How We Test

How the Reviews desk tests skincare apps, tools, and tracking devices. Minimum four-week test windows, defined criteria, declared relationships, and honest negative reviews kept live.

What we test

The Reviews desk tests:

  • Skincare apps: routine builders, ingredient scanners, AI skin scanners, photo tracking, dupe finders, cycle-sync apps, condition trackers (acne, rosacea, eczema, vitiligo, melasma).
  • At-home tests: microbiome kits, DNA-based skincare tests, cortisol panels, hormone panels.
  • Devices: UV reminder apps, sleep-skin trackers, facial massage apps, observational scanners.
  • Tool reviews: ingredient checkers, label scanners, pregnancy-safe-skincare apps.

What we do not test

  • Prescription skincare. That belongs with a clinician, not a review desk.
  • In-office treatments (lasers, peels, microneedling, RF). We cover them editorially but we are not equipped to test them at consumer scale.
  • Anything that requires a clinical-grade study (blinded controls, n > 30, ethics-board approval) to evaluate honestly.
  • Anti-aging efficacy claims. They require blinded, photo-graded, long-window studies we cannot replicate.

How a test works

  1. Minimum 4-week test window. Anything shorter is impression, not review. For condition-tracking apps, we run 6-8 weeks to span at least one cycle of relevant skin behavior.
  2. Defined criteria. Before testing, the desk sets the evaluation rubric for that tool category. We publish the rubric on the review page so readers can weigh the same dimensions we did.
  3. Multiple testers where applicable. For apps that test on different skin types, we recruit 3-5 testers across skin types and ages.
  4. Privacy assessment. Every app review includes a privacy section: what data is collected, who it’s shared with, how to opt out, whether the privacy policy matches the app’s actual behavior.
  5. Verdict by use case. We don’t score on a 1-10 scale. “Is this good?” is almost always “for whom, for what.” The verdict section explains who the tool fits and who should pass.

Evaluation criteria

Every Reviews desk piece is graded on the same five dimensions, with sub-criteria specific to the tool category:

  1. Usability. First-week friction. Five-week friction. How often we forgot to use it. How often we wanted to.
  2. Scientific basis for claims. Does the app or device claim something the literature supports? If it claims to “balance hormones via UV exposure,” that’s a flag.
  3. Privacy posture. What’s collected, what’s shared, what’s monetized.
  4. Value relative to free alternatives. Most paid tools are competing with a Google Calendar reminder and a phone camera. We test against that baseline.
  5. Durability of value. Does the value hold across the test window, or does the novelty wear off in week 2?

Disclosure

  • Free units. When we receive a product or test kit free, we declare it in the review’s first paragraph.
  • Paid testing. When we pay retail for the unit, we declare that too.
  • Affiliate. The Journal does not currently use affiliate links. If we add them, every link will be marked, and existing reviews will not be retroactively monetized.
  • Brand relationships. If a writer has worked at a brand being reviewed (including formerly), the relationship appears in the byline.

Negative reviews stay up

Approximately half of our app and tool reviews end with “skip this one.” We keep them published. We do not quietly delete a bad review when a brand updates the product or asks. If a product genuinely improves, we re-test in 6-12 months and update the existing review with a clearly-marked “Re-tested” section — the original verdict stays visible.

Re-testing cadence

App reviews are re-tested annually because the underlying product changes. Test kits are re-tested when the methodology or panel changes (typically 12-24 months). Device reviews are re-tested when the firmware ships a meaningful update or the price moves significantly.

Questions

Have a tool you want us to review, or a question about our methodology? Email [email protected].