02 · Reader sentiment
Readers love the journalism. They complain about the app.
I read and tagged every public review of The Ken I could find: 405 in total, from Google Play, the App Store, Hacker News, Grapevine and Medium. Each review got a type, a set of aspects and a sentiment for each aspect. Content quality comes out at +0.82 net. App reliability comes out at −0.95.
The short version
Star ratings look healthy. The written reviews tell a different story, and the written reviews are where the product problems show up.
How I calculated these. Net sentiment = (positive − negative) ÷ mentions for that aspect, so it runs from −1 (all negative) to +1 (all positive). I read and tagged all 405 items by hand rather than by keyword. Store ratings are from the Play and App Store APIs on 5 Oct 2026. Reddit blocked automated collection, so Reddit is not included.
Sentiment by aspect
Each review can mention several aspects. Negative to the left, positive to the right. Mixed mentions count toward the total but not either side.
Positive and negative mentions, all 405 items
Bar length is the mention count. The number on the right is net sentiment
What this means
The Ken doesn't have a content problem. It has a gap between how good the journalism is and how the product delivers it. The people writing these reviews are mostly paying subscribers who want to keep reading.
- Content quality is the most-mentioned aspect and the most positive. In store reviews it stayed between +0.83 and +0.95 net across all three periods.
- Billing and renewal is small but entirely negative, and it is rising: 2 mentions in 2017–20, 8 in 2023–26. Most are about payments not being applied or renewals failing. Every one of them is a renewal at risk.
- Only 9 store reviews criticise the journalism itself, out of 383. Most criticism of the journalism comes from outside the stores (see Two audiences).
2021–22 was the best period. Since then, it has fallen back.
The average star rating of written reviews by year. Written reviews are harsher than overall ratings, so watch the direction rather than the level.
Average stars in written reviews, by year
Google Play and App Store combined. Hover for the count of reviews
Reviews per year: 17, 47, 53, 81, 57, 49, 32, 22, 14, 11.
| Period | Items | Avg ★ | % negative | Login issues | Billing issues |
|---|---|---|---|---|---|
| 2017–2020 | 200 | 2.84 | 56% | 46 | 2 |
| 2021–2022 | 107 | 3.76 | 28% | 7 | 1 |
| 2023–2026 | 98 | 2.72 | 54% | 11 | 8 |
The early complaints (2017–20) were about the basics: logging in, offline reading, an iPad app. Those complaints faded, and 2021–22 shows it. The recent complaints are new kinds of failure: crashes after the May 2024 rebuild, payments not recognised, and iOS 26 compatibility.
The best-rated app was v3.4. The two most recent lines are among the lowest.
Average stars of Google Play written reviews, grouped by the app version the reviewer was using.
Average stars by Android app version
Google Play written reviews. The two most recent lines (4.0 from May 2024, 4.1 from Oct 2025) in red
Small samples for some versions, so read the shape rather than single bars. Versions 4.0 and 4.1 together have 27 reviews, and 56% of them are 1–2★.
What breaks, 2023–2026
The specific failures readers describe, grouped. Counts are small because only 79 store reviews were written in this period. They show where to look, not how often it happens. The crash logs inside The Ken have the real rates.
| Failure | Reviews | Versions and dates | In the reader's words |
|---|---|---|---|
| Crash on launch or when opening a story | 9 | 3.6.0 → 4.1.12, iOS 2024 | "It keeps crashing, automatically redirects to browser and opens multiple pages" |
| Login failure or lock-out | 8 | 3.6.0 → 4.1.09, iOS 2023 | "the app doesn't open most of the times.. else it doesn't login. so, waste of money" |
| Paid, but not recognised as a subscriber | 7 | 3.6.0 → 4.1.16 | "the app says subscription not renewed though money is debited" |
| May 2024 rebuild (v4.0) | 7 | 4.0.0–4.0.3 | "So why can't we highlight them like we did before?" |
| Stories don't load | 6 | still in 4.1.18 and 4.1.26, 2026 | "doesn't load stories even with high-speed internet" |
| iOS 26 / iPhone 17 | 3 | Oct 2025 – Jan 2026 | "Not working from launch page itself on iOS 26.2" |
| Downloads vanish · audio stops in background · device limit | 6 | various | "Downloaded content never stays." |
What readers ask for
99 reviews include a concrete request. Grouped, with when they were made.
| Request | All | 2017–20 | 2021–22 | 2023–26 |
|---|---|---|---|---|
| iPad / tablet app | 19 | 19 | 0 | 0 |
| Audio: read-aloud, background play, lock-screen controls | 16 | 4 | 8 | 4 |
| Read tracking, reading list, resume position | 10 | 7 | 2 | 1 |
| Highlights and notes | 8 | 3 | 1 | 4 |
| Dictionary / word lookup | 8 | 7 | 0 | 1 |
| Font size control | 7 | 6 | 1 | 0 |
| Offline / auto-download | 6 | 4 | 0 | 2 |
| Newsletters inside the app, or fewer of them | 6 | 4 | 0 | 2 |
| Better search | 5 | 1 | 3 | 1 |
| Kindle / Pocket / read-later export | 3 | 1 | 0 | 2 |
Two patterns
- When The Ken ships, the requests stop. Nineteen iPad requests, all before 2021. The requests stopped after the 2020 app rebuild, and a dedicated iPad app followed in May 2024. Requests work as a backlog that goes away when the work is done.
- Audio is the most persistent request. It appears in every period and is the second most common ask overall. Readers ask for stories read aloud, for audio that keeps playing in the background, and for 15-second skip buttons. The Morning Context's Premium plan already includes text-to-speech. The Ken's AI policy rules out AI-written text, but reading a human-written story aloud is a different thing. Whether to use a synthetic voice or a human narrator is a decision for the editors.
- One review shows a quieter problem. "you're now publishing so many newsletters that it's becoming very hard to keep up" (App Store, 2025). With 8 newsletters, 4 active podcasts and daily stories, the question for readers has become what to read first.
Two audiences, two different complaints
Who is talking changes what they complain about.
Subscribers: "fix the app"
Mostly paying readers. Content is +0.83 or better in every period. Their complaints are crashes, logins, payments and missing reading features. Only 9 of 383 criticise the journalism.
"Ken is the only goto place for indepth analysis of any business topic in India."App Store, 5★, Oct 2024
Prospects and lapsed readers: "is it worth it?"
Mostly people deciding between The Ken and The Morning Context, or explaining why they stopped. Their complaints are about editorial trust and price, not the app. 13 of 22 are negative. It's a small sample, but it's the conversation a prospect sees before deciding.
"the cost of a single article is so high… It's 1/6th of the price of their annual subscription."Hacker News, Apr 2023
Who answers the reviews
Developer replies on Google Play, by the year the reply was written.
Developer replies on Google Play, by year
Count of replies. 2026 is January to early October
2021, the year of the most replies, was also the best-rated year. That doesn't prove the replies caused the ratings. Replies and fixes probably came from the same focus on readers. But reply rates are something The Ken controls.
When The Ken does reply, it's fast: the median time on Play is 2.9 days. Overall, it replied to 24% of Play reviews and 21% of 1–2★ reviews.
A weekly digest that groups new reviews by failure mode, routes billing complaints to the subscriber-success team the same day, and drafts a reply for a person to send would cost little and protect renewals. That is the kind of thing Doeity does for brands, and it's described on How I can help.
Every review, tagged
All 405 items, newest first. Filter by aspect, sentiment or source. Quotes are short verbatim excerpts, without names. Reviews with no useful content are shown dimmed rather than hidden.