What People Actually Complain About in Period Apps (2026 Review Data)

Every “best period tracker” list ranks apps by features. Almost none of them look at what people say once they have actually lived with one for a year. The complaints are more useful than the feature grids, because they tell you where an app stops fitting — and that is the thing you cannot find out from a screenshot.

Where these numbers come from

The figures below come from an analysis published by Unstar.app, which categorised 1–3 star reviews across the six most-downloaded period apps: Flo, Clue, Stardust, Natural Cycles, Apple Health and Glow. It was published on 27 April 2026 and updated on 4 August 2026. The percentages are theirs. We have not re-run their analysis, and nothing in this article is our own research.

One caveat to hold onto before reading anything into these. Low-star reviews are not a survey: people who are quietly satisfied rarely write one, so this measures what upsets people enough to complain, not how common those experiences are across all users. That still makes it the best available picture of where these apps stop working for someone — which is exactly what you want when you are choosing.

Privacy and data sharing — 28%

The largest category by a wide margin, and the one where the complaint is most often about the business model rather than a bug. An app that is free to you is being funded somewhere, and reviewers have got noticeably better at asking where.

Unstar.app notes that distrust of Flo persists despite its Anonymous Mode, which is a genuinely interesting finding: the feature exists, and reviewers still do not believe it. Trust, once lost, is not recovered with a settings toggle. They also identify Apple Health as the privacy default on iOS — the only app in their analysis with no third-party data exposure and no separate account — and suggest that if post-Roe legal exposure is your main concern, Apple Health is the safest starting point, with Flo's Anonymous Mode and Clue's GDPR footing as the next options.

What this means for choosing: privacy is not one question but three. Who holds the data, who could compel it, and what funds the app. An app can be strong on one and weak on another. What to check in a tracker's privacy policy covers how to answer those without being a lawyer.

Prediction accuracy — 19%

The second-largest category, and the one most often misread as an app being broken. Unstar.app singles out Clue for predictions dominated by mean cycle length — a detail worth understanding, because a mean is dragged around by a single unusual cycle in a way a median is not. One missed log, or one unusually long cycle, and the average moves for months afterwards.

Here the honest answer is uncomfortable. If your cycles are genuinely irregular, no app predicts them well, because there is nothing stable to predict from. That is not a failure of engineering; it is the condition of the data. An app can be more or less graceful about admitting it — saying “we are not sure” rather than showing a confident date — but none of them can fix it.

If that is you, the useful shift is from prediction to pattern. What actually causes irregular cycles and how to track a cycle that will not settle both cover getting value from logging when the forecast is never going to be reliable.

Subscriptions and paywalls — 16%

Unstar.app describes premium creep at Flo and a Plus paywall at Clue covering features that had previously been free, and notes that apps in this category have moved features to paid tiers steadily since 2023. They also flag Natural Cycles as premium-only, and Glow as running aggressive ads on its free tier.

It is worth separating two complaints that arrive in the same words. Charging for an app is not a betrayal — building and maintaining one costs money, and a subscription is a business model that does not involve your data. What reviewers object to, reading the pattern, is specifically retraction: a feature they had been using, now behind a paywall it was not behind last year. That is a different grievance from “this costs money”, and it is the one that produces the star rating.

Symptom and mood logging — 12%

The most concrete category of the six, and the one where complaints are least likely to be trade-offs. Unstar.app records limited symptom fields in Stardust and minimal symptom logging in Apple Health, alongside general frustration with logging interfaces across the category.

Apple Health is a fair illustration of why “limited” is not automatically “bad”. It is a general health platform with a period feature, not a dedicated cycle tracker; depth was never the point, and its strong privacy position is partly a consequence of that restraint. Stardust is a different case — symptom depth is a reasonable thing to expect from an app whose whole purpose is cycle tracking.

This is the category to test deliberately in your first week, because it compounds. Every day you log into a structure that does not fit your body is a day of data you cannot fully use later. What is actually worth logging every day works as a checklist to test any app against.

Notifications and wellness content — 10%

Notification volume is named specifically as a Flo theme in the analysis. The broader category covers both push frequency and in-app content reviewers did not ask for.

This one is almost entirely a growth-mechanics complaint rather than a health one. Notifications drive daily-active-user numbers, and those numbers drive valuations. An app that pings you about content is usually optimising for something other than your cycle.

Crashes and sync bugs — 8%

The smallest category, and the only one that is unambiguously a defect rather than a decision. Its size is the interesting part: these apps are, on the whole, technically reliable. The things that make people give a period app one star are mostly choices somebody made on purpose.

Which of these should actually change your choice

The distinction that helps when deciding is not how large a category is, but whether the complaint describes a failure or a trade-off someone made deliberately.

Four questions to ask before you commit

• What funds this app? If it is free with no subscription and no obvious sponsor, read the privacy policy for third-party sharing before you log anything.
• Does the symptom list have room for what I actually get? Test it in week one. The cheapest thing to check, and the most expensive to discover late.
• What does it say when it does not know? An app showing a confident date for an irregular cycle is telling you something about its honesty, not its accuracy.
• Can I get my data out? Cycle history is worth years. An app that cannot export it has made leaving expensive.

If you have already narrowed it down, we have written in more detail on Flo, Clue, Natural Cycles and Stardust.

Where we sit, honestly

We build Dawn Phase, so treat this as the interested party it is — and one section is all it deserves in an article about other people's reviews.

Dawn Phase is free, carries no ads, and does not sell or share cycle data with anyone, including research partners. Predictions use a median rather than a mean, and are withheld entirely once the data is too stale to stand behind. Full CSV export and a doctor-ready report are on the free tier. Those are direct answers to four of the six categories above, and we would rather say it plainly once than imply it in every section.

What we do not have is worth saying too. We are young and our user base is small, which is a real risk you are entitled to weigh — an app with millions of users is more likely to still exist in five years. There is no Apple Watch app. Reminders are email rather than push notifications. Offline support works in the browser and on Android, but not yet inside our iOS app. And we make no fertility or contraception claims of any kind: if you need a cleared contraceptive method, Natural Cycles is a different category of product and should be evaluated as one.