Keyword rank tracking with the context

Every tracker will tell you that you are fourth. Ours tells you what is above you, because fourth under a generated answer, a video row and four ads is not fourth in any sense your client would recognise.

A rank tracker checks where your domain appears for a set of keywords, on a schedule.The number on its own has become unreliable as a proxy for traffic, because what sits above the first organic result changes constantly. So we record the position and the shape of the page it sat on, and report them together.

The problem with a position on its own

Third place used to be a reliable description of what a searcher saw. It is not any more, and the reason is that the space above the first organic result has been filling for a decade and filled fastest in the last two years.

The same position, on two different results pages, can mean the top of the screen or below three folds of other content. A report showing a client moved from position five to position three, on a query where a generated answer now occupies the first screen, is technically accurate and actively misleading. That is the failure this page is about.

AI OverviewA generated answer above everything. The most displacing feature there is, and the one no position number accounts for.

Featured snippetOne result promoted above the rest, often answering the query outright.

People also askAn expanding block that pushes organic results down and absorbs the follow up question.

Paid resultsUp to four before the first organic result on commercial queries.

Local packA map and three businesses, which takes most of the first screen on anything with local intent.

Video and image carouselsHorizontal blocks that break the vertical list and reset the reader.

Shopping and knowledge panelsProduct grids and entity boxes, both of which change what first place is worth.

What we record on every check

Four things per keyword per run, kept as four things rather than collapsed into a score.

Best organic position: The domain's highest placement in the depth we looked at. Not appearing at all is recorded as exactly that, rather than as a blank or a zero, because "not in the top hundred" is a real answer and treating it as missing data hides trends.
The URL that ranked: Which page placed, which is how you find out that the wrong one is ranking for your money term.
SERP features present: Everything above and around the result, so a position can be read against what is actually on the screen.
The other domains in the top ten: Who you are actually competing with on this query, which is frequently not who you assumed.

Search Console is a second opinion, not a second input

Connect Search Console and you get its numbers next to ours, clearly labelled and never merged. This is a deliberate decision and it is worth explaining, because most tools in this category quietly blend the two.

A measured position is one location, one language, one device, at one moment. A Search Console average position is the mean across every impression your site actually received, blending every location, device and personalisation that occurred. Those are two different facts about two different populations. Averaging them produces a number that describes neither, and the reason it is popular is that a single number is easier to put in a report than two that disagree.

When they disagree, that disagreement is usually the most informative thing on the screen: it generally means you rank well somewhere you are not getting impressions, or the reverse.

The measurement that sits next to this one

Rank tracking answers where you place in a list. It does not answer whether an assistant names you when somebody asks it what to buy, and that is now a separate question with a separate instrument.

That measurement is a rate rather than a position: take a fixed prompt set, run it against the assistants, and count how often you are named and how often a page of yours is cited. It has to be re-run, because the answers vary between runs on an unchanged site. We do both in the same product, and we keep them apart, for the same reason we keep Search Console apart.

AI visibility monitoring covers that half. Between them they answer the two questions a client actually has, which are whether anybody can find you and whether anybody recommends you.

Related free tools

Index checkerWhether the page can rank at all Canonical checkerThe most broken thing we measure Technical SEO checklistOrdered by measured failure rate Keyword research methodChoosing what is worth tracking

Fair questions

What is a keyword rank tracker?

A tool that checks where a domain appears in search results for a set of keywords, on a schedule, so you can see movement rather than a snapshot. The useful ones record what else was on the results page at the same time, because a position on its own does not tell you whether anybody could see you.

How do I track keyword rankings across multiple sites?

Set each site up as its own project with its own keyword set and its own market, then read them side by side rather than merging them. Positions are only comparable within a single domain, location and language, so a single blended number across several sites is not a measurement of anything. Agencies running many clients want per project tracking with one place to see all of them, not one combined average.

Is rank tracking still useful with AI Overviews?

Yes, and only if the tracker records the features. Ranking third has meant different things at different times, and it means something very different on a page where a generated answer occupies the first screen. A tracker that reports the number and not the context is measuring something that has quietly stopped correlating with traffic.

How often should rankings be checked?

Weekly is right for most sites. Daily checking mostly measures noise: results vary by data centre, personalisation and ongoing testing, so day to day movement of a place or two is usually not a change in anything. Every SERP check costs money, and spending it on daily resolution buys precision you cannot act on.

Why do rank trackers disagree with Search Console?

Because they measure different things. A tracker takes a measured position from a specific location, language and device at a moment in time. Search Console reports an average position across every impression your site actually received, which blends locations, devices and personalisations. Both are correct and they answer different questions, so we keep them separate and never average them.

Can I check one keyword for free?

Yes. A free account covers a working set of keywords with scheduled checks, which is enough to see whether the approach is telling you anything before you pay for more. Every check costs us a real API call, which is why depth and volume scale with the plan rather than being unlimited.