How we review.

We do not use one hidden rubric for every product on the site. Each major roundup starts with a plain-language thesis about what “best” should mean in that category, then scores every pick against criteria we publish in the review itself—usually in a scorecard you can read row by row.

Start with a thesis

Before we assign numbers, we state what we think actually matters for daily use in that product type. That thesis is not a marketing slogan; it is the lens we use when we compare models and explain tradeoffs.

On our home blood pressure monitor roundup, for example, we frame the category around repeatability at home—not specs on a box:

The best blood pressure monitor is the one you can use correctly over and over again.

A glucose meter roundup might emphasize strip cost and logging; a pulse oximeter roundup might weight motion sensitivity and display legibility. The thesis can change from category to category because the failure modes do.

The scorecard is the methodology

On flagship reviews, the scorecard section is where we disclose how we ranked products. You should not have to guess what we measured. Each product gets:

Named categories chosen for that review (typically four to six rows, depending on what matters for the product type).
A score from 1 to 10 in each category, with one decimal place where needed.
A short note in each cell explaining why we assigned that score—not just a number in a table.
An overall score shown as the unweighted average of those category scores, rounded to one decimal. We do not apply secret weights after the fact.

If a category is not in the scorecard, it did not drive the ranking on that page. That is intentional: we would rather be narrow and defensible than pretend every device was judged on the same generic checklist.

Criteria change by product type

We do not believe a single global formula (for example, “accuracy 40%, usability 25%”) fits thermometers, cuffs, wearables, and CGM systems equally well. Features that deserve their own row on one roundup may be irrelevant on another.

For our 2026 home blood pressure monitor review, the published scorecard rows are:

Setup friction — wrapping, aligning, and app pairing before a trustworthy first reading.
Comfort & fit — cuff feel and whether regular use stays tolerable.
Readability — how easy the result is to read on the device itself.
Power convenience — batteries, charging, and how often power gets in the way.
Repeatable accuracy — how hard it is to mis-position the cuff badly enough that day-to-day numbers stop being comparable.

Another category might swap in rows such as app export, strip availability, or validated clinical status. When we add or rename a row, we update the scorecard on that page so the methodology and the ranking stay aligned.

Practical takeaway: read the scorecard on the review you are shopping from. That table, not this page alone, is the contract for how that specific ranking was built.

What goes into the scores

Category scores draw on hands-on use, manufacturer specifications where they are checkable, published validation or accuracy studies when they exist, and the practical friction we observe during normal home use. For blood pressure monitors with listing on the U.S. Blood Pressure Validated Device Listing (or equivalent registries), we treat that validation status as relevant context when we discuss accuracy—not as a substitute for whether a given buyer will use the device correctly every day.

Where independent validation is missing, incomplete, or not applicable, we say so in the notes rather than implying lab-grade certainty we do not have.

Product selection

For each category we cover, we build a candidate list from clinically relevant or widely used models: validated devices where registries exist, strong sellers at major retailers, and established manufacturers—not brands that paid for placement. Products enter the pool because readers are likely to consider them, not because we already know the winner.

Medical review

Reviews with practical health guidance are checked for clinical accuracy before publication. Reviewers flag incorrect claims about what a device measures, what those measurements mean, and how they should be interpreted. If a reviewer disagrees with a claim we wanted to make, the claim does not run.

Re-verification

We re-check flagship reviews when models are discontinued, recalled, materially repriced, or superseded by better evidence. Products can move down or off a list when the scorecard no longer reflects defensible buying advice. Where a page shows a last-verified date, that is when we last confirmed the ranking and scorecard still matched our current judgment.

What we do not do

We do not accept payment to review a product or to rank it higher.
We do not publish “sponsored” reviews disguised as editorial.
We do not hide material drawbacks on products we recommend.
We do not rank a product above competitors solely because of an affiliate relationship. When we have a commercial tie to a brand, our affiliate disclosure applies; the scorecard still has to support the placement.

How we update rankings

Rankings are not static. When we update a page, we re-read the thesis, the scorecard notes, the order of picks, and any FAQ answers so they still tell a coherent story. A score change should be traceable to a row in the scorecard, not a vague “feel” adjustment.

Corrections

Found a factual error in a score, note, or claim? Tell us. We will fix it and note the correction when it affects how a reader should interpret the review.