What a Competitor's Reviews Tell You

A competitor sitting at 4.3 stars tells you almost nothing on its own. Two companies with identical scores can have entirely different problems, and the score will not distinguish them.

What carries information is the distribution and the trend. A 4.3 built from mostly fives with a hard cluster of ones describes a product that works well until a specific failure hits. A 4.3 built from a mass of threes and fours describes a product nobody hates and nobody loves, which is a positioning opportunity rather than a quality one.

Review velocity

The number of new reviews a company receives per month, tracked as a series rather than a total, which distinguishes real growth in customers from a one-off solicitation campaign.

Also known as: Review volume trend

Review velocity matters more than either the score or the distribution. Volume per month, and the ratio of negative to positive within each month, shows whether a competitor is improving, degrading or simply growing. A rising complaint rate at constant volume is the single most actionable pattern in this data.

One furious review about a refund is noise. Fifteen reviews across six months describing the same refund process is a decision. Companies fix the complaints that threaten revenue, so a complaint that survives half a year is either expensive to fix or invisible to whoever would fix it.

Group what you find into buckets: onboarding, billing, support responsiveness, reliability, a specific missing capability. Count the mentions per bucket. That count is a competitive map of their weaknesses, built entirely from primary sources, and it costs about an hour.

The most valuable finding is a complaint that your product does not produce. That is not a talking point for an ad, it is a reason customers switch, and it belongs in your sales conversations and your comparison content in the customers' own words rather than yours.

Five star reviews are usually dismissed as marketing noise, which is a mistake. They are the cheapest source of customer language you will find, and unlike a testimonial on a homepage nobody edited them for brand voice.

Collect the recurring phrases. If twenty reviewers praise the same thing in similar language, that is the job the product is hired for. A gap between what a company promotes and what its happy customers praise is one of the most exploitable findings in competitive research, because it means their own marketing is pointed at the wrong benefit.

Use that language in your own copy where it is true of you. This is not imitation, it is aligning to how buyers describe the problem, which is the most reliable way to make a value proposition land.

A new failure mode appearing in reviews often precedes churn by weeks and precedes any public acknowledgement by months. It is the only competitive signal in this guide that reliably arrives before the competitor themselves has framed it.

Sudden volume spikes need interpretation rather than celebration. A jump usually means a review request campaign, a platform migration, or an incident. Reading the content of the spike tells you which, and each one implies something different about their operations.

Splitting the output that way stops the analysis becoming a list of things to feel good about. A complaint bucket you share is not ammunition, it is a category-wide problem, and whoever solves it first gets to talk about it first.

Manually, the maintenance pass is monthly: open each competitor profile, read the newest negative reviews, update the bucket counts, note anything new. Half an hour per competitor, and the discipline is keeping the same buckets so the numbers stay comparable.

Reviews are also one of the easiest sources to automate, because the pages are public and structured. Oppira tracks Trustpilot reviews for the competitors you follow and surfaces changes in volume and sentiment alongside their social and ad activity, which puts the complaint trend next to the marketing push that may have caused it.