India App Quality ReportQ4 2026 edition

The storefront bug rate is the share of the reviews Google Play shows on an app's page that describe the app malfunctioning. Across all 187 apps it is 10.7%. Banking, UPI, lending, investing and insurance apps together run at 15.5%, against 8.3% for everything else. Banking leads the large categories at 23.4%, and 10 of its 11 apps had at least one malfunction review on show. At the other end, quick commerce & grocery (1.6%) and shopping (3.6%) storefronts are nearly free of them: 81.9% of their complaints are about orders, delivery, refunds, fees or support.

Every review was coded for the one main problem it describes. Software malfunctions make up 30.8% of all complaints on these storefronts; the rest are about service, fees, support or delivery (53.4%) and design, ads or missing features (15.8%).

Crashes and freezes are most common on insurance storefronts (8.2% of reviews). Login and OTP failures peak in banking (7.1%), where device binding, MPINs and SIM checks add steps that can fail. Payment failures peak in travel & tickets (4.3%).

The 14 UPI and payment apps we read show a storefront bug rate of 15.0%. 7 of their 240 storefront reviews (2.9%) describe a failed payment; their most common malfunction is crashes and freezes (13 of 36 malfunction reviews). Per review, travel apps show failed payments more often (4.3%).

Most failed payments are reported where a UPI or card payment sits inside another app's booking, order or repayment flow. Of the 42 reviews describing a failed, stuck or double-debited payment, 34 are on apps other than UPI, payment and banking apps: 11 on travel apps and 6 on lending apps, where repayments or EMI autopay fail. A typical review: the money leaves the account, the app shows a failure or nothing, and the booking, order or repayment doesn't go through. Some of these failures start at the bank or the payment network rather than in the app; to the user, the app is where it broke.

The median rating across the 187 apps is 4.5★, and only 12 are rated below 4. Ratings compress quality into a narrow band near the top, which is why the storefront reviews add information the stars don't.

Ratings and storefront reviews tell different stories. Across the 158 large apps, the rank correlation between rating and storefront bug rate is −0.19 (Spearman), a weak relationship. 10 apps rated 4.5★ or higher have malfunction reviews in at least 1 in 5 of their storefront slots.

For apps with 1 crore or more installs we read up to 18 storefront reviews (three in each of six languages). These are the apps where the largest share of those reviews describe a malfunction.

At the other end, these large apps had no malfunction reviews on show in any of the six languages: IDFC FIRST Bank (4.9★), Kissht (4.9★), Moneyview (4.8★), ConfirmTkt (4.8★), AbhiBus (4.8★), Zepto (4.7★).

Google Play shows developer replies under reviews. 71.0% of malfunction reviews have one. 76.9% of those replies are boilerplate: sorry for the inconvenience, write to our support address, share your registered number. 21 asked the reviewer for the phone model, Android version, app version or a screenshot, 14 acknowledged a bug or promised a fix in an update, and 2 said the team could not reproduce it.

We read the “What's new” text of every app. For 70 of the 159 apps that show one, the latest note says only a version of “bug fixes and performance improvements”. Apps update often: the median app was updated 28 days before we read it, and 67.4% had shipped an update in the previous 30 days.

15 malfunction reviews (4.7%) say the problem started after an update; 6 of them are on banking apps. These are regressions: the previous version worked.

We read every app's storefront in English and Hindi, and apps with 1 crore+ installs in Bengali, Marathi, Telugu and Tamil as well. Hindi reviews are the most critical: 49.8% are negative against 32.2% for English, and much of that is about service, fees and support (31.8% of Hindi reviews against 18.0% of English). In Bengali, Marathi, Telugu and Tamil, the reviews Google shows are mostly short praise, and only 4.6% describe a malfunction. Within the same 162 apps read in all six languages, reviews in Indian languages describe malfunctions slightly less often than English reviews (8.8% against 11.1%).

Google Play lets people mark a review as helpful. These malfunction reviews, all posted in the last 12 months, collected the most helpful votes. Each is paraphrased in our own words.

What we read. Google Play shows a handful of reviews on every app's store page, picked by Google as most relevant; this is the first thing a person deciding whether to install an app reads. On 23 October 2026 we loaded the India store page (gl=in) of each app in this edition and recorded every review shown: up to three per language. Every app was read in English and Hindi. Apps with 1 crore (10 million) or more installs were also read in Bengali, Marathi, Telugu and Tamil. We also recorded the star rating, rating count, install band, last update date and the “What's new” text. 189 apps were planned; 187 were read successfully (Fi Money and Doubtnut are no longer listed on Google Play).

Which apps. The most-installed consumer apps serving India in 16 categories, chosen by install band and category leadership, including government services. The full list is in the dataset.

How reviews were coded. Each review was coded with an AI model against the written codebook below: its language, sentiment, the one main problem it describes, whether that problem is a software malfunction, whether it started after an update, any phone or Android version named, and the type of developer reply. Coders worked from the original text in its original language; the dataset keeps only the codes and a short paraphrase, never the review text or the reviewer's name. To check the coding, a second AI coder that had not seen the first coding re-read and coded 63 randomly sampled reviews from 14 apps, one per category. The two agreed on whether a review describes a malfunction 92.1% of the time (Cohen's κ = 0.78) and on the exact problem type 84.1% of the time. The second coder marked slightly more reviews as malfunctions (16 against 13), so the rates in this report lean conservative.

Storefront bug rate. Malfunction reviews divided by all storefront reviews read. A review counts as a malfunction when its main problem is a crash or freeze, a login or OTP failure, a payment failure, a feature not working, slowness, or wrong data. Complaints about fees, support, delivery, refunds, loan decisions or content are recorded but not counted as malfunctions. 95% ranges are Wilson intervals.

Limits. Storefront reviews are Google's selection, not a random sample of all reviews, and Google changes the selection over time. They lean toward long, detailed and frequently upvoted reviews, and some are years old: the median review was 92 days old, and the bug rate for reviews posted in the last 12 months was 11.7% against 8.5% for older ones. Large apps contribute up to 18 reviews and smaller apps up to 6, so per-app figures are small samples. Bengali, Marathi, Telugu and Tamil were read only for large apps, so as a check we recomputed every category using only English and Hindi reviews, which were read for every app: the same three categories come out on top. A review describes what a user experienced; we did not test the apps, and some failures (payments especially) start outside the app. Ratings and install bands are as displayed by Google Play.

Disclosure. 99tests sells app testing. Some companies in this dataset are or have been 99tests customers; every app was read and coded the same way.

Both files are free to use under CC BY 4.0: reuse them with credit to “99tests India App Quality Report” and a link to this page. The apps file has one row per app; the coded reviews file has one row per review with its codes and our paraphrase, without review text or reviewer names.