If you’ve ever looked at Google Search Console and seen a page with an “Average Position” of 8 but almost no clicks, you’ve run into the most misleading metric in the entire report. Average position isn’t wrong, exactly. It’s just measuring something different from what most people assume it’s measuring, and for SaaS marketing teams making budget or content decisions off that number, the gap between the two can be costly.
What average position actually calculates
Average position is a weighted average of the rank at which your page appeared across every impression, for every query, on every day in the date range. Two things fall out of that definition that aren’t obvious from the label:
- It’s per-query, not per-page. A single URL that ranks #2 for a five-person-a-month long-tail term and #61 for a 10,000-search head term doesn’t show “position 2 and position 61.” GSC blends them into one number for the page, and that blended number depends entirely on how many impressions each query generated.
- It’s an average, so outliers get buried. A page that ranks #1 for eight obscure variants and #90 for the one term that actually has volume will often show an average position in the teens or twenties, which reads as “almost page one” when the reality is that the page is functionally invisible for the query that matters.
This is why you’ll see pages with a great-looking average position and a click-through rate near zero. The average is being pulled up by queries nobody is searching for, while the query with real demand sits far enough down the results page that almost nobody scrolls to it.
Where this actively misleads SaaS teams
The failure mode we see most often: a content or SEO team reports “average position improved from 34 to 22” as a win, when what actually happened is the page picked up impressions on a batch of ultra-long-tail queries (often branded misspellings, internal search term leakage, or a competitor’s product name appearing in a comparison post) that happen to rank well simply because almost nothing else targets them. The metric moved. Nothing that drives pipeline changed.
The inverse failure is just as common: a page that’s genuinely improving on the terms that matter, climbing from position 45 to position 18 on a real head term, can show a flat or worse average position if it simultaneously loses a handful of low-value long-tail impressions elsewhere. The number goes the wrong direction while the thing you actually care about is working.
What to look at instead
Average position is fine as a directional signal at the site level over a long time horizon, where the noise from individual queries washes out. It’s the wrong metric for anything more specific than that. For page-level or campaign-level decisions, three views tell you what’s actually happening:
- Position filtered by query, not blended. In the GSC Performance report, filter to the specific query you’re targeting and look at that query’s position in isolation. This is the number that maps to “can a human actually find us.”
- Position distribution, not the average. What percentage of impressions land in positions 1-3, 4-10, 11-20, 21+? A page can have an identical average position to another page while having a completely different distribution: one clustered just off page one (fixable with a moderate push) and the other scattered across page three through eight (a different problem requiring a different fix).
- Impressions-weighted CTR against expected CTR for that position band. If your actual CTR is meaningfully below the typical curve for your position, that’s a title/snippet problem, not a ranking problem, and no amount of further ranking work fixes it.
The practical takeaway
If average position is the only number a report shows you, ask for the query-level breakdown before drawing a conclusion from it. A single blended number across every query a page happens to rank for cannot tell you whether you’re winning the term that generates revenue or just accumulating noise from terms that don’t. For SaaS companies where a handful of head terms drive most of the pipeline-qualified organic traffic, that distinction is the entire ballgame.
