How Free Journaling Apps Actually Make Money

2026-10-26 · Yesterself

Every app you did not pay for is paid for by someone. That is not a conspiracy, it is arithmetic: servers, app store fees, and a developer's rent are real costs, and a download of zero dollars does not cover them. So the honest question is not whether a free journaling app makes money. It is which of the handful of available methods it uses, and whether that method requires anyone to look at what you wrote last night.

There are essentially four. Learning to tell them apart takes about ten minutes per app, and most of the evidence is sitting in plain sight on screens you have already scrolled past.

Model one: paying subscribers subsidize you

The most common arrangement in journaling software. A minority of users buy a premium tier, and that revenue covers the free tier for everyone else. The free users are not the product; they are the pool that a small percentage of future subscribers is drawn from, plus word of mouth.

This model has a structural property worth naming: the company's incentive is to make the free tier good enough to keep you and the paid tier valuable enough to convert some of you. Nothing in that chain requires reading your entries. In fact, reading them adds cost and legal exposure without adding revenue.

How you spot it from the app's own screens. There is a clearly priced subscription, usually annual and monthly, listed in the app store under in-app purchases. Somewhere in settings there is a "restore purchases" row. The upgrade screen names specific features you would gain rather than vague "unlock your potential" copy. And critically: there are no ads anywhere, including on the free tier.

A warning inside this model: watch the shape of the pricing. A subscription that is only offered weekly, or a trial that requires a card and converts in three days, is a sign the company's plan is churn rather than retention. That is a business problem, not a privacy problem, but it tells you something about how long the app intends to be a good citizen.

Model two: advertising

Advertising is not automatically sinister, but journaling advertising is a special case. The value of an ad depends on targeting, and the thing that would make an ad on a journaling screen unusually valuable is the content of the journal. That is the tension. Most ad-funded apps do not feed your prose into an ad auction — but the mechanism by which ads get sold means an advertising SDK sits inside the app, and those SDKs typically collect device identifiers, coarse location, session behavior, and sometimes an interest profile assembled elsewhere.

How you spot it. Obvious cases announce themselves with a banner. Less obvious cases: an in-app purchase literally named "Remove Ads," interstitials that appear after you save an entry, a "watch a video to unlock" flow, or a privacy label listing "Advertising Data" or "Identifiers" as collected and linked to you. Screenshots on the store page sometimes have a suspicious empty strip along the bottom edge where a banner was cropped out.

Model three: the data itself is the product

The rarest of the four in journaling, and the one people most fear. It comes in grades. At the mild end, an app aggregates anonymized usage patterns and sells or publishes them. In the middle, a "wellness" app shares behavioral signals with partners. At the far end, entry content is processed to build a profile that is monetized directly, or used as training material for a model.

The important insight is that this model is not usually hidden behind a lie. It is disclosed, in the terms, in language designed to be skimmed past: a license to "host, store, reproduce, modify, and create derivative works" broader than delivery requires; a partners list; a phrase like "we may share de-identified information."

How you spot it. Three specific screens. First, the sign-up screen — an app that demands an email address and password has created a permanent identity to hang a profile on. Second, the permissions requests: a journaling app asking for contacts, precise location, or advertising tracking is asking for something writing does not need. Third, the terms of service, specifically the license grant paragraph and the sharing paragraph. Our walkthrough on reading a journal app's privacy policy in ten minutes covers exactly which paragraphs to read and which to skip.

The strongest technical counter-evidence is encryption you hold the key to. If the server only ever receives ciphertext, there is nothing to profile, whatever a future business plan might wish for. That is a claim you can check rather than trust — see what makes a journal app end-to-end encrypted and whether the people who make your journal app can read it.

Model four: nobody is paying the bills yet

This one is easy to miss because it looks like model one with the price tag removed. An investor-funded app can run free for years on someone else's money, with no ads, no data sales, and genuinely good behavior — because the plan is growth first and revenue later.

The risk is not that the app is reading you today. It is that the eventual answer to "how will this make money" has not been chosen yet, and you will be inside the app when it is. Historically, the two most common resolutions are a sudden paywall on features you already depend on, and an acquisition, where the acquiring company inherits your archive under whatever terms its own lawyers prefer.

How you spot it. There is no paid tier at all, and no ads, and the app is polished and staffed. Look at the website footer for a funding announcement or a press page; look at the app store listing for a company name and search it. None of this means "avoid." It means keep a current export.

The test that outranks all four

Whatever model an app is in, one question sorts the trustworthy from the rest: can you leave with everything, right now, for free?

An app that exports your complete archive in an open format — plain text, Markdown, JSON — on the free tier, with no upgrade prompt in the way, has made a costly commitment. It cannot hold your writing to ransom later, because you already have a copy. An app that offers export only to subscribers, or only as a locked PDF, or "coming soon," has quietly told you that your archive is leverage.

Do the export on day one, not on the day you get worried. Then do it every few months. It takes two minutes and it turns every question on this page from an anxiety into a preference.

Where Yesterself sits

Yesterself is model one, deliberately and only. The free tier is meant to be usable forever; premium adds sync, voice, stewards, and the Season of Selves retrospective, at 29.99 a year or 4.99 a month after a 21-day trial that asks for no card. There is no advertising and nothing is used to train anything.

The structural part matters more than the promise. There is no account — no email, no password. Identity is twelve BIP39 words you hold. Entries are end-to-end encrypted, so the server stores ciphertext it cannot read; there is no profile to build because there is nothing legible to build it from. And export is free forever, in open formats, on the free tier. Yesterself is not released yet — the waiting list on the site is the only thing to join today.

Apply the four-model test to Yesterself the same way you would to anything else. That is the point of knowing it.

Common questions

Are free journaling apps safe?

Safety depends on the business model, not the price. A free app funded by paying subscribers has no reason to read your entries; one funded by advertising or data licensing does. The app's own screens usually tell you which it is.

Does a free app have to sell my data?

No. Plenty of free tiers exist because a minority of users pay for a premium tier, and the free tier costs very little to run. Free is not evidence of harvesting, and paid is not evidence of safety.

How can I tell if an app shows ads before installing it?

Look for an in-app purchase list that includes something like remove ads, check the screenshots for banner space at the bottom of the screen, and read the developer's own description for the word free supported by advertising.

What is the strongest single signal of a clean model?

That the app never asks for an account. If there is no email address and no password, there is no stable identifier to attach a profile to and no mailing list to rent.

What should I do if an app changes its model after I have been using it for years?

Export your archive first, in an open format, then decide. An app that makes export hard at exactly that moment has answered your question.