A rank report can be accurate to the decimal and still describe a market you do not sell in. The location setting decides what every number means, and when it goes wrong, it goes wrong silently.
Nobody sets this up badly on purpose. Someone configures a rank tracker in good faith, types a town name the tool does not recognise, or picks a plausible-looking entry from a long dropdown, and the tool starts reporting. The numbers arrive on schedule, they move up and down in believable ways, and everyone reads them as the truth about their market. Whether they are depends entirely on a setting most people looked at once.
There is no single "position 7"
The idea that your site holds one true position for a query is a convenient simplification, and it stopped being true years ago. What Google shows depends on where the searcher is, what device they are on, and to some extent what they have searched and clicked before. Two people typing the same words in the same minute can see differently ordered pages, and both of them are seeing the real results.
This is not a flaw in the tools; it is the reason the tools exist in the form they do. Because there is no single answer, every serious rank tracker asks the same question when you set it up: where should we measure from? A country at minimum, often a city, sometimes a device type. The measurement is then an honest answer to a precise question — "where does this page rank for this query, as seen from this place, on this device" — and nothing more.
That precision cuts both ways. It means the numbers are meaningful, and it means the location setting is not a preference buried in a form. It is the thing that decides what the numbers mean at all.
Can a rank tracker measure the wrong country without showing an error?
Yes. Rank tools take a location setting when a measurement is set up, and an unrecognised or mistyped location can silently fall back to measuring a different market — even a different country. The numbers that come back look perfectly plausible, because they are real positions in a real market; it is simply not your market. There is no error and no warning, so the only defence is to check that the results describe the search landscape you actually operate in.
It is worth pausing on why this failure mode is so hard to notice. A tool that failed loudly — an error message, an empty report — would get fixed the same day. A tool that falls back quietly produces everything a correct report produces: positions between 1 and 100, competitors with real domains, movement from week to week. Every individual number passes inspection. The report as a whole is about somewhere else.
The practical consequence is uncomfortable. A Danish business can spend months reading its positions as seen from another country, celebrating improvements and worrying about drops that its actual customers never experienced. Decisions get made on those numbers — which pages to rewrite, which keywords to chase — and the decisions are not wrong so much as aimed at the wrong audience.
How do I know if my rank report is measuring the right market?
Compare the report against what you can see. Search for one of your tracked queries from your own market and look at who appears, then look at the competitors the tool lists for the same query. If they broadly match, the tool is looking at the same results page you are. If the report is full of domains you have never met — foreign competitors for a local query — the tool is almost certainly measuring from somewhere else.
The competitor list is the strongest tell, because it is the part of the report that cannot be plausible by accident. Positions from the wrong country look like positions from the right one; competitors do not. If you run a plumbing firm in Aarhus and your rank report says your main rivals for "plumber near me" are companies in another country, the number next to your own domain is answering a question nobody asked.
Your own search is not a perfect instrument either — your history and location nudge what you see — so do not expect the positions to match exactly. You are checking the shape of the results, not the decimals: the same kind of businesses, the same language, roughly the same names near the top. A rough match means the measurement is pointed at your market. No match at all means it is not, however healthy the numbers look.
What does average position in Search Console mean?
It is an average taken across every query, every searcher location and every device that showed your page, so it is a summary of many different rankings rather than a ranking itself. A page can hold position 3 in one country and position 40 in another and report an average around 20 — a position it may never actually have occupied anywhere. The number is useful for spotting movement over time, not for knowing where anyone in particular sees you.
This is the same location problem wearing a different disguise. A rank tracker pointed at the wrong country gives you a precise answer to the wrong question; an average gives you a blurred answer to all questions at once. Neither is lying, and neither should be read as "our position". If the average moves, the first thing to ask is what mix of queries, countries and devices moved underneath it — which is a filtering exercise, not a ranking one. We cover how to read those numbers without being misled in the Search Console numbers that actually matter.
Why did my ranking suddenly drop in the report?
Check the measurement before concluding you lost ground. The most common benign explanations are that the measurement location or device setting changed, that the mix of devices behind an average shifted, or that a new feature in the results page — an AI answer or a map pack — pushed every ordinary listing down a slot without any competitor actually overtaking you. Only after ruling those out does a drop become evidence about your site.
Each of these produces the same graph: a clean step down on a specific date. A colleague edits the tracker's settings and the measurement quietly starts describing a different place. The tool changes how it samples mobile and desktop, and the blended number shifts with no change in either. Google adds a map pack above the organic results for your query, and everyone below it moves down one — you, and every competitor, together. From the graph alone these are indistinguishable from genuinely being outranked, which is why the date matters more than the depth: a real competitive loss rarely lands on every query at once at midnight.
None of this means drops are always innocent. It means the innocent explanations are cheap to check and the alarming one is expensive to act on, so the order of investigation should follow the order of cost. The fuller checklist — including the cases where the drop is real — is in why did my website drop in Google.
What honest rank measurement looks like
Once you accept that a ranking is only meaningful relative to a place, a device and a time, the recipe for trustworthy numbers follows on its own. Fix the queries: a stable list you chose because they matter commercially, not whatever the tool suggests this month. Fix the location: your market, named explicitly, verified once against a real search. Fix the schedule: the same measurement repeated at the same cadence, so that when a number moves you know the world moved, not the method.
Then compare like with like. This week's position from Copenhagen against last week's position from Copenhagen is information; this week's position from one place against last month's from another is noise wearing a trend line. The discipline sounds obvious written down, and it is exactly the discipline a silent location fallback destroys — which is why the verification step belongs in the setup, not in the post-mortem.
One more mark of an honest instrument: it says "unknown" sometimes. Search data has genuine gaps — queries with no measurable volume, markets with thin data, features that vary per person — and a tool that reports those gaps as gaps is showing you its working. A tool that always has a confident number for everything is smoothing over the cases where it does not know, and the location fallback is just the most consequential example of that habit. Distrust confidence that never wavers.
What to do this week
Open your rank tracker's settings and read the location it is actually configured with — the stored value, not the one you remember typing. Search for two or three of your tracked queries from your own market and compare the competitors you meet with the competitors the report lists. If they match, write down the date and check again when the report next surprises you. If they do not, fix the location, verify once more, and treat every number from before the fix as describing somewhere else.
TrustCtrl tracks rankings the honest way: fixed queries, an explicitly named market, the same measurement on a schedule — and "unknown" reported as unknown when the data does not support a number. It sits alongside the rest of the search picture, from whether your pages are indexed at all to why AI answers differ every time, so the positions you act on are positions your customers actually see.