When people give a Quran app one or two stars, the reason they give most often is advertising, followed by paywalls on the main feature, then by problems with the translation and by bugs. We found this by reading about 14,400 low-star reviews of 18 Quran and Islamic apps on Google Play, along with thousands more from other places. Underneath the numbers sits a plain wish, which is that people want to open the Quran on their phone without being sold to, and they want the app to stay as good as it was on the day they installed it.
Why we read the bad reviews
A five-star review usually tells you that someone is happy, but it rarely tells you why. A one-star review tends to be different, because a person who is upset will often explain exactly what went wrong, sometimes after using the app every day for years.
We wanted to know why people leave Quran apps, because that is one of the most useful things a builder can learn before writing any code. When the same complaint appears across many apps from many makers, it points to a habit in the whole field, and a habit is something we can decide to avoid from the very beginning.
There was a second reason as well. The Quran is unlike any other content on a phone, and the people who open a Quran app are often doing so for worship, for memorisation or to teach their children. When that time is interrupted, the frustration runs deeper than it would with a game or a shopping app, and the reviews show it clearly.
How we gathered them
The main sample was about 14,400 low-star reviews from Google Play, taken from 18 Quran and Islamic apps; the exact count in that large sample was 14,393. Alongside it we looked at about 5,300 reviews of apps for memorisation and teaching, and at about 340 App Store reviews, which we copied by hand because there was no simpler way to collect them.
We also read public discussions on Reddit and Hacker News through their public archives. Four threads were read in full and about 30 more at the level of their titles, and we counted 469 titles that were about reciters in some way.
One point matters for everything that follows. The counts we give here come from keyword matching, which means a review was counted under a theme when it contained words linked to that theme. The method is good at showing which themes are large and which are small, but it cannot give fair percentages, so we do not quote any. Please read every number below as a way of ranking the themes against one another, and never as a precise measure of how many people feel a certain way.
We name none of the apps we studied. The purpose of the work was to learn, and it would not be fair to set one maker’s worst reviews beside another’s.
Ads came first
Advertising was the largest theme by a clear margin, with about 2,260 reviews spread across 12 different apps. Because it appeared in so many apps, we read it as a habit of the field, and we do not think it is the fault of one or two makers.
What made these reviews stand out was the setting in which the adverts appeared. Some reviewers described being interrupted while reading or listening to the Quran, and several said this felt wrong to them in a way that ordinary advertising elsewhere did not. Some had used an app for a long time and left only when the adverts became too frequent to bear.
For anyone building in this field, the lesson is a simple one. Advertising is a business decision, and in a Quran app it carries a cost that never appears in a spreadsheet, because it reaches into the time a person has set aside for worship.
Paywalls and features taken away
The second theme was paywalls on the main feature, with about 960 reviews across 10 apps. These were rarely complaints about paying in general. They were about discovering that the very thing the app had been installed for, the reason it was chosen in the first place, had been placed behind a payment.
Close beside this was a smaller theme with a clear message, which was updates that removed something, with about 290 reviews across 13 apps. A person learns where everything sits in an app and builds a daily habit around it, and then an update takes a feature away or moves it behind a payment. The anger in these reviews seemed to come more from a sense of broken trust than from the change itself.
We think these two themes belong together. Both describe an app that changed its terms after people had come to depend on it, and both can be avoided if a builder decides early on what the app is for and then keeps to that decision.
Translation and bugs
Translation problems came up in about 900 reviews, which was more than we expected before we began. People care deeply about understanding what they read, and a translation that is missing, hard to switch on or poorly displayed weighs heavily on them.
Bugs came up in about 650 reviews and covered the familiar range of problems found in any app, such as crashes, audio that stopped part of the way through, and settings that did not stay where they were set. These complaints matter, though they are the kind that careful testing can reduce, and they ranked below ads and paywalls.
A growing question: which reciter is this?
One finding stood out because it has grown every year we looked at. Requests along the lines of “which reciter is this?” were counted at 22 in 2021, 53 in 2024, 84 in 2025 and 113 so far in 2026.
People hear a recitation in a short clip, at a masjid or in a recording that a friend has shared, and they would like to know whose voice it is. These counts are small beside the review themes, so we read them only as a direction of travel, but a number that rises steadily in this way still tells us something about what people are searching for.
What we could not hear
Every study has edges, and these are ours. We could not reach Facebook groups or X, where much of the conversation about Quran apps takes place. Reviews written in Arabic, Urdu, Indonesian and Turkish were barely sampled, which means the picture we give leans towards English speakers, and the wider Ummah may well see some things differently.
We also found nothing on how teachers and students use voice notes, even though many Quran teachers work this way every week. The most useful next step would be to sit down with three or four teachers and ask them directly, and we hope to do that, إن شاء الله.
What it means for how we build
The research pointed towards keeping an app offline, with any online part kept separate and switched on only by the person who wants it. Our own apps work that way, and nothing in them was changed or decided because of the study. Our prayer times app, Miqa, which is live on Google Play, asks for no internet permission at all and has no accounts, no adverts, no analytics and no tracking. Mirtal, our recitation app, is in closed testing and was built on the same rule. What the study gave us was a clearer picture of why that rule matters to the people who use these apps.
Counting thousands of reviews by keyword also showed us how limited keyword counting is, so we built a tool that sorts reviews by their main complaint, and it is right to be open about how AI is involved in it. The review sorter uses Laya, a small open AI model by NandhaKishorM (Apache 2.0), which we fine-tuned on Quran app reviews. We labelled 899 reviews by hand and kept 301 of them back to test the model. In that test it was right 73% of the time against 47% for keyword counting, and it was right 92% of the time on the reviews where it was sure of itself, which was about six reviews in ten.
Our own pass mark was 75%, and the sorter fell short of it, so we keep it for this one job and send every review it is unsure about to Claude, a larger AI model, to read in full. It runs on our own computer, works offline and never prints the full text of a review.
The full study is on our research page about Quran app reviews, with its method, its counts and the conversations it could not reach, so that anyone building in this field can weigh the findings for themselves.