Home / Insights / Growth & Measurement

Growth & Measurement

How to Measure SEO and GEO Without Vanity Metrics

Build reporting on four things: Search Console's defined metrics, analytics key events, revenue or qualified-lead outcomes, and honest notes about the limits of AI-surface visibility. Google itself warns against optimizing for clicks instead of the overall value of visits.

Define the metrics before the dashboard

Search Console defines impressions, clicks, average position, and CTR as clicks divided by impressions, with position shown only for Google Search results. [Search Console Help, What are impressions, position, and clicks?]

An AI Overview occupies a single position and all links in it share that position; clicks to external pages in AI Overviews and AI Mode count as clicks. [Search Console Help]

Building a report before agreeing on these definitions is how most dashboards end up unreliable: two people can look at the same word, position, and mean different things by it once AI surfaces are mixed into the data.

  • Search Console excludes data from Search Labs experiments. [Search Console Help]
  • Search Console's UI covers 16 months; longer histories require the Search Analytics API or bulk data exports. [Google Search Central, Debugging search traffic drops]
  • A short written glossary of exactly how each metric in your report is defined saves significant time whenever a new team member or partner needs to be brought up to speed.

Outcome layer

A key event is an event measuring an action important to business success, and any collected event can be marked as one. [Google Analytics Help, About key events]

Only key events are eligible for import as Google Ads conversions, and linking or marking can take 24 to 48 hours to propagate. [Google Ads Help, About conversion measurement]

Measurement discipline tends to matter more in slow periods than in fast-growth periods, since a rigorous reporting model is what tells you honestly whether a slowdown is seasonal, competitive, or self-inflicted.

The AI-surface honesty section

AI-feature appearances are folded into overall Search traffic in the Performance report under the Web search type, not reported separately. [Google Search Central, AI features and your website]

Google advises considering conversion indicators, sales, signups, engaged audience, information lookups, rather than focusing too much on clicks. [Google Search Central, Top ways to ensure your content performs well in Google's AI experiences]

Resist the temptation to build a bigger dashboard before fixing the definitions underneath it. A small, well-defined report beats a large, ambiguous one every time a real decision is on the line.

Diagnosing changes

Use the same traffic-drop framework for every anomaly: check whether the drop is site-wide, page-group, or single-page, and check seasonality before concluding something broke.

Keep a simple log of every site change alongside the traffic data, since correlating a specific deploy or content change with a subsequent shift is far easier with a timestamped record than with memory alone weeks later.

Share the same report format with every stakeholder, internal team and outside vendor alike, so that a disagreement about performance is a disagreement about facts rather than a disagreement about which numbers to trust.

A one-page monthly report template

One page, four sections: Search Console metrics, analytics key events, revenue or qualified-lead outcomes, and a short honest note on AI-surface visibility limits, is enough for most founder-led businesses to act on.

Resist the pressure to add more metrics to this page over time. A report that grows every month eventually stops being read carefully, which defeats the purpose of keeping it to one page in the first place.

Revisit the reporting model itself every few months, not just the numbers inside it, since a metric that made sense a year ago can quietly stop reflecting what the business actually needs to know.

Metrics to retire

Retire any "AI citation count" as a headline metric, since no fetched primary source in this research documents an audited citation metric from any engine, and treating an unaudited number as a KPI invites false confidence.

Also retire raw traffic as a standalone success metric on its own. Paired with conversion and revenue data it is useful context, but on its own it rewards exactly the kind of low-value, high-volume content Google's own guidance warns against.

Where a vendor's reporting disagrees with your own, treat the discrepancy as a definitions problem first, and trace both numbers back to their source before assuming either one is wrong.

Frequently asked questions

Can I report AI Overview clicks separately?

No; they are included in overall Search performance (Google, https://developers.google.com/search/docs/appearance/ai-features).

Is average position a ranking?

It is a relative position measure for Google Search results (Search Console, https://support.google.com/webmasters/answer/7042828).

How far back does data go?

16 months in the Search Console interface (Google, https://developers.google.com/search/docs/monitor-debug/debugging-search-traffic-drops).

Should traffic be the headline metric?

No. Google recommends judging the value of visits and conversions instead (Google, https://developers.google.com/search/blog/2025/05/succeeding-in-ai-search).

Can I count AI citations?

No audited primary source documents an official citation metric from any engine in this research.

Primary sources

Policy and statistical claims in this guide are grounded in the sources below. Access dates and policy details can change, so verify regulated guidance before acting.

Seven clients. One of them could be you.

Tell us what you're building and what's stalled. If we're not the right fit, we'll say so on the call - and point you somewhere better.

Start a project