How to Review a Product You Do Not Own (Without Pretending You Do)

Every affiliate publisher hits this within their first month. You want to write “the best cordless drills for cabinet work”, there are nine worth covering, and buying nine drills costs more than the article will ever earn.

The advice you will find splits into two bad answers. One says buy everything, which does not scale past a hobby. The other quietly teaches you to write in the first person about products you have never touched — which is a deception, and which search engines have spent two years getting better at spotting.

The third answer is to do real research, be specific about what kind of research it was, and let that stand on its own. Done properly it produces a better article than most hands-on reviews, because one person using one drill for a week is a sample size of one.

The line: what you may and may not claim

This is not a grey area, and it is worth being precise because the penalty is not just an SEO problem.

The research method, step by step

This takes two to four hours per roundup and produces material that most hands-on reviewers never gather.

Not one per article — one per category. If you are writing about cordless drills, own a cordless drill. It does not have to be the one you recommend.

This single purchase changes everything downstream. It gives you the vocabulary, tells you which specifications actually matter in use, and lets you recognise when an owner review is describing a real problem versus a user error. Without it you are summarising language you do not understand, and it shows.

It also gives you one legitimate first-person thread to run through the article: “My own drill is the 18V version of this, and the thing I did not expect was how much the clutch settings matter for hinges.” That is true, useful, and clearly scoped to what you actually own.

Most people read the five-star reviews. The failures are where the information is.

• Sort by lowest rating, then read forty. You are looking for repetition, not outrage. One person with a broken unit is noise; fourteen people describing the same part failing is a design fault.
• Note when it fails. “Stopped working after eight months” appearing repeatedly is the single most useful fact you can give a reader, and it appears in no specification sheet.
• Separate product faults from expectation faults. “Too small” usually means the listing is misleading about size — which is still worth telling readers, but it is a different problem from a motor burning out.
• Check the three-star reviews. These are the most honest ones on any platform. People who liked it with reservations write the most useful prose.

Copying a spec table off the listing adds nothing — the reader can see that themselves. The value is in knowing which two numbers matter for their use case, and saying so.

That requires understanding the category, which is why step 1 exists. For a drill it might be clutch range rather than peak torque. For a monitor, panel type rather than refresh rate. For a mattress, edge support rather than thickness. Finding the one specification that everybody prints and nobody explains is the fastest way to write something better than the competition.

Two more things worth checking that almost nobody does: whether the manufacturer publishes the figure at all (silence is informative), and whether two models that look different are actually the same unit with a different battery.

• The manufacturer’s manual. Usually a public PDF. Maintenance requirements, replacement-part availability and warranty exclusions all live here and never appear on the product page.
• Support and parts. Can you still buy the consumable in two years? For anything with filters, blades or bags this is the whole ownership cost.
• Enthusiast forums and subreddits. The place people go when something breaks. Search the model number and read what comes back.
• Video reviews. Watch several, especially long-term follow-ups. You may not cite them as your own testing, but you can attribute: “several long-term reviewers report the coating wearing at the contact point.”
• Safety recalls and standards. A recall history is a fact worth surfacing and almost nobody checks.

State plainly how you reached the recommendation. This is what converts “review of a product I have not seen” into “research summary”, and it costs one paragraph.

Some publishers worry this admission costs them credibility. In practice it does the opposite — it is more confident than an unsourced opinion, it is checkable, and readers who have been burned by fake reviews recognise it immediately. It also puts a genuine differentiator on a page that would otherwise look like every other roundup.

Categories where you have to buy it

The method above has a hard limit. Some categories carry consequences that research cannot responsibly cover.

These are also the categories where owning one product genuinely differentiates you, because most competitors have not either. If you are choosing a niche partly on how sustainable the content is, this is worth factoring in — see the niche scoring guide.

Getting to genuine hands-on, affordably

The research method is a floor, not a ceiling. Every product you actually own makes the whole category’s content better. Practical ways to increase that without buying everything:

• Buy the one you recommend most. Your top pick appears in every article in the category; owning it pays for itself in credibility across all of them.
• Buy secondhand. A used unit tells you everything about build quality and wear that a new one cannot, often at a quarter of the price.
• Borrow deliberately. Ask readers, friends or a local group. People are generous about lending things they are proud of owning.
• Ask manufacturers once you have traffic. Review units are common. Disclose it — an undisclosed incentivised review is a serious problem in both FTC terms and, since July 2026, in Google’s review-snippet guidelines.
• Do not buy, use and return. It is retail fraud, it is obvious in the returns data, and it will eventually cost you the Amazon account you are trying to earn from.

Where you do have genuine experience, make it unmissable rather than burying it in paragraph nine — that is the substance of E-E-A-T for affiliate sites, and it is the difference between a page that reads as credible and one that reads as assembled.

Common questions