Findings · 2026-09-10 · By VSNARY | Emmanuel Orta

A newspaper’s schema coordinates put it in China

One missing minus sign. The address in the same schema block names a US city; the longitude below it points at northern China, and a range check calls it valid.

On 10 September 2026 we read the structured data on the front page of a large regional US newspaper. Inside a LocalBusiness node describing the paper sits a GeoCoordinates block. The latitude is right. The longitude is the paper’s own city centroid with the minus sign missing. The postal address in the same node is correct. Plotted as written, the coordinate lands in northern China, thousands of kilometres from the newsroom the address describes.

We have not named the paper: we have not yet contacted it, and the point of this page is the defect class, not the publisher. Anyone can check their own site for it in a minute.

What the page declares #

The node is not stale markup left in a comment. It carries the newsroom street address, the postal code, the switchboard number, the newsroom email address and an opening-hours block. Everything in it is correct except the one field a human never reads. Coordinates are the only property in that block written purely for machines, and they are the only thing that is wrong.

That is the general shape of this kind of defect. Nobody proofreads a number that never renders. No page on the site shows the mistake. A reader cannot see it, an editor cannot see it, and analytics will never report it. The only party that reads it is a parser, and a parser has no opinion about whether an American newspaper belongs in Asia.

Where a coordinate like this comes from #

The pair is not a measurement of the building. It is the city centroid to four decimal places, the value templates, plugins and tutorials reach for when they need the city. We see the same centroid elsewhere in our own records: a tree-care site in that city declares exactly the same pair, correctly signed. The newspaper’s building sits a short drive from that centroid, not on it.

So the likely history is ordinary. Someone pasted a city centroid into a template field, and somewhere between the paste and the render the minus was lost: a spreadsheet column typed as text, a CMS field that strips leading characters, a copy that took the digits and not the sign. The value stayed plausible the whole way. It has four decimal places and a real latitude. It simply points at the wrong hemisphere.

Why a range check passes it #

Our own report passed this coordinate as clean when we found it, and that is the part worth publishing. The usual coordinate checks look at the shape of the number: is the latitude within plus or minus 90 and the longitude within plus or minus 180, is the pair sitting at 0,0, do two locations share one coordinate, are there enough decimal places to mean anything. A sign flip passes all of them, because the unsigned value is a perfectly legal longitude.

A rule that only knows the shape of a number cannot tell you the number is about the wrong place. Catching this needs a second fact to disagree with, and that fact was in the same node the whole time, in addressCountry and addressRegion.

Which coordinate checks catch a sign flip? #

Only the last check catches it, and it is the one you can do by eye. When we found this, our report did not yet compare the coordinate with the declared address; this case is why we want it to.

What it costs, stated carefully #

We are not going to claim this changes where the paper ranks. We cannot see inside anyone’s ranking system, and a newspaper of that size has far stronger location signals: a Google Business Profile, decades of citations, an address printed on the masthead. A sensible consumer weighing one declared coordinate against all of that will discard the coordinate.

What is checkable is narrower and still worth knowing. Anything that reads this node and trusts the coordinate over the address gets northern China. That includes the naive case, which is more common than it sounds: a script that wants a latitude and longitude and takes the pair the page hands it. A map that plots sites where they say they are will put this one in Asia, because that is what the page says.

How common is it? #

We found it by accident, while mapping every measured domain that declares its own coordinates. On 10 September 2026 that was eight of the 233 domains we had measured; most sites are not local businesses and correctly declare none. Seven of the eight sat where their address said. One sat in Asia, and it was the newspaper.

Eight records is far too few to quote a rate from, and we will not. The interesting part is not the frequency. It is that a defect this total can sit on a major publisher’s front page and produce no symptom anyone would notice.

How to check yours in about a minute #

1. Find the block. Open your front page, view source, and search for GeoCoordinates.
2. Read the sign. Anywhere in the Americas the longitude must start with a minus sign. South of the equator the latitude must too.
3. Compare with the address. Read the address in the same node and ask whether the two agree about which side of the planet you are on.
4. Ask where the pair came from. If it is your city’s centroid rather than your building, every business in your city is claiming the same point and none of them is distinguishable by it. Use the coordinate of your actual door.

If your site declares no coordinates at all, that is not a defect. A coordinate you cannot keep correct is worse than none, and an accurate postal address does more work. Our NAP consistency guide covers checking address, phone and coordinates together, and the LocalBusiness schema guide covers what to declare at all. The last time a defect in our own reading reached a report is written up too, and the dataset page shows how every finding rate we publish is computed.

What this post does not claim #

It does not claim the paper is hard to find, that its search performance is affected, or that anyone there is careless. It claims one thing, which any site owner can verify on their own pages in a browser: a coordinate can sit thousands of kilometres from the address published beside it, and a check that only tests the shape of the number will call it fine.

Questions this post answers

Why would a newspaper's coordinates point to China?

The longitude in its schema is the city centroid with the minus sign missing. Western-hemisphere longitudes are negative, so dropping the sign moves the point to the same longitude east of Greenwich, which for much of the western United States falls in northern China or Mongolia.

Does a schema validator catch a missing minus sign on a longitude?

Usually not. Validators and range checks confirm the value is a number between -180 and 180, and the unsigned value is still legal. Only a check that compares the coordinate with the declared country or region catches it.

How do I check my own GeoCoordinates?

View the source of your page, search for GeoCoordinates, and read the longitude. Anywhere in North or South America it must be negative. Then confirm the point matches the street address declared in the same node.

Is it a problem if my site declares no coordinates?

No. Coordinates are optional. An accurate postal address does more work than a coordinate you cannot keep correct, and a wrong coordinate is worse than none.

Should I use my city's centroid as my business coordinate?

No. A centroid is the same point for every business in the city, so it distinguishes nobody. If you declare coordinates, use the location of your actual building or entrance.

Does a wrong coordinate hurt search rankings?

Nobody outside a search engine can say that, and a large site has much stronger location signals. What is certain is that any tool or script that trusts the coordinate over the address will place the business in the wrong place.

Related findings

Comments

Comments are read before they appear. Nothing is published automatically, and no account is needed.

Writing about this? Facts, live figures and marks — every number on that page is dated and traceable to a scan.

All findings · The dataset · How the dataset works