"Say a word nobody dares ask the meaning of": a French PM sold a daily-word app to the French with TikTok carousels

A Paris indie developer took the "daily word + lock-screen widget + subscription" formula that already works in English and moved it to French, teaching words French people themselves don't recognise. It has added over 7 thousand installs on Google Play in one week, and its French education chart rank is still climbing. The product itself has nothing new. What's worth studying is how it posts on TikTok, how it collects money, and whether the same approach works in another language.

A five-slide TikTok

On Aug 14 of 2026, the TikTok account @proustapp posted a five-slide carousel. Slide one is a dark oil painting of a brooding man, captioned: "Comment je me sens quand je place un mot que personne connaît et que personne ose demander ce que ça veut dire" (How I feel when I drop a word nobody knows and nobody dares ask what it means).

The next three slides are vintage rural-path illustrations, one word each: "oaristys" (a tender, intimate conversation between lovers), "Emparadiser" (to make someone feel as happy as in paradise), "Avoir le cœur en écharpe" (to be heartbroken). The last slide reads "Apprends avec l'application Proust !" with a mustachioed Proust portrait below.

By Nov 6 the post had 870k views and 72k likes (TikTok profile, 11-06). Two weeks later, on Aug 29, Proust's Android version went live on Google Play, and the same day the account posted another one that reached 640k views.

What it is

Proust is an "obscure word app" for native French speakers. Opening it gives you a feed of word cards: rare words, idioms, "why do we say this", each card covering etymology and a story rather than a dry definition. There are 18 categories, short quizzes, evening review reminders, chapter levels and mini-games, daily streaks, and a "word of the day" widget for the home and lock screen.

It's made by MKa Studio, registered in Paris's 15th arrondissement as a sole proprietorship, run by Mohamed-Amine Kazouini. His LinkedIn lists him as a product manager overseeing a dozen or so retail apps (from search summaries, not verified). The iOS version launched in Feb 2026 3, and the Play version came nearly 7 months later.

The model isn't original. The US app "Vocabulary" by Monkey Taps has done it since 2016: word of the day, widgets, streaks, a subscription paywall on first open. It has 6.86M installs on Play. The difference is that Monkey Taps teaches English vocabulary, and users in each country just see a translated interface. Proust teaches the French language to the French, takes its name from Proust, and its whole tone is "cultured, refined".

Where users come from

Posting TikToks on its own, three or four a day. @proustapp now has 32k followers and 170 posts, nearly all of them the five-slide carousels above: an opener meme of "I used a rare word and the room went silent", three word cards, and an app ad at the end. From late August to early September it posted almost daily, with a string of hits: 870k (08-14), 640k (08-29), 350k (09-09), 230k (09-19). The Play launch got its cold start from that wave.

It rides an existing content format on French TikTok: "Say these rare words and you're smarter than 90% of people". A personal account doing that, @francaisavecmalyk, has 450k followers, and in Oct 29 one of its "5 rare words" posts reached 1.07M views. Proust turned that format into an app, and every post ends with a download link.

After October, Instagram ads took over. From October the account slowed to a post every few days, mostly getting a few thousand to twenty thousand views. One October post still reached 13k, and the November one only got 110 5 399. Organic reach clearly weakened, but installs kept climbing: Play total installs went 10-29 to 2,968,10-31 to 5,089,11-03 to 8,922,11-05 to 10,256 (Google Play install counts, 11-05). The App Store France rating count went from 10-28 to 11-06, roughly 10 new ratings a day, from 274 to 368.

Paid ads most likely filled the gap. An iOS review reads "c'est payant pour mettre le mot du jour sur l'écran verrouillé comme sur leur pub Insta" (you have to pay to get the word of the day on your lock screen, like in their Insta ad, 10-28), and another in English says "The ad is presenting that app as free and it is not" (10-22). How much was spent, what the creative looked like, and when buying started are all invisible from outside, so this is inference.

Doing it again in another language. On iOS there's already a Brazilian Portuguese version, "Proust: Palavra do Dia" (11-06 has 38 ratings), and an Italian one, "Proust: Vocabolario italiano" (which ranked #82 on Italy's free education chart). The Play update notes from Nov 4 say "Proust now speaks Italian". The numbers are still small, but the direction is clear: once French worked, the template got copied to other languages.

Media, X, Reddit and indie developer circles have basically not discussed it.

Where the money comes from: lock the card, then the trial

The Play listing says "The app is free, and will always be free", but in practice a trial pops up on install, 3 or 7 days, and you have to enter a card. The subscription runs between $7.99 and $18.99 (Google Play in-app price range).

Users are very clear about this. Since October, almost every one-star review on the App Store France storefront is about it:

"Pas de formule gratuite, obligé de rentrer sa carte." (No free tier, you have to enter your card. 10-21) "Même si c'est sans engagement on est obligés de mettre la carte pour faire l'essai à 0€." (Even if it's no commitment, you still have to put a card down for the €0 trial. 10-22) "Les mots sont vraiment intéressants, le concept est top. Seulement j'aurai vraiment aimé qu'on puisse mettre les widgets sans avoir besoin de payer." (The words are really interesting, the concept is great. I just wish you could use the widgets without paying. Samsung, 10-23)

That last review spells out the design. The ads show the lock-screen widget, and the widget happens to sit behind the paywall. Users are pulled in by what they want most from the ad and then asked to start a trial on first open. Some forget to cancel when the trial ends and become paying users. This is the standard Monkey Taps-style playbook, and it costs reputation: Play rating 4.3 stars (541 reviews, 11-06), iOS France 4.58 stars (372 ratings), and the negative reviews are almost all about the paywall.

The result: on the French Education Top Free chart, the app fell from #10 on 10-29 to #28 on 11-06, while on the Top Grossing chart it climbed from #38 on 11-01 to #32 (Google Play charts). Downloads are shrinking while revenue is growing. No one has published how much it earns, and the founder has never shared revenue figures.

The praise centres on the content: "Très bonne appli, tout est soigné" (Very good app, everything's polished, 08-25). An early user in February said "J'aime bien mon petit widget pour voir du vocabulaire que je ne connais pas" (I like my little widget for seeing words I don't know).

Who it's competing with

The last line is the most telling. From June to November 2026, at least seven or eight similar small apps appeared in France. They're decent, and each has single-digit ratings. Valoria also opened a TikTok account, with only 10 followers. Same genre, same timing, and only Proust got traction. The difference isn't in the app, it's on TikTok.

It's similar in other languages. Searching the App Store in Nov 6: the German store for "seltene Wörter" and "Wort des Tages", the Spanish store for "palabra del día", the Polish store for "słowo dnia", the Italian store for "parole rare", all have new similar apps this year, with rating counts mostly between 0 and 30. The top results in each store are Monkey Taps translated versions (2,291 ratings in Germany, 1,357 in Spain) and a few old dictionary apps. The spots in each country are still open, and the people who want them are already going for them, but nobody has built the content account.

If it were you

B. Worth a try: the playbook is clear and cheap, but this is a content business, not an app business. Building an app takes a few weeks, and seven or eight people in French already proved that alone doesn't work. Proust won because it can make TikTok carousels, put the widget behind the paywall, and later add ads. How much it earns is unknown. And the ceiling of the French market alone is limited, which is why it is already expanding into Brazil and Italy.

Who it suits: speakers of languages like German, Spanish, Polish or Dutch, ideally people who already read widely and enjoy etymology, who can write word cards that native speakers find stylish rather than translated. You need to be willing to post three or four carousels a day for two months. Experience running Meta ads is a bonus.

Who it doesn't suit: anyone wanting to do English (Monkey Taps and a crowd of older products already hold that space); people who can code but don't want to make content daily; anyone who can't live with one-star reviews from the card-on-open setup but still wants to make money from it (without a hard paywall, this kind of app rarely earns much; with one, you have to absorb the one-star reviews).

How to build it: follow Proust's structure: scrolling word cards (word, definition, etymology, one usable example sentence), categories, short quizzes, daily streaks, and home and lock-screen widgets. Use a literary figure everyone in your country knows; for German it could be a Goethe-style name, for Spanish a Cervantes-style one. Give a 3-day or 7-day trial on first open, put the widget behind the paywall, and keep a daily limit on the free tier. Launch on iOS first, prove it works, then go to Play. That's the order Proust used.

How to get users: Copy the structure of the 870k-view clip: slide 1 is a meme, "I used a rare word and nobody in the room dared to ask what it meant," over a classical oil painting; slides 2–4 are three cool obscure words; the last slide shows the app name and icon. First find personal accounts in your own language already doing "saying these words makes you sound smart," see which of their posts blew up, then do the app version of that. For the first two months, post three or four a day. Once organic reach weakens, boost your best-performing carousels on Instagram, pushing the lock-screen widget.

When to launch: Get listed by mid-December and start posting. January is the "new year, better me" season, and self-improvement apps usually get their best stretch of the year (inferred). The next window is September, back-to-school.

Signs you're reading this wrong: ① You've posted 30 carousels and none has passed 50k views: there may be no demand for this format on French TikTok, or the content is off. Stop and fix the content before buying ads. ② Proust drops out of the top 50 on the French Google Play education Top Grossing chart in December, or its daily iOS rating growth falls to single digits. That means the revenue from the open-screen card-linking paywall won't last. ③ Monkey Taps launches native-language vocabulary content for each country: that position is gone. ④ Apple or Google tightens review on forced open-screen trials and card-linking upsells: this monetization method has to change.

---

Data and sources: Google Play store page (France, 11-06), installs and charts (10-29 to 11-06); App Store French page, rating count and public reviews (as of 11-06); App Store search results for German, Spanish, Polish, Italian and Brazilian storefronts (11-06); TikTok @proustapp and @francaisavecmalyk profiles and posts (11-06); Monkey Taps and Eloqa websites; developer info from the Play page's "About the developer," French company registry and LinkedIn search snippets. Ad spend is only based on claims in user reviews, not verified in the ad library. Items marked "inferred" are inferences.