How to Read Average Position in Google Search Console

How Search Console average position is calculated, why it rarely matches your real ranking, and how to read it with filters and impressions.

Author: SEOLOK Editorial TeamPublished: Updated: 6 min read

Search Console performance chart with clicks, impressions and average position lines

Short answer

Average position in Search Console is the average of your site's topmost position across every impression it received in search results. Because different queries, countries, devices and days are blended into one number, it often differs from your actual ranking for a specific keyword. Read it with query, page, country and device filters applied, always alongside impressions, and as a trend over time rather than a single figure.

Open the Performance report in Search Console and you'll see four boxes: total clicks, total impressions, average CTR and average position. The first three are fairly self-explanatory. Average position is the one most often misread. "My average position is 14, so I'm on page two" is a conclusion that frequently has little to do with reality.

This article explains how Search Console average position is calculated, why it can produce confusing numbers, and how to turn it into something you can actually act on.

What is average position and how is it calculated?

According to Google's Search Console documentation, position is where your result appears in search results, with 1 being the top. If several of your pages show up for the same search, the topmost one counts for that impression. Average position is the average of those positions across impressions.

Google's deep dive into performance data describes position simply as the average position in search results for the URL, the query or the website in general, and the Search Analytics API returns the same metric as "position", the average position in search results.

The key concept is impressions. Position is only calculated from searches in which your site was actually shown. Searches where you didn't appear have no effect on the average.

A simple example

The table below shows a week of entirely hypothetical data for one page:

QueryImpressionsPosition for that query
"accounting software"1003
"free bookkeeping software"5012
"e-invoicing accounting software"85030

Weighted by impressions, this page's average position works out roughly as:

(100 × 3 + 50 × 12 + 850 × 30) ÷ 1,000 = (300 + 600 + 25,500) ÷ 1,000 ≈ 26.4

Yet the page ranks 3rd for the keyword you care about most, "accounting software". A high-impression query where you sit far back drags the whole average down. Concluding that the page is failing because "average position is 26" would be the wrong call here.

Why average position can mislead you

1. Very different queries collapse into one number

As the example shows, a page- or site-level average reduces queries in very different situations to a single figure. Ranking near the top for a few queries and far back for many looks like being "somewhere in the middle".

2. New impressions can lower the average

When your site starts appearing for new keywords, it usually enters at the back first. The result: impressions rise while average position "gets worse". In reality, that's a positive sign that your visibility is growing.

The reverse happens too. Drop out of queries where you ranked poorly and average position "improves", even though your total visibility may have shrunk. Never read average position in isolation from impressions.

3. Countries and devices get mixed together

Without filters, mobile searches in one country sit in the same average as desktop searches in another. You could rank 2nd in one market and 40th in another; the average shows you neither.

4. Date ranges smooth everything out

A three-month average position compresses every rise and fall in that window into a single number. A sharp drop last week may barely register.

5. Anonymized queries aren't in the table

Google explains that queries not issued by more than a few dozen users over a two-to-three-month period are anonymized for privacy and never appear in the tables. They are included in chart totals but excluded whenever you apply a query filter. This is one reason summing table rows won't match the chart.

6. Row limits

Per Google's documentation, the Search Console interface exports a maximum of 1,000 rows, while the API can return up to 50,000 rows per day per site per search type. On very large sites, some query and page rows may simply never appear.

Six steps to reading average position properly

1. Filter first

Narrow the data down before trusting the number:

  • Query filter: a single query or a group (for example, non-brand queries).
  • Country filter: your target market.
  • Device filter: look at mobile and desktop separately.
  • Page filter: a single URL or a folder.

With a single query, country and device selected, average position comes much closer to your real ranking for that keyword.

2. Read it next to impressions

Judge every position change together with the change in impressions:

ImpressionsAverage positionLikely reading
RisingWorseningYou're starting to appear for new queries, from further back
RisingImprovingBoth visibility and rankings are up; the best case
FallingImprovingYou dropped out of weak queries; overall visibility may be down
FallingWorseningPossibly a genuine ranking loss; worth investigating

3. Watch the trend, not a single number

Switch on "Average position" in the chart and follow the line over time. The Compare tab lets you put two periods of equal length side by side. Google's traffic drop guide likewise advises not to focus too much on absolute position and to treat clicks and impressions as the real measure of success, while suggesting you reassess your content if you see a dramatic, persistent drop in position.

4. Go down to the query and page tables

Instead of the overall average, open the position column in the Queries and Pages tabs. Seeing which queries sit in positions 1–3, 4–10 and 11–20 tells you far more than one average.

5. Focus on the opportunity zone

The most valuable insight usually lives in queries with lots of impressions sitting between positions 4 and 20. Google already sees your site as relevant for them, but you're not at the top yet. Improvements to titles, content depth and internal links tend to pay off fastest here. The free keyword opportunity finder lists these queries for you.

6. Cross-check with rank tracking

To be sure a position really changed for an important keyword, compare the Search Console average with a ranking measured on a fixed device and location. Our guide to what Google rank tracking is explains where the differences come from, and why Google rankings keep changing covers the natural day-to-day movement.

Search Console position vs rank tracking

Search Console average positionRank tracking
SourceReal user impressionsA measurement under set conditions
CoverageEvery query you got impressions forThe keywords you choose
Location/deviceAll mixed unless filteredThe chosen device and location
Keywords you don't appear forNot includedMeasured, shown as "not in top X"
FreshnessA few days behindAs of the measurement
Best used forOverall trends, finding opportunities, click analysisExact position and competitors for target keywords

They answer different questions. Using them together beats replacing one with the other.

Quick questions

If my average position is 1.0, am I first for every search? It means your topmost result was in position 1 for the searches where you received impressions. Unless you filtered, it says nothing about keywords you target but don't appear for.

Why is average position a decimal, like 7.4? Because you appeared in different positions across different impressions, for example 7th in some searches and 8th in others.

How far back can I look? The Performance report shows the last 16 months. Longer-term analysis requires storing the data somewhere else; see why Search Console data is limited to 16 months.

Average position in SEOLOK

SEOLOK's Search Console analytics connects read-only through your Google account, imports the last 16 months of daily data (query, page, country, device, clicks, impressions, CTR and average position) and then adds each new day. The monthly report lists high-impression searches in positions 4–20 as "opportunities" and estimates the extra clicks you could expect if they reached the top 3. So instead of staring at a single number, you see average position together with impressions and clicks, as a list you can act on.

Reading average position well means moving from "am I doing well or badly?" to "which query, which page, and what should I improve?" Filter the data, read it with impressions and follow the trend; those three habits alone get a lot more out of Search Console.

Free tool

Keyword opportunity finder

Analyses your Search Console query and page data and lists, in priority order, searches with lots of impressions but few clicks, keywords in positions 4–20 and pages due a refresh.

Use the tool

Want a free SEO report for your site?

Our team reviews your rankings, technical health and opportunities and prepares a report for you.

A free SEO report for your site

Leave your details; our team will prepare a report on your rankings, technical health and opportunities and get in touch.

Tags:search consoleaverage positionperformance reportgoogle rankings

SEOLOK Editorial Team

SEO and product team

The SEO and product team behind SEOLOK. We base our articles on Google Search Central, web.dev and Search Console documentation and on what we learn building our own measurement stack. Every article lists its sources and last update.

Search Console

Why Is Search Console Data Limited to 16 Months?

Why Search Console only keeps 16 months of performance data, what you lose when days roll off, and how to archive history with the API, BigQuery or a tool.

SEOLOK Editorial Team · · 7 min read

Rank tracking

Why Do Google Rankings Keep Changing?

Why Google rankings fluctuate day to day: algorithm updates, competitors, shifting intent, location and device, and how to tell noise from a real drop.

SEOLOK Editorial Team · · 7 min read