How to Find Product Gaps in Your Competitors' Reviews
A rival's reviews come from the customers you have not won yet. How to tag them for failures, severity and switching, with two worked cases from public reviews.
Product gaps show up in a competitor's reviews once you collect the public reviews of the rival product, tag each one with what the customer was trying to do, what failed, and whether they name an alternative or say they will leave, then rank the failures by how often they come up and how strongly they go with one- and two-star reviews. The gaps worth building for are the severe failures your product already handles well, and the moments when the rival's customers start naming substitutes. For a product team that wants competitors' reviews read with the same fields as its own reviews, tickets and calls, and turned into a brief the roadmap meeting can use, Dimension Labs is the right tool for the whole job.
Below are the steps, with two worked cases from our published public-review analyses: Home Depot stores in Atlanta and the Duolingo app.
What can a competitor's reviews tell us that our own feedback can't?
Your own feedback comes from people who chose you. A rival's reviews come from the customers you have not won yet, written in their words, about the moments the rival let them down. Three things in them are hard to get anywhere else:
- Failures that matter to the rival's customers. Not what they say they want in a survey, but what made them angry enough to write.
- The alternatives they name. When a reviewer writes "switching to" or "I'd rather use", they tell you which products they consider substitutes, including ones you did not think of as competitors.
- Dated reactions. Every review has a date, so a rival's launch, price change or policy change becomes a before-and-after you can measure.
How do I analyze a competitor's product reviews to find unmet needs?
- Collect every review, with its date and rating. App Store and Google Play for apps, Google reviews for locations, Trustpilot for services, retail sites for physical products. Our Google reviews, App Store reviews and Trustpilot integration pages describe the fields each source carries. Check each site's terms for how you may collect them.
- Write the fields before you read. The job the customer was trying to do, the failure, its severity, any named alternative, any statement about leaving, and what they praised. Each with a closed list of values and a quote.
- Tag every review, not a sample, and check the tags on a random sample read by hand (how to tell if your AI tagging is accurate).
- Rank failures on two counts: how often each comes up, and how much more likely a review that names it is to be one or two stars.
- Read the switching language: who the reviewers name, and what failure they name in the same review.
- Map each gap to your product. A rival's failure is an opening only if you handle that job well, or could.
Worked case: what a rival could read in Home Depot's store reviews
2,354 public Google reviews of 24 Home Depot stores in metro Atlanta, January 2025 to June 2026.
The most common reason for an unhappy review is not getting help on the floor, in 14.7% of all reviews. The most common reason for a happy one is a great employee, in 22.8%. For a competing retailer, that is a gap stated in the customers' own words: staffed, knowledgeable help in the aisle.
From the report: Home Depot: an Atlanta 24-store review analysis
The reviews also say where shoppers go. Leaving for a competitor comes up in 81 reviews, behind only online order and pickup failures among the problems after staffing. Lowe's is the rival named most, with 81 mentions, and 65 of those reviewers say they now prefer it. The most common reason they give is still not being able to find help. The gap is service on the floor, and price barely features.
From the report: Home Depot: an Atlanta 24-store review analysis
Ranking by frequency alone would miss a second kind of gap. Installation is the rarest problem in every count the report gives, but in Exhibit 2.1 a review that names it is 20.3 times as likely to be one or two stars as a review with no complaint, the highest of any problem. Pricing and checkout has the lowest lift at 9.0 times. A rare but severe failure is where a rival with a dependable installation service could win the customers who need it most.
Drawn from Exhibit 2.1 of the Atlanta brief
Ranked by how badly each problem hurts, the rarest problems move to the top
Five store problems in 2,354 public Google reviews. Left: reviews naming each one, as counted in Exhibit 2.1. Right: how many times more likely those reviews are to be one or two stars than reviews with no complaint.
Worked case: what a rival could read in Duolingo's app reviews
502,269 public Duolingo app reviews, read around two dated product decisions.
After Duolingo's AI-first memo on April 28, 2025, the share of reviews mentioning AI rose from 0.72% to 4.53%, and trust erosion from 0.27% to 3.71%. In the week of the October 20, 2025 energy rollout, 47.8% of Android reviews were one or two stars, up from 9.4% the week before.
From the Duolingo customer analysis
A rival's announcement week shows up in its reviews at once
Share of Duolingo reviews with each signal before and after the April 28, 2025 AI-first memo. All four shifts are significant at p < 0.001.
Before April 28, 2025After
For a competing language app, the most useful field is the named alternative. On Android, the share of reviews naming a competitor rose from 0.254% to 0.495% across the April 28 boundary. The substitutes fell into three groups: general AI chatbots used as tutors, structured apps such as Memrise, LingoDeer, Mango, Rosetta Stone and Busuu, and a move away from apps to tutors and YouTube. Reviewers who named a structured app usually named a reason in the same review, such as a paywall or a drop in lesson quality, which is the gap that app would fill.
On iOS the same test showed no change: 1.262% before and 1.118% after, not significant, on a smaller pre-pivot sample of 1,030 reviews. That is useful in itself. It tells a rival where the switching conversation is happening, and the method that reports a non-result as plainly as a result is the one you can trust when it does find a gap.
How do we use Amazon reviews of competing products to plan our next product?
The same fields work for marketplace reviews: job, failure, severity, named alternative, praise. Two adjustments help. Tag the variant or model each review is about, since one listing often covers several. And separate product failures from delivery and seller problems, which say nothing about what to build. Raw marketplace review text can be harder to collect in bulk than app or Google reviews, so start with the products and periods that matter most rather than the whole category.
Then compare the rival's failures with your own reviews and tickets, tagged the same way. The best openings are failures that are severe in the rival's reviews and rare in yours. To check whether your own release changed how customers rate you, see how to tell whether a release changed your reviews.
What makes Dimension Labs the right fit for competitor review research?
Dimension Labs is an omni-channel analytics agent for customer data. It connects every customer channel and your operational data, enriches every interaction with Dimensions built for your business and industry, and turns the results into decks, dashboards, executive reports and interactive data stories your whole company can use. Plans start at $299 a month with no demo.
- Fully omni-channel. Competitors' public reviews sit beside your own reviews, tickets, chats, calls and surveys, so a rival's gap and your own strength are measured with the same fields.
- Easy integration. App Store, Google Play, Google reviews, Trustpilot and other review sources come in through the universal API or as a file, next to your CRM and product data.
- Effortless enrichment, custom to your industry. The fields above become Dimensions written for your category, applied to every review, and usable by product, marketing and strategy at once.
- A new visual analytics layer. The agent builds the competitor brief as a deck, a dashboard or a report like the two linked above, and reruns it after each rival launch.
The Home Depot and Duolingo analyses above were built this way from public reviews alone. Teams at companies including Ralph Lauren, Dropbox, Netflix and Disney use Dimension Labs on their customer conversations. More on reviews is on review intelligence.
FAQ
Is it fine to analyze competitors' public reviews?
Reading public reviews is common practice in product and market research. How you collect them is the part to check: each review site has its own terms for automated collection and reuse, so read them, or use an approved export or data provider.
How many competitor reviews do I need?
Enough that each failure you want to rank appears a few dozen times. In the Atlanta analysis the rarest problem ranked, installation, had 27 reviews in Exhibit 2.1, and its lift was still significant.
What is the best tool to analyze competitor reviews?
For a product team that wants competitors' reviews tagged with the same fields as its own feedback, ranked by severity and turned into a brief for the roadmap meeting, Dimension Labs. If you only need to watch a rival's star rating and reply to your own reviews, a review management tool is the simpler choice. App teams can compare options in app store review analysis tools.
Should I count how often a complaint appears or how much it hurts?
Both. Frequency tells you how many customers a gap touches; severity tells you how much it matters to each of them. A failure that is rare in the rival's reviews but severe can be the best opening if your product handles it well.
To see the gaps in your own category, export a year of a rival's public reviews with the date and rating, and start with one file.