Rank Trackers and Google Search Console Disagree: The Conditions Differ

Your rank tracker says position 10. The average position in Google Search Console says 18. Same keyword, same week, and the numbers don't line up. Neither one is wrong. They measure under different conditions, so not lining up is the natural outcome.

Contents

Neither number is wrong: Google says so officially
Four things that make the numbers differ
AI search ended the idea of a single rank
How RevenueScope helps
FAQ

Summary / References / Related articles

TL;DR#

Google answers the question "the report says position 5, but position 8 is what I see when I search" officially: with variables such as search history and location, the position you see can differ from the average[1]
Average position is the average across the searches where your site appeared; a tracker's number is a fixed point taken under fixed conditions
Four things create the gap: the population being aggregated, personalization, the point in time, and how the data is collected
Ahrefs writes in its own help center that there is no such thing as a truly objective ranking[5]
The generative AI performance report added in June 2026 carries impressions only — neither position nor clicks are included[7]

1. Neither Number Is Wrong: Google Says So Officially#

Google has already published an answer to this mismatch.

The Google Search Console help center carries the item verbatim: the report shows an average position of 5, yet searching for it yourself shows position 8. The answer is that variables such as search history and location exist, so the position visible in your own search can differ from the average[1]. The gap is documented not as a defect but as expected behavior.

The reason becomes clear once you look at what the average position is an average of. When your page appears more than once in a single search, only the topmost position is recorded. Searches where your site didn't appear in the results aren't part of the calculation at all. What remains is averaged across every search where you were shown.

The calculation above is the one printed in the Google Search Console help center[1]. Six positions were held, and only two survive into the report. The average of those two, 2.5, is the number that reaches the screen.

And searches where you didn't appear never enter the calculation in the first place. There are months where average position falls precisely because exposure widened. Searches that had never shown your site start showing it, and those lower positions join the average. Nothing changed on your side, and the number still drops.

The rank tracker side isn't trying to reproduce that average. It fixes a location and a device, then returns the position found in the results collected under those conditions. One is the average across the searches where your site appeared; the other is a fixed point under fixed conditions. Two numbers with different definitions. If it's the clicks rather than the position that don't add up, the cause is a different one — start with how to read clicks that don't add up.

2. Four Things That Make the Numbers Differ#

The gap comes out of the population being aggregated, personalization, the point in time, and how the data is collected.

The first is the population. What Google Search Console collects is the portion of real user searches where your own site was shown[1]. What a rank tracker holds is the result collected under the conditions you set. Ahrefs states in its own help center that it cannot pull search results from the same IP network as your readers[5]. In Semrush, the location and device set on a campaign at the start cannot be changed later[6]. Different populations produce different averages.

The second is personalization. Google explains that search results change with factors such as location, language settings, device type and the search being run right now[2]. The Google Search Console help center itself reports that searching the same keyword may not surface your site in your own search[3].

The third is the point in time. Positions move day to day. Even within one day, morning and afternoon can land in different places. The further apart the moment the tool collected and the moment the user searched, the further apart the numbers[5].

The fourth is how the data is collected. Each tool designs its own way of reaching search results, and the relationship with Google's terms is a matter under discussion. That topic is handled in how rank trackers collect data and where Google's terms come in, so it gets a mention here and no more.

In the same help article, Ahrefs reports that there is no such thing as a truly objective ranking[5]. The side supplying the positions is explaining that a position is a conditional number. Try to settle which one is correct and the ground you would compare them on isn't there.

On the sample data, compare average position against estimated revenue

3. AI Search Ended the Idea of a Single Rank#

An AI Overview is counted as one position within the search results. Every link listed inside it is assigned that same position[1]. Ten links can be carried there, and the number recorded is the same one.

In AI Mode, each follow-up question is counted as a new search[1]. The person is continuing a single piece of research, and the record holds several searches.

Then in June 2026, a generative AI performance report was added to Google Search Console[4]. What appears there is impressions only — neither position nor clicks are included[7]. Google itself is publishing no position metric for the AI surface.

So a position has become a proxy metric defined differently on each surface. Position 10 in ordinary search results and position 10 inside an AI Overview cannot be compared on the same basis. In AI Mode, a position doesn't exist at all. Once the assumption of a single rank is gone, the question of which position 10 is correct has no answer. Ranks holding steady while clicks fall comes out of the same place. How visits arriving through AI Mode get recorded is covered in where AI Mode traffic shows up in GA4.

So what do you judge with instead? Stop trying to reconcile the two positions, put how much each keyword earns at the center, and the gap stops interfering with the decision. Whatever the position, protect the keywords that are making revenue. Push back the ones that aren't. That decision doesn't need the two positions to agree.

What Google Search Console gives you, though, stops at clicks, impressions, CTR and average position. How much revenue a keyword drove doesn't appear there.

RevenueScope solution

A keyword at average position 4, and a keyword at 9. The one earning revenue was the one at 9. Follow the position numbers alone and the reversal goes unnoticed.

RevenueScope adds estimated revenue to the per-keyword results from Google Search Console. Clicks, impressions, CTR, average position, the most-clicked page and estimated revenue appear on the same row for each keyword. Clicks and estimated revenue carry a comparison against the previous period. Estimated revenue is a conservative approximation, calculated by multiplying revenue per search-referred session by clicks. AI-referred traffic is captured too, separated by answering source such as ChatGPT or Claude.

Here is one month for Fictional Store B.

Search keyword	Position	Clicks	Estimated revenue
sensitive skin toner	9	300	¥240,000
what is toner	4	500	¥30,000
toner refill	15	80	¥90,000

Note: the demo screen doesn't hold a fixed table like the three rows above. It runs on the sample store's sample data for that day, so both the keywords on show and their positions change (the table above is a rounded teaching example).

"Sensitive skin toner", the one sitting lower, earns 8× what "what is toner" at position 4 earns. People who came to look up a definition click and don't buy; people who searched from a problem buy even from a lower position. The one to work on next is the one at position 9. Not lowest position first, not most clicks first: protect the keywords with the largest estimated revenue, then grow the keywords sitting just below them in position. That is a decision the gap between the two positions can't push around.

In the sample store, do the keywords that rank highest also earn the most?

FAQ#

Frequently asked questions#

Q. Which position should I trust in the end?

A. Split them by job. To know how your site looks on average, use the average position in Google Search Console; to follow a trend under fixed conditions, use a rank tracker. Neither has to be dropped. Bear in mind, though, that neither number answers how much revenue a keyword generated. Budget and effort get allocated on other material.

Q. Does the same thing happen with rank trackers built in other markets?

A. Mechanically, every tool fixes the conditions of collection. As long as a location, a device and a collection time are set, the result is a different number from the average across the searches where your site appeared. Designs differ by product, though, so no claim is made here about how any particular tool behaves.

Q. Should we stop using rank trackers?

A. No. Measuring under the same conditions every day is a design suited to following a trend. Because location and device are held level, a rise and a fall can be compared. The average position in Google Search Console moves on a change in who is searching alone, so day-to-day comparison carries noise. The two are tools with different jobs.

Summary#

The reason a rank tracker and Google Search Console don't agree is that they measure under different conditions. Google explains that the position in your own search can differ from the average[1], and Ahrefs writes that there is no truly objective ranking[5]. AI search arrived on top of that, and Google itself publishes no position for the AI surface[7]. A single rank that holds across every surface no longer exists.

So rather than spending time closing the gap, moving the axis you judge on is faster. Once you know which keywords are making revenue, the positions worth lifting narrow on their own (how to pick ranks worth chasing).

If a position rises by two, how much does revenue rise? Answer that with a number and the gap between the two positions has stopped being a problem.