Google ranking: how do I find out my site's position?

There is a question every marketing manager asks in the first week of an organic search project, and that rarely gets an honest answer: where on the results page is my site right now?

The honest answer is that there is no single place. Your Google ranking position varies with who is searching, from where, on which device and with what history. Two people in the same city, in the same minute, can see different orders.

That does not make measurement useless. It makes measurement a matter of method. Anyone who understands what each tool measures stops arguing about the number and starts discussing the trend, which is where content decisions actually happen.

The question has also become more urgent in recent months. The tracking tools the industry has relied on for a decade started diverging from Google's own panel more often, right as AI-generated summaries took over the top of the screen.

What is Google ranking position and why does it move so much?

Google ranking position is the place a page holds in the list of organic results for a specific query. It is not a fixed attribute of the site, but the outcome of a calculation made at the moment of the search, weighing the query, the location, the device and the recent history of the person searching.

That last part is what confuses people. The Performance report documentation records that you may not see your site when you repeat the same query, because results are specific to the time, place, device and recent history of the person searching.

Caption: Google ranking position is not a fixed number: it depends on who searches, from where and on which device.

In other words, the movement is not a flaw in the measurement. It is a feature of the product.

Add to that the changes in the algorithm itself. Broad updates reorder results at scale, and not every fluctuation comes from a confirmed update. Much of what the industry calls volatility is the system continuously reassessing pages.

There is also a structural factor that weighs more than most people assume. The results page stopped being a list of ten links. It now holds AI-generated summaries, question blocks, carousels, videos, local panels and shopping results.

Position three today can sit visually lower on the screen than position five did two years ago, because more elements push links down. Measuring the position number alone, without looking at what happened to clicks, leads to wrong conclusions.

That is why the correct reading treats position as one of the health signals of organic traffic, not as the scoreboard of the game.

How do I check my website's position on Google in Search Console?

Search Console is the official and free source for checking your Google ranking position. The path is the Performance report, where the average position metric is off by default and has to be switched on. From there, the data opens by query, page, country, device and date.

The steps are short. Inside a verified property, open Performance, tick the Average position box in the top panel and choose your analysis window.

The table under the chart organizes the data into tabs. The Queries tab shows which terms your site appears for and at what average position. The Pages tab does the same by URL.

Filtering is what turns the report into a decision tool. Without a filter, the average position of the whole site is a number with almost no use, because it mixes brand queries with generic ones, and those two behave in opposite ways.

One definition is worth knowing because it changes the interpretation. Google documents that average position corresponds to the position of the highest-placed result from your site for that query. If three of your pages appear at positions 4, 12 and 30, the value counted is 4.

Two limits need to be clear before any report. The first is data trimming: queries with very low volume may not appear in the list, due to anonymization. The second is the interval, since the data usually lags by a few days.

There is also a recurring point of confusion. The panel dedicated to artificial intelligence features in Search Console does not deliver position, but impressions, as detailed in the AI report in Search Console. Anyone looking for position stays in the traditional Performance report.

Why does the rank tracker position not match Search Console?

Rank trackers and Search Console measure different things, which is why they will never coincide. The tracker simulates a search under controlled conditions and records the position in that simulation. Search Console aggregates what actually happened with real people, across every context, and returns an average.

An example makes the difference concrete. Imagine a query where your site appears at position 3 for someone searching in São Paulo on mobile, and at position 11 for someone searching in Recife on desktop.

A tracker configured for São Paulo and mobile will report 3. Search Console will report something between 3 and 11, weighted by the real volume of impressions in each context. Both are right, answering different questions.

Here is how the two sources compare on the criteria that matter most in a report:

Table: Comparison between official average position and position measured by tracking tools.

The gap tends to widen whenever Google changes the structure of the results page. Every time the platform makes automated collection of search data harder, third-party tools have to adapt, and the adaptation window shows up as a hole or as noise in the dashboards.

That is the moment of highest risk in a content operation. Not because the data vanished, but because someone can read a broken gauge as a performance drop and change strategy for a reason that does not exist.

The practical rule is simple: before reacting to a position drop in the tracker, check Search Console. If the official panel does not show the same movement, the problem is probably in the collection.

What does Google's /goto redirect change about measuring position?

Google's /goto redirect is an intermediate stop added to the links on the results page: instead of pointing straight at the destination site, the link now runs through google.com/goto with the address encoded. Google confirmed the rollout on August 26, 2026, describing it as a technical measure against results scraping.

For anyone searching, the effect is small. The destination address still shows beneath the snippet title, and what disappears is the URL preview in the browser bar before the click.

For anyone measuring position, the effect is large. The encoded address cannot be decoded locally, so finding out where each result points now means following the link all the way through.

The cost shows up as request volume. Derek Perkins, of Nozzle, calculated that resolving every link in a five-page ranking takes between 500 and 1,000 requests, because Google blocks HEAD requests and forces a full GET that follows the redirect.

Request volume was already climbing before this. In September 2025 Google removed the num=100 parameter, which multiplied by ten the number of calls needed to assemble a ranking, and that multiplication now applies per results page as well.

Comparability of the historical series is the point where vendors disagree, and both readings belong in the decision. Industry reports suggest that data collected before June 2026 may not be comparable to what came after, because the collection method changed.

SISTRIX said the opposite about its own database: the completeness and comparability of its data would not change, and only its refresh cycles would run slower during the transition.

Because those claims come from different vendors, the question has no single answer, and it should not have one. The answer that matters is the one from the tool you pay for: ask your vendor whether the collection method changed, and from which date.

Search Console was not affected by the change. Its data does not come from scraping the results page but from Google's own infrastructure, which is what keeps it the auditable source for the report.

Analytics deserves extra scrutiny during this period. Following the redirect at scale can log visits that no person made, which tends to muddy the traffic-source reading while the transition runs.

The change ends up reinforcing a separation that already existed. A third-party tracker serves the reading of movement and competitors, and Search Console serves the number that goes into the client report.

Diagnostics deserve the same separation. Confirming how Googlebot reaches the site solves a different class of problem, the crawling one, which often gets mistaken for a position drop when the tracker wobbles.

Which Google ranking position belongs in the report?

The Search Console position is the one that belongs in the report, because it is official, covers real searches and does not depend on third-party collection. The tracker stays useful, but as an instrument for daily monitoring and competitor comparison, not as the number that supports an account review.

That choice solves a political problem before it solves a technical one. When the number in the report comes from a source the client can audit alone, the conversation stops being about whether the tool can be trusted.

Build the report with three cuts of position, not one. The first is average position for non-brand queries, which measures real gains. The second is position for the priority pages of the funnel. The third is variation over the period, which shows direction.

Brand queries deserve a separate cut for a specific reason. They almost always sit in the top positions and pull the site average up, creating a sense of performance that generic search does not support.

One last formatting recommendation. Always bring position next to impressions and clicks, on the same line of the report. Position alone invites the wrong reading, because moving from 15 to 8 without gaining a single click is a very different result from moving from 4 to 3 and doubling traffic.

How does AI search change the reading of Google ranking position?

AI search does not replace Google ranking position, it changes what the position delivers. Generated answers are assembled from pages that already rank well, so staying well positioned remains the entry condition. What changes is the click-through rate attached to each position.

Google is explicit about what it takes to appear in these features. The official page on AI features in Search states that there are no additional requirements and no special optimizations needed to appear in AI Overviews or AI Mode, and that the fundamentals apply: an indexable page, compliance with policies and helpful, reliable content.

That dismantles the idea that AI search requires a parallel discipline. John Mueller, from Google's Search Relations team, reinforced the point in August 2026: from our point of view, there is nothing really special you need to do for generative AI responses in search.

What changed in a measurable way was the click, not the mechanics. A study by SparkToro using Similarweb clickstream data found that 68.01% of Google searches in the United States between January and April 2026 ended without a single click.

That number explains why stable position can coexist with falling traffic. Rand Fishkin, co-founder of SparkToro, sums up the way out as investing in brand awareness and influence on the platforms where your audience already spends time.

On the measurement side, Google reports that clicks coming from pages with these features enter the Performance report under the Web search type. There is no separate cut that isolates how much came specifically from the generated summary.

The practical consequence is a loss of granularity with no official solution yet. You know the click happened, but you do not know whether the reader came through the summary or the blue link, which weakens any attempt to attribute results to a specific format.

In September 2026, Google itself acknowledged the limit of this metric. John Mueller said the old 1 to 10 scale has become hard to map for site owners, according to the Search Engine Roundtable report published on September 10.

The reason is structural. The tool counts the generative block as a single position, so being cited inside an AI answer and holding the first blue result can produce similar records with very different meanings for the reader.

More revealing than the diagnosis was the request that came with it: Google invited the community to suggest better ways to report the position. A company does not open a public consultation about a metric it considers settled.

It is also worth saving the history while it still exists, because the report window keeps moving and Google does not rebuild a lost period. The archiving routine is covered in the Search Console data guide.

In that scenario, the signal that best protects results is authority. Content with named sources, verifiable data and real first-hand experience has a better chance of being chosen as the basis of the answer, which is the logic behind E-E-A-T in practice.

It is worth following how the interface evolves, because it moves clicks directly. The behavior of links inside Google AI Mode keeps changing, and each adjustment redistributes attention on the screen.

Which metrics should I track alongside Google ranking position?

Google ranking position alone does not support a decision. It needs company: impressions, clicks, click-through rate and conversion. Together they answer not only where the site sits, but whether sitting there produces results, which is the question the business actually asks.

Impressions show demand. If impressions grow while position stays flat, the market is searching more, and the opportunity grew without any extra effort.

Clicks show appeal. Good position with low clicks points to a weak title and description, or to an AI summary resolving the question before the click.

Click-through rate is the most useful crossing of the four. It lets you compare pages at different positions and identify which one earns more than its position would suggest.

Conversion closes the chain. Without it, you can celebrate traffic growth that generates no contacts at all, a common scenario when content attracts an audience outside the profile.

There is one more indicator few people track and that deserves a place in the routine: how many distinct queries bring the site into the top three positions. Growth in that count signals gains in breadth, even when average position stands still.

When the fluctuation comes with a broad, simultaneous drop across several pages, the explanation may lie in an algorithm update, and the difference between the types of Google update helps calibrate the response.

How can Google ranking position guide SEO strategy?

Google ranking position becomes strategy when it stops being a scoreboard and turns into a priority queue. The most profitable use of the data is finding pages that already sit near the first page and improving them, because the effort to move from position 12 to 6 is far smaller than creating a new page.

Start with the cut that returns the most in the least time. In Search Console, filter queries with average position between 8 and 20 and meaningful impressions. Those pages already have algorithmic recognition and need little to reach where clicks happen.

The second cut is the opposite and equally valuable. Queries at good positions with low click-through rate are not a content problem, they are a packaging problem. The work there is title and description, and the result shows up within days.

The third cut looks at gaps. Queries where the site appears at a very low position with high impressions signal demand the current content does not serve well, and they are usually the best new topics for a content marketing plan.

Once triage is done, the work of lifting position follows the familiar repertoire of organic positioning: search intent properly served, clear structure, real depth and a decent reading experience.

One methodological precaution closes the loop. Record the position before touching the page and define the reassessment window in advance, between four and eight weeks. Without a baseline and a deadline, any variation becomes an argument for any thesis.

If the operation involves many pages and several owners, turning this routine into a documented process matters more than any new tool. That kind of discipline is what sustains consistent SEO work over time.

Frequently asked questions about Google ranking position

Is chasing Google ranking position worth it?

It is, as long as you chase the right position, measured at the right source. Chasing the average position of the whole site wastes time on a number that mixes everything together. Chasing position on the queries that bring the audience you want is another story, and that is where organic search pays back.

The shift in posture that makes the difference is small and uncomfortable. Swapping the question where are we ranked for the question which queries bring the people who matter, and where do we stand on those, reorganizes every content priority.

Anyone who makes that swap gains a side benefit. It becomes much harder to be fooled by an outdated tool, by algorithm fluctuation or by a pretty report that does not talk to the number of contacts generated.

And there is a gain in peace of mind that rarely enters the calculation. An operation that measures well stops chasing every fluctuation and starts investing where the return is predictable, which cuts rework and team fatigue.

If your institution has been tracking position for a while and still cannot connect that data to enrollments or contacts generated, the problem usually sits in the measurement routine, not in the content. It is worth talking to our team to review how your organic traffic is being measured.

Knowing the position is the easy part, moving it is the whole point of search ranking work.