Fact-Checking Policy
Last updated: August 2026.
Who checks a piece before it publishes
The writer checks their own work first, going back through the draft claim by claim and writing down where each one came from. Not from memory. The draft reaches an editor with its sources attached, and a claim that cannot be traced to a document, a named person or a measurement is cut or reworded into what we know.
A second technical reader is compulsory on anything a reader could act on with tools in their hands: torque settings, fluid capacities, service intervals, tow ratings, payload arithmetic, any recall or defect claim, and any procedure. For a procedure that reader is usually somebody who does the job for a living, and the last thing before a maintenance piece publishes is a call to somebody with the vehicle on a hoist. Plenty of pieces stop here for a week, and some never restart.
Primary sources, citations, and what we cannot get
The order of preference is dull and it holds. A regulator’s recall entry beats a news report of it. A bulletin with its number and revision beats a photograph of a dealer’s printout, which beats somebody’s account of what the printout said. A compliance plate photographed on the vehicle beats a brochure figure, because the plate is what the vehicle was built and rated as.
Interviews with people who work on these vehicles sit alongside documents, not under them. A fleet technician who has replaced the same component eleven times knows something no press office has written down. Secondhand material is a lead only: another outlet’s report, a forum thread, an owners group post, a press release.
Where a source is online and likely to stay there, the article links to it, on the sentence making the claim rather than in a pile at the bottom. Much of what we rely on is not linkable, since manuals sit behind subscriptions and bulletins go to dealers, so we describe the document, give its number and revision, and say how we saw it. A claim with no source shown is one we failed to support.
One limit is worth stating. Isuzu Owners Club is an independent, unofficial publication, not affiliated with, endorsed by, sponsored by or connected to Isuzu Motors Limited, its subsidiaries, its dealers or its distributors, so we have no privileged access to build sheets or internal engineering material. We check with documents anybody can obtain and people who agree to talk to us, in the service of helping owners keep these vehicles running and find each other. None of it is official guidance. Isuzu, D-Max and MU-X are trademarks of their respective owners, used here descriptively to say which vehicle a piece is about, and we claim none of them. Where the manufacturer’s documentation contradicts us it outranks us and we correct, and warranty questions and recall registration belong with Isuzu or an authorised dealer.
Numbers, which is where this site gets caught out
Payload and towing arithmetic is our commonest source of error and the reason the site exists. The braked towing figure on the brochure and the gross combination mass in the handbook cannot both be spent at once, tare and kerb weight are defined differently market by market, and a bull bar, canopy and drawer system take their cut before anybody sits in the vehicle. So every sum shows its working: which plate ratings, which market, what we assumed for occupants and accessories, and what is left. Readers redo them, which is the point.
Failure rates get the same suspicion. A small fleet with three failures out of twelve vans is twelve vehicles, not a rate for the model, and we say so rather than minting a percentage quoted back at us for years. With studies we read the original rather than the summary: what was measured, on how many vehicles, whether the finding supports the claim hung on it, and who paid for it.
Photographs, video and forum posts
A photograph of a failed part proves that a part failed somewhere. It does not prove the model, the market, the year or that anything is widespread. Before an image is used as evidence we reverse image search it, look at whatever metadata survived the messaging app, and ask for the original file, not the screenshot. Then we ask for the dull extra frames: another angle, the whole vehicle, the compliance plate, and whether it had been modified, a question that has killed several dramatic pictures sent in good faith.
A screenshot of a forum post is not a source either. Before anything from a comment thread is cited we contact the person who wrote it and ask for what sits underneath: the job sheet, the invoice, the scan tool data. If they will talk on the record they are quoted by name, and our editorial standards set out when we allow otherwise. Counting posts is not a measurement, and a hundred people repeating each other is one source.
The same standard, whoever is talking
This subject is more political than it looks. Emissions rules, low-emission zones, fuel duty and diesel bans all produce numbers, and those numbers arrive from manufacturers, government agencies, environmental campaigners and industry bodies defending the sector. Every one is checked the same way: the figure, not the messenger.
Where a number cannot be checked we say who asserts it and that we could not verify it, which is duller than picking a side and more use to somebody buying a van. We take no money from any party, campaign or state body, and how the site pays for itself is on our ownership and funding page.
When the checking fails
Sometimes it does, and saying what happens then is the point here. The error is fixed in the article with a dated note, on the terms in our corrections policy, and anything a reader could act on comes out the same hour it is confirmed wrong. Reports go to one inbox, read by the editors, at support@isuzuownersclub.com.
Then somebody answers the second question: how it got through. That answer is logged with the correction, and where a check would have caught it, the check joins the list above. Two of the rules on this page arrived that way. We work to the principles behind the International Fact-Checking Network code, non-partisanship, transparency of sources and funding, open methodology and an open corrections policy, without claiming to be a signatory.
