DEV Community: yaaay

Dating app users complain about price more than any other app category

yaaay — Mon, 26 Oct 2026 15:54:53 +0000

Dating apps get the same bug reports as every other app: messages that don't load, profiles that vanish, photos that won't upload. But when we compared their reviews with those of 204 other apps, one thing stood out. Dating app users write about money far more than anyone else.

The numbers

We took the latest 300 App Store reviews for each of 10 dating apps: Tinder, Bumble, Hinge, OkCupid, Match, Grindr, Feeld, HER, Coffee Meets Bagel and Hily. That is 3,000 reviews from the US, UK, Canadian and Australian storefronts, fetched in October 2026.

An AI classifier tagged each review with zero or more of seven themes: bugs, pricing and subscription, feature requests, UX, performance, crashes and ads. 1,931 of the 3,000 reviews contained at least one complaint.

1,170 reviews complained about price or the subscription. That is:

39.0% of all 3,000 reviews
60.6% of the reviews with any complaint

Across all 214 apps we have reports for (67,350 reviews), the pricing figure is 15.9% of all reviews. Dating is the highest of every category with at least three apps. Budgeting apps come next at 30.9%, then food delivery at 29.1%.

Pricing was the most common complaint, or tied for it, in 9 of the 10 apps. The exception is Feeld, where bugs (134 reviews) edge out pricing (132).

App by app

App	Reviews that mention price or subscription (of 300)
Hily	150 (50.0%)
Match	145 (48.3%)
Coffee Meets Bagel	139 (46.3%)
Feeld	132 (44.0%)
HER	129 (43.0%)
Grindr	121 (40.3%)
Bumble	118 (39.3%)
Tinder	100 (33.3%)
OkCupid	83 (27.7%)
Hinge	53 (17.7%)

Hinge is the outlier. Its 300 reviews average 2.88 stars, against 1.92 for Tinder, 1.68 for Bumble and 1.45 for Match. Fewer of its reviewers are angry, and fewer of the angry ones are angry about money.

One caveat matters here. "Latest 300 reviews" covers very different time spans. Tinder's 300 start on 29 Aug 2026. Bumble's start on 23 May 2026. Match gets so few new App Store reviews that its 300 go back to December 2020. So treat the Match figure as several years of opinion, and the Tinder figure as the last two months.

What people actually say

Most of the price complaints are not "it costs too much" in the abstract. They fall into three patterns.

Paying and still getting nothing. This is the most common shape. The user paid, and the thing they paid for (matches, replies, visibility) didn't arrive.

"I was getting matches when I first joined now even with subscription and paying for boost there are no matches." (Tinder, 1 star, Canada)

"You dont get matches,you dont get messages,its expensive an not worth it" (Hily, 1 star, UK)

Free features moved behind the paywall. Long-time users notice when something they used to get for free now costs money, and they say so.

"Features that used to be free before are all now locked behind a paywall. The fact that i can only see 12 people for free is literally insane compared to a fee years ago" (Grindr, 1 star, US)

"I remember when this app actually felt usable without paying." (HER, 1 star, Canada)

"Makes no sense why the core features of this app are behind a paywall. But don't worry, they give you a couple free likes before asking you to pay up." (Hinge, 1 star, Canada)

The price itself, quoted. Some reviewers name the number.

"now you either commmit to months on end or pay 28$ a week- are you insane? This app is now ridiculously expensiv[e]" (Bumble, 1 star, Canada)

"This is expensive to use and not worth the prices at all." (Coffee Meets Bagel, 1 star, US)

Quotes are trimmed for length but otherwise left as written.

Why dating apps get this more than other apps

A few reasons fit the data, though we can't prove them from reviews alone.

First, the product is an outcome, not a tool. A budgeting app works the moment you open it. A dating app only "works" when another person replies. When that doesn't happen, a subscription feels like paying for nothing, and the review says exactly that.

Second, the pricing is layered. Subscriptions, weekly plans, boosts, super likes and "see who liked you" are all separate. Reviewers often mention two or three of them in one sentence, which tells you they're doing the maths and don't like the total.

Third, the free tier keeps shrinking. "Used to be free" shows up across several apps, not just one. Users who have been on an app for years compare it to the version they remember.

What a dating app team could do with this

These are suggestions drawn from the review text, not results we measured.

Separate "no matches" from "too expensive". Many price complaints are really about results: the user paid and nothing happened. Tag those apart from complaints about the price level, because the fixes differ.
Watch reviews after any free-tier change. "Used to be free" is a direct reaction to a change. If you trim the free tier, expect it in your next 300 reviews and count it.
Read the weekly-plan complaints carefully. When people quote a weekly price, the plan structure itself is the complaint, not just the level.
Study the outlier. Hinge sits in the same category with roughly a third of Match's price-complaint share. Its reviews are a cheap way to see which paywall choices users tolerate.

How we did it

Reviews: latest 300 public App Store reviews per app, from the US, UK, Canada and Australia, via Apple's public reviews feed. Fetched October 2026.
Classification: an AI model tagged each review with any of seven themes. A review can have several. "Complaint" means the review has a theme and is either rated 3 stars or lower or was marked as a serious issue.
Category comparison: 214 apps in our reports, 67,350 reviews. Only categories with 3 or more apps are compared.
Limits: App Store only, English-speaking storefronts, and the time windows differ by app (see above). Reviews come from people motivated enough to write one, so they lean negative.

We collect the patterns for the whole category in our guide to why users complain about dating apps.

Want the same breakdown for your app or a competitor's? Revunda's free check reads the latest reviews and sorts the 1 to 3 star ones into themes. No signup.

How to read your app's one-star reviews without fooling yourself

yaaay — Mon, 26 Oct 2026 15:14:31 +0000

Most teams read one-star reviews the same way: scroll, wince, remember the angriest three, and file a ticket. That gives you a list of anecdotes. It doesn't tell you which problem costs you the most stars.

This is the method we used to sort 29,785 one-star reviews from 274 big-brand apps into themes. It needs a spreadsheet or a short script. No AI model is required, and you can check every step.

Step 1: pick a fixed window and write it down

The App Store and Google Play show the newest reviews first. "The latest 300 reviews" sounds like a fixed sample, but it isn't. For a busy app it covers a few weeks. For a quiet one it can cover years: in our dating app reports, Tinder's latest 300 App Store reviews started in August 2026, while Match's went back to December 2020.

So choose a date range first, then take every review inside it. In our study that was 25 Oct 2025 to 24 Oct 2026. If you compare yourself with a competitor, use the same range for both.

Step 2: count shares, not reviews

A theme with 200 reviews in a big app and a theme with 20 in a small one can be the same problem. Divide by the number of one-star reviews in the window and compare percentages.

Set a floor, too. We only ranked an app if it had at least 30 one-star reviews in the year, and only reported a sector or country with at least 100. Below that, one bad week can swing the share by ten points.

Step 3: define each theme as a pattern you can read

We used one regular expression per theme, matched against the review title and text. Here are three of the ten, exactly as we ran them:

Login & account access: \blog ?in\b|\blogin|\bsign(?:ed|ing)? ?(?:in|on)\b|\bpassword|verification code|\botp\b|\b2fa\b|two[- ]factor|face ?id|touch ?id|fingerprint|locked out (shortened here; the full pattern is in the study)
Broke after an update: (?:since|after) (?:the |this |last |latest |recent |new |an? )?(?:update|upgrade)|(?:new|latest|recent|last) update
Ads: \bads\b|\badvert|\bcommercials?\b

A pattern is crude, and that is the point. Anyone on the team can read it, argue about it and change it. If someone says "we have a crash problem", you can count it in a minute instead of debating it.

A review can match several themes. A review that says "since the update I can't log in" counts for both, which is what you want.

Step 4: check a sample of matches by hand

Patterns catch things they shouldn't. Before trusting any number, pull a few random matches per theme and read them.

We checked 6 random matches for each of our 10 themes. 55 of the 60 were on topic. Two of the misses show the kind of error to expect:

"crashed" in a review about a taxi driver, not the app
"checkout" in a complaint about an upsell, not a failed payment

If more than one in five matches for a theme is wrong, tighten the pattern before you report it.

Step 5: look hard at what matched nothing

In our study, 59.4% of one-star reviews (17,699 of 29,785) matched none of the ten themes.

That is normal, and it is worth reading. Much of it is short ("useless", "worst app") and carries no signal. But some of it is a theme you haven't named yet. If you see the same complaint five times in the unmatched pile, add a pattern for it.

Also expect lower match rates outside English. Our patterns are English, so we only used English-language storefronts. Even then, Indian reviews often mix English and Hindi, which lowers the match rate there.

Step 6: split by store and by country

Averages hide the useful part. Three examples from the same data:

By country. Customer service was 21.3% of one-star reviews in India, against 8.0% in the US and 2.5% in Ireland. Login was 2.8% in India and 14.5% in Canada.
By store. 44.0% of App Store reviews in the sample were one-star, against 29.4% on Google Play. The top theme was login on both.
By sector. "Broke after an update" was 3.3% overall but 10.3% in smart home and security apps. Ads were 3.6% overall and 24.7% in news and weather.

Be careful with country comparisons. They mix in which apps people use. India's sample leans towards shopping apps, and shopping apps attract service complaints.

Step 7: separate app problems from business problems

The three biggest themes in our study were login (9.5%), customer service (9.1%) and charges, fees and refunds (8.4%). Crashes came fifth at 4.6%.

Only one of the top three is clearly an engineering problem. Service and billing complaints are about refunds, hold times and surprise charges. The app is just where people go to say it. Route them to the team that can fix them, or they will stay in your reviews no matter how stable the build is.

Step 8: rank by share, then read the top reviews in full

Now sort themes by share and read the 10 to 20 most recent reviews for the top two. This is where the anecdotes become useful, because you know they represent a real share of the pain. Write down the specific failure each review describes ("code never arrives", "Face ID stopped after update") and count those too.

Mistakes this method catches

"Crashes are our biggest problem." Often false. In our sample, login and service each beat crashes by about two to one.
"Reviews got worse this month." Check the window first. A quiet app's "latest 300" may span years.
"Our competitor gets the same complaints." Compare shares in the same date range, or the comparison means nothing.
"The AI summary says users want X." Maybe. A pattern count lets you check how many reviews actually say it.

What you need

An export of your reviews with date, rating, country and text. Both stores give developers their reviews through the console and an API. For competitors' iOS apps, Apple's public review feed works.
A list of 8 to 12 themes as patterns. You can start from the ten we published.
An hour for the first pass, then minutes each month after.

The full pattern list, the hand check and the per-app tables are in the one-star review study. Copy the patterns freely; the data is CC BY 4.0.

Short on time? Revunda's free check reads your app's latest reviews and sorts the 1 to 3 star ones into themes for you. No signup.

What 29,785 one-star app reviews complain about

yaaay — Mon, 26 Oct 2026 14:32:51 +0000

When a big app gets a one-star review, most teams guess the cause is a crash. We checked. Across a year of one-star reviews for 274 App Store and Google Play listings from 227 large brands, crashes came fifth. The top three were login trouble, customer service and money.

What we looked at

We pulled the newest public reviews for each listing on 24 Oct 2026: banks, insurers, carriers, airlines, delivery apps, retailers, streaming services, betting apps, health apps, smart home apps and government apps. 76,521 reviews were dated between 25 Oct 2025 and 24 Oct 2026. Of those, 29,785 one-star reviews came from English-language storefronts (US, UK, Canada, Australia, India, Ireland and New Zealand).

Each review's title and text was matched against one plain regular expression per theme. No AI model was involved, and the patterns are published with the study, so anyone can rerun them. A review can match several themes or none. A hand check of 60 random matches found 55 on topic. The misses were things like "crashed" in a story about a taxi driver.

The ranking

Theme	Share of one-star reviews	Reviews
Login & account access	9.5%	2,821
Customer service	9.1%	2,712
Charges, fees & refunds	8.4%	2,506
Slow, loading & lag	4.9%	1,470
Crashes & won't open	4.6%	1,373
Ads	3.6%	1,085
Broke after an update	3.3%	975
Redesign & confusing UI	3.2%	946
Notifications & alerts	2.0%	581
Payment or checkout fails	1.0%	307

59.4% of one-star reviews (17,699) matched none of the ten themes. Many of those are short ("useless", "worst app ever") or are about the company rather than the app.

Login is the top complaint, and it isn't close in some sectors

Login problems lead overall at 9.5%, but the average hides the spikes. In energy and government apps, 24.5% of one-star reviews are about getting in. For insurers it is 18.0%. For parcels and post, 14.1%. For banks and payments, 13.6%.

The apps with the highest login share, among those with at least 100 one-star reviews in the year:

myGov: 42.1% of 259 one-star reviews
My O2: 34.9%
BMO Mobile Banking: 31.0%
Royal Mail: 30.2%
State Farm: 30.0%
NHS App: 28.2%

These are apps people are often required to use. You can't switch tax offices because the app logs you out. So the frustration ends up in the review box. Verification codes that don't arrive, Face ID that stops working after an update, and forced password resets all show up in the text.

If your app sits behind an account, read your one-star reviews for the words "log in", "code" and "password" before you read them for "crash".

Customer service complaints are about the business, not the code

Customer service sits at 9.1% overall. These reviews rarely describe a bug. They describe a refund that never came, a chat bot that loops, or a phone line nobody answers. The app is just where people go to complain.

The country split is the most striking part of the whole dataset:

India: 21.3%
Canada: 8.4%
United States: 8.0%
United Kingdom: 6.3%
Australia: 4.4%
Ireland: 2.5%

Part of this is app mix. Shopping and groceries are the largest sector in the Indian sample, at 32.0% of its one-star reviews, and shopping apps attract service complaints. MakeMyTrip (44.2%), Nykaa (42.4%) and Flipkart (33.9%) lead the service ranking. But the gap is large enough that a team shipping to India should expect its review page to double as a support queue.

Money complaints cluster in tickets, taxes and streaming

Charges, fees and refunds come third at 8.4%. In taxis, transit and fuel apps the figure is 15.1%. In streaming, sport and tickets it is 13.9%. Among single apps, Fandango (35.9%) and TurboTax (35.7%) lead the apps with 100 or more one-star reviews.

These reviews tend to be specific. People quote the amount, the date and what they expected to pay. That makes them more useful than most one-star reviews: a pricing page or a checkout screen can be fixed, and the review often tells you which line confused the customer.

Updates hurt smart home and banking most

"Broke after an update" is only 3.3% overall. In smart home and security apps it is 10.3%, the highest of any sector. Banks and payments follow at 5.8%.

Blink Home Monitor is the extreme case: 34.3% of its 140 one-star reviews in the year blame an update. BMO (20.7%), Barclays on iOS (17.6%) and Ring (17.2%) follow. When a camera app or a bank app changes, people notice on the same day, because they open it for something that matters to them.

Ads only matter in a few places, but there they dominate

Ads are 3.6% of one-star reviews overall. In news and weather apps, the figure is 24.7%. Channel 4 (64.8% of 108 one-star reviews), FOX News (43.6%), AccuWeather (30.2%) and ITVX (30.0%) top the list. If you run an ad-funded app, the overall average says nothing about your risk.

iOS users leave more one-star reviews

44.0% of App Store reviews in the sample were one-star, against 29.4% on Google Play. The top theme was the same on both stores: login and account access. We would not read too much into the gap. The two stores sort and surface reviews differently, and our sample has far more App Store listings (227) than Google Play listings (47).

Limits

The apps are big consumer brands, not a random sample of all apps.
Regular expressions miss sarcasm, other languages and vague complaints. That is why 59.4% matched nothing.
Sector shares reflect which apps we picked in each sector.
Shares only count sectors and countries with at least 100 one-star reviews, and apps with at least 30.

What to do with this

If you run an app, the cheap test is to export your last few hundred one-star reviews and search them for the ten patterns in the study. If login or service beats crashes for you too, the fix may be in your account flow or your support team, not your crash logs.

The full tables, charts, CSV and the exact patterns are in the study on Revunda. The data is CC BY 4.0, so you can reuse the charts with a link.

If you want this for your own app, Revunda's free check reads your latest reviews and sorts the 1 to 3 star ones into themes. No signup.