We publish long reviews in unrelated niches, score them against a standard anyone can read, and buy everything we test.
Niche Review Press started in a spare room in 2024, after both of us spent a year writing category guides for a site that measured success in published words. The reviews we filed there were fine. The ones that ran were shorter, faster, and built from manufacturer copy, because a guide that takes six weeks to test cannot compete with one that takes an afternoon to assemble.
So we built a press where the six weeks is the product. We pick small categories, we buy the things, we use them for a month or more, and we publish the scoring sheet beside the verdict. Nine issues in, that has meant covering subjects with nothing in common: CRM software, cargo bikes, bookkeeping services, microphones, espresso machines. The niche changes. The standard does not.
Reader subscriptions pay for roughly two thirds of what we do. The rest comes from our own savings and from a small annual licensing fee paid by three trade publications that reprint our comparison tables with attribution and no editorial input.
There are no affiliate links on this site. We have turned down sponsorship from four companies whose products we have reviewed, and we will list them by name if anyone asks at admin@nichereviewpress.com.
Testing cost us $21,400 last year. Every receipt is in the annual accounts we publish each January.
Ines spent eleven years implementing business software for companies between five and two hundred people, which is where the habit of breaking things on purpose comes from. They run our software and services testing, built the scoring model, and edit everything that runs here, including their own work after a week in a drawer.
Hollis came from ten years of audio engineering and a side habit of commuting by cargo bike in bad weather. They handle anything with a physical test rig: microphones, espresso machines, bikes, and the measurement bench in the corner that the rest of the room is built around.
No company has ever seen a review before it published. We do not send questions to manufacturers before a verdict is written, we do not accept loan units unless a product cannot be bought at retail, and when we do borrow something we say so in the byline and return it inside 30 days.
No advertiser, subscriber, or licensing partner has any say over what we cover or how it scores. If any of that changes, the change will be announced on this page with a date, before it affects a review.
Read the scoring methodologyEverything is bought at retail unless it cannot be. Loans are disclosed in the byline and returned within 30 days.
No review runs before a product has been used for a month in the conditions it is sold for. The exact number of days is printed in every byline.
Criteria and weights are fixed before testing starts and published with the piece. See the methodology.
A score can change after a significant product update. When it does, the old score, the new score, and the reason sit at the top of the piece with a date.
Factual errors are fixed at the top of the article, not in a footnote, and the original wording is quoted.
Roundups name the products we rejected and why. A shortlist without the rejections is a catalogue.