Free to explore: filter winning sites by DR, traffic and niche  ·  Try the live explorer →

Average position in Search Console

A ladder showing rank

What is average position in Google Search Console? Most people treat it as “where we rank”, one number that should go down over time. It is an average of many rankings across many queries, devices and countries, and it moves when any of those change. A site can gain ground on every query it cares about and still watch its average position get worse. Knowing how the number is built is the difference between reading it and being misled by it.

What is average position in Google Search Console, exactly

Google’s Performance report metrics documentation defines position as “A relative ranking of the position of your link on Google, where 1 is the topmost position, 2 is the next position, and so on.” Average position is then “the topmost position occupied by a link to your property or page in search results, averaged across all queries.”

Two details in that sentence carry most of the meaning. First, only the topmost link counts: if your site appears twice on the same results page, the higher of the two is the one recorded. Second, the averaging happens across queries. Every query you showed up for, however irrelevant, contributes.

A position column in a report

How to calculate average position in Google Search Console

Google gives a worked example in the same Help article. Your property appears for two queries, with several links on each results page.

QueryPositions where your links appearPosition recorded
Query 12, 4, 62 (the topmost)
Query 23, 5, 93 (the topmost)
Average position(2 + 3) / 2 = 2.5

Six links, two queries, one number: 2.5. The lower links at positions 4, 5, 6 and 9 do not affect the result at all. Only the best placement on each query counts, and then those best placements are averaged.

Several results from one site

The calculation is simple, which is exactly why it misleads. A single figure cannot tell you whether it came from two queries at 2.5 or one query at 1 and another at 4. The average erases the distribution, and the distribution is usually what you need.

A calculator averaging numbers

Why the site-level number misleads

The received wisdom says a falling average position means you are losing rankings. Often it means the opposite. Here is an illustrative case with round numbers, not drawn from any real site, using the same simple averaging as Google’s example.

  • Before: your page ranks for two queries, at positions 2 and 4. Average position is 3.
  • After: both still rank at 2 and 4, and the page starts appearing at position 30 for a new, related query. Average position becomes 12.

Nothing got worse. You gained visibility for a third query, and the average quadrupled. This is the same mechanism that makes impressions rise while clicks stay flat, covered in our guide to impressions vs clicks. New low rankings add impressions and drag the average down at the same time.

The reverse trap is just as common. Lose a handful of deep, low-value rankings and the average improves, even though nothing you care about moved.

Branded queries distort it further. Searches for your own name tend to put you at or near the top, and they can make up a large share of the queries on a smaller site. A strong brand week pulls the average up; a quiet one lets the long tail pull it down. Neither tells you anything about the non-branded pages you are trying to grow, which is why the first filter worth saving is one that excludes your brand terms.

A query-level filter

Filter before you read it

The fix is to stop reading average position at site level. In the Performance report, tick Average position and then narrow the data before drawing conclusions.

  1. Filter by query. Average position for one exact query is close to what most people mean by “rank”. Use an exact-match query filter rather than “contains”, which reintroduces the blend.
  2. Filter by page. Combine the query with the URL you expect to rank. If several of your pages compete for the query, the page filter shows which one Google prefers.
  3. Split by device. Mobile and desktop results can differ, and a site-wide average mixes both.
  4. Split by country. Rankings vary by market. A blended figure can hide strong positions in one country and weak ones in another.
  5. Compare periods, not values. Use the date comparison and look at the direction of change for the same filtered query.
Mobile and desktop compared

The last point echoes advice Google gave when it first published these definitions: pay attention to changes in position rather than the absolute value. A query that moves from 8 to 5 over a quarter is a clear signal. A site average that moves from 18 to 21 is usually not. Our guide on how to use Google Search Console covers setting up these filters and comparisons in more detail.

The Pages tab is a useful second view. With a query filter applied, it lists every URL of yours that appeared for that query and the position each one held. If two pages swap places from week to week, the average for the query can look stable while Google is still deciding which page to show, and that is worth resolving with clearer internal links or a merge.

What average position cannot tell you

The limitation is that even a carefully filtered figure is still an average over the period. A query that ranked first for half the month and tenth for the other half can look like a steady fifth. Search Console gives you the daily chart for that query, and it is worth looking at before you describe any page as “ranking fifth”.

A trend line of position

It also cannot tell you what the results page looked like. Position is a relative ranking of links; it does not describe the features, panels or summaries around your link, and those decide how much attention a given position actually earns. A link at position 3 can sit far down a crowded page. That is why click-through rate at a given position varies so much from query to query, as our breakdown of CTR by position explains. For tracking a defined set of keywords over time, a dedicated rank tracker complements Search Console rather than replacing it; see how to track keyword rankings. Use average position to spot direction, filtered tightly, and check the live results page before acting on it.

Frequently asked questions

What does average position mean in Google Search Console?

It is the topmost position occupied by a link to your property or page in search results, averaged across all queries. Position 1 is the top of the results.

How is average position calculated?

Only your highest link counts for each query, and those values are averaged. In Google’s example, topmost positions of 2 and 3 on two queries give an average of 2.5.

Why did my average position get worse when traffic went up?

You probably started ranking for new queries at low positions. They add impressions and pull the average down, even though your existing rankings held.

Is a lower average position number better?

For a single filtered query, yes: 1 is the top. At site level a lower number is not reliably better, because it depends on which queries are included.

The takeaway Average position is a blend of queries, devices and countries. Filter it to one query and one page, compare periods rather than values, and treat the site-wide figure as context, not a score.