Analyze what customers say about a competitor

Each place people talk about a product attracts a different crowd, so Kite reads several before drawing conclusions. Trustpilot reviews often come from paying customers after something went well or badly, and they're full of comments about billing, cancellations, and support. Hacker News draws technical buyers who argue about pricing, reliability, and whether the product does what its homepage claims. App Store reviews come from people using the product on their phones. Reddit threads are where buyers ask which product to choose and explain why they left the one they had.

A theme that shows up in two or three of these places is much stronger evidence than one that appears in only one. If Trustpilot reviewers complain about slow support and Reddit users say the same thing when recommending alternatives, that complaint is real and widely felt.

Kite tells you which sources had something to say and which didn't. A competitor with no iPhone app has no App Store reviews, and a company nobody discusses on Hacker News is a finding in itself. Kite reports those gaps plainly rather than leaving you to guess why a source is missing.

Kite pulls the latest 200 Trustpilot reviews for the competitor's website, which is usually the largest single body of customer opinion about a company. For each review, it reads the date, the star rating, the title, the full text, whether Trustpilot marked it as verified, and whether the company replied. It also reads the company's overall Trustpilot rating and its total number of reviews.

Those details change what the reviews mean. The dates let Kite see whether a complaint is recent or fading. The replies show whether the competitor answers unhappy customers or ignores them, which is a useful fact if your own team responds quickly. When the 200 reviews are only part of a much larger total, Kite says so and tells you the date range they cover, so you know you're looking at the last few months rather than the company's whole history.

Two hundred recent reviews is also the answer to the question of how to summarize hundreds of Trustpilot reviews. You don't need to read thousands to find the patterns. A few hundred dated reviews, read carefully and grouped into themes, show you what customers are saying now.

Kite searches Hacker News for stories and comments that mention the competitor or link to its website. For each story, it sees the title, the points, and the number of comments, and it can tell when a story was the company's own launch post. Then Kite reads the full discussion under the most active threads, up to several hundred comments each, because the most revealing opinions are usually in the replies rather than the headline.

If the competitor has an iPhone app, Kite finds it on the Apple App Store by matching the app's listed website to the competitor's domain, so it never confuses the company with a different app of the same name. It reads the app's current rating, how many ratings it has, its latest version and release date, and the star ratings and titles of up to 50 of its most recent reviews. A sudden run of one-star reviews after a new version is often the clearest sign that an update broke something customers cared about.

For software companies with a public GitHub organization, Kite can also check how often the main project ships new releases. That adds context when Hacker News commenters say a product has stalled or is moving fast.

Reddit is where buyers ask strangers which product to pick, and the answers often name a competitor's weaknesses directly. A thread titled "Alternatives to [competitor]?" is a list of reasons people want to leave.

With a connected Reddit account, Kite searches across Reddit for posts that mention the competitor, such as questions that ask for alternatives or compare it with other products. It reads each post together with every comment under it, so it sees not only the question but also which answers people upvoted and agreed with. It can also check a commenter's public profile and karma, which helps it tell a long-standing member of a community from an account created yesterday.

Without a Reddit connection, Kite can still find Reddit threads through web search. Connecting an account lets it read whole threads and every comment rather than only the pages search turns up.

Once Kite has read everything, it groups what customers say into a small number of themes. Each theme is a specific thing people praise or complain about, such as "charged again after cancelling," "support takes days to reply," "the mobile app logs you out," or "setup is fast and the templates are good." Kite counts how many reviews and comments mention each theme, notes where they came from, and checks whether the theme is growing, steady, or fading over time.

For example, Kite might find that 38 of a competitor's latest 200 Trustpilot reviews mention unexpected charges, that most of those were written in the last three months, and that the company replied to almost none of them. That's a much more useful finding than an average rating of 3.4 stars.

Kite also separates complaints about the product from complaints about the company. A buyer who says a feature is missing has a different problem from one who says they couldn't reach anyone to cancel. Both matter, but they lead to different marketing.

Every theme comes with two or three real quotes and a link to where each one was posted, including at least one critical and one positive voice when both exist. You can check any claim yourself, and Kite marks clearly which findings come straight from the reviews and which are its own interpretation.

A competitor's weakness only helps you if your product is genuinely better at that thing. Kite compares each complaint theme with what it knows about your product, your pricing, and your customers, then sorts the themes into three groups: weaknesses you can credibly claim to fix, weaknesses you share, and weaknesses that don't matter to your buyers.

The first group becomes work. If customers keep complaining that the competitor's pricing is confusing and your pricing is simple, Kite can propose a line for your homepage, a section for your comparison page, or an ad aimed at people searching for alternatives to that competitor. Kite can also write a "how we win and how we lose" page for each competitor so your sales team knows which complaints to listen for on calls.

The praise matters too. When customers consistently love something the competitor does well, Kite tells you, because that's a strength you'll need to match or a fight you shouldn't pick.

Kite saves the findings in your team's knowledge base, next to its other research on that competitor. Later work, such as a comparison page or a new ad campaign, starts from what customers actually said rather than from guesses.

Customer opinion changes when a competitor raises prices, ships a bad update, or fixes an old problem. Kite can set up a standing watch for new public reviews of a competitor, or for new reviews on a particular site such as G2. It checks on a schedule, no more often than once an hour, and tells you when something new comes in, in the Slack thread where you asked or by email if your team doesn't use Slack.

You can also ask Kite to repeat the whole analysis on a schedule, such as once a quarter. Each run reads the latest reviews and compares the themes with the last one, so you can see which complaints are growing and which the competitor has fixed.