: out of ten buyer questions, how many assistant replies name your company, and which sources do they cite? AI Knows Us runs those questions across ChatGPT, Gemini, Perplexity, Claude and Google AI Overviews and records the results, which gives the board a line that explains some of the brand growth.