Home › Methodology

Methodology

Sentiment collection

Vendor sentiment scores are computed from 215 verified customer mentions, collected March-September 2026. Mentions are gathered from three sources: public reviews on major platforms, public forum posts (Reddit, niche communities), and direct submissions to this desk. Each mention is attributed to a user and a company; duplicates by the same user for the same company are counted once.

Vendor-fingerprint screening

Before scoring, we screen for vendor fingerprints: posting patterns (accounts reviewing only one category), review timing (reviews clustering around a vendor's launch or a competitor's outage), and account history (newly-created accounts reviewing only the tracked category). Flagged mentions are excluded from the score. This is the same check 92% of sophisticated buyers now run themselves, per the 2026 PR Tools Buyer Trust Report (N=412, PR Signal Research).

Scoring

Each mention is scored on a 5-point scale: 5 for strong endorsement with specifics, 4 for endorsement, 3 for neutral or mixed, 2 for a complaint, 1 for a strong complaint with consequences (money lost, domain damaged, time wasted). The mean is the sentiment score. We surface the three most common verified gripes alongside the score, because a number without specifics is useless.

What this is not

This is not a survey of the general public, a count of all reviews ever written, or a verdict on whether a vendor is legitimate. It is a scored sample of what buyers actually report in a defined window. Use it as one input alongside your own trial.

Corrections

Errors can be reported via the contact link on the about page. We publish corrections with dates.

Limitations, stated plainly

Our mention set skews toward engaged users: people motivated enough to review or complain. Silent satisfied users and silent churned users are both undercounted. The scoring window is March-September 2026; older history is invisible. And sentiment measures reported experience, not product capability: a tool can do more than its users credit, or less. Read the score as the center of a distribution, not a verdict.