50k installs get it to #38 on the UK Top Grossing chart: SeaChat isn't a hit, it's a big company's test run on a small chart

SeaChat is a women-oriented AI roleplay app. It launched on Sep 2 and by Nov 6 had charted in the top 42 of the Google Play Entertainment Top Grossing charts in the UK, Netherlands, Poland, Sweden and Canada. It looks like a small team's dark horse, but it's actually SeaArt (海艺), the AI image site, spinning out one of its in-site features as a standalone app, and in this one niche they shipped five packages at once. The real scale is small. What indie developers can learn from it is the chart threshold and the pricing, not the product itself.

One error-report review, and a ranking ahead of Character AI

On Sep 15, a user named Carolina_ Girl left a Google Play review: "I keep getting Actor turn. failed: redis set: oom: command not allowed when used memory…… What is this and how do I fix it?" 20 people marked it helpful. Surfacing a server memory error directly to users shows the app was launched in a rush.

Seven weeks later, on Nov 6, the same app ranked #38 in the UK Google Play Entertainment Top Grossing chart, ahead of Character AI at #42. Character AI has 70.7M installs worldwide; SeaChat has 49,954 installs (Google Play installs, 11-06), a gap of roughly 1,400x.

Put these two facts side by side and you have SeaChat's whole story: the product isn't great, but the ranking looks good. Why the ranking looks good is the part worth learning.

What it is

On the store it's called "SeaChat: Live Your Story", with the one-line pitch "Meet AI characters who remember you". Opening it doesn't give you a blank chat box but a row of character cards: a cover, a one-line setup, and an opening scene you can click into and keep playing.

What the characters look like, you can see on the site's "Most Popular" page (seachat.ai, 11-06):

The lineup is nearly all women-oriented romance and "forbidden" themes, with a "Limitless" adult-content switch at the top of the page. The official TikTok bio literally says "Otome rp chat app", otome roleplay.

The creation side borrows the concepts of SillyTavern and JanitorAI: character cards, lorebooks, presets. Some creators write things like "set memory to at least 20k" or "recommend such-and-such language model" in their character bios. This is for heavy users.

Who's behind it

The Play developer account is seaart-seabuzz, and the "About the developer" section on the store page says HONGKONG SEABUZZ CO., LIMITED, with the email domain seabuzz.ai. SeaArt is a site for AI image and video generation. In February 2026 it issued a press release saying MAU exceeded 30M at the end of January, with annualized revenue near $50M (self-reported).

SeaArt has a whole string of apps on Google Play. The AI character chat category alone has five packages, split by audience:

This is a playbook Chinese companies going global know well: one backend and one content library, a new name and a different audience, several packages in the market, add budget to whatever runs and cut whatever doesn't. The SeaArt site already had an "Otome Chat" feature and a character library it claims is 3M+. SeaChat had only 50k installs two months after launch, yet its top characters already have tens of millions of messages, so these characters and data were very likely brought over from the parent site (inference). Another piece of evidence: two pairs of characters on the Most Popular page (Leor and Léonard, Zero and Zeno) have word-for-word identical bios, which looks like cards copied in bulk.

Where users come from

The channels you can see are all small:

• Name piggybacking: The subtitle "Live Your Story" is word-for-word identical to that of Tipsy: Live Your Story (4.88M installs, #3 in UK Entertainment Top Grossing), a leading women-oriented interactive story app. Anyone searching for Tipsy has a chance to see it.
• SEO: seachat.ai has a "Character.AI alternatives" blog and AI character chat landing pages, and every character has multilingual detail pages that can be found in search.
• Official TikTok account @seachat_rp: 2,835 followers, 173 videos, with a few hundred to a thousand views per video. The highest-view video was posted on Sep 29 and got 1,001 views. The Oct 29 Halloween event (limited-time rewards, exclusive avatar frame) got only 458 views. Not a main channel.
• User acquisition: daily installs held steady at 2,000–3,000, concentrated in the UK, Netherlands, Poland, Sweden, Canada and Australia. The free chart only reached the top 100 occasionally, yet the Top Grossing chart kept climbing. This looks like small-budget campaigns aimed at European and Commonwealth women users (inference, ads not verified).
• Community and press: On X and Reddit, almost nobody has mentioned it in the past two months, and there's no coverage at all.

Where the money comes from, and how much

The in-app purchase range was $0.99–$299.99 in Oct 29, and from Oct 30 it became $0.99–$599.99. The top tier doubled, which means they're testing higher-priced tiers.

No public revenue figures. The third-party site AppGazers estimates October US-store revenue under $5,000, with 6,700–20,200 downloads. In the US it only reached #90 in Top Grossing; the places where it made the top 40 are smaller European markets.

The key is the chart threshold. In Nov 6, around ranks 37–42 on the UK Entertainment Top Grossing chart, you have Magic (1.29M installs), My Muse (1.38M), dotdotdot (380k) and Character AI (70.7M), with SeaChat at 50k installs sitting in the middle. That means the middle of the category Top Grossing chart in markets like the UK and Netherlands has a surprisingly low bar, probably only a few hundred dollars a day in revenue (estimate). The UK ranking was climbed: #77 in Oct 29, #38 in Nov 6.

What users say

Rating from 3.1,167 reviews (UK store page, 11-06). The most-liked reviews are mostly complaints about the experience:

"Totally doesn't follow the storyline. No matter how many times I correct it, it only uses the details it wants." (Tisha Crabtree, 9-09,16 people found this helpful)

"Mediocre. Old chats can't be deleted, you can only edit and roll back, and editing doesn't delete the chats after it, unlike Emochi. No branching either." (Jerez Sy Juco, 10-17)

Add the redis error at the top. Users stay because the character themes fit them, not because the product works well.

Who it's competing with

This category is packed. The broad-audience leaders Character AI, PolyBuzz and Talkie are all in the tens of millions of installs. Emochi (FlowGPT) is the closest in approach to SeaChat, also moving from a web community to an app, but it launched two years earlier and has 29.7M installs. In women-focused romance, Tipsy is #3 in UK Top Grossing. Among same-period newcomers, Dulci (Sep 18), which launched 16 days after SeaChat, already has 220k installs and 92k new in 7 days, four to five times faster. Whispy AI has 550k installs and is #13 in UK Top Grossing. Below them is a long list of similar products like Crushie, Magic, Afizzy, Mira and Popia. The competitor Charmsy has already published an article titled "SeaArt Otome Chat alternative" to siphon off traffic.

The selling points SeaChat pitches, "characters bring their own stage" and "creators publish worlds with character cards and lorebooks", are shared by its rivals. What truly sets it apart is only two things: the otome positioning plus the adult toggle, and the SeaArt backing. Anyone can build the first; you don't have the second.

If it were you

Verdict: Don't build like SeaChat. It charted thanks to its parent company's ready-made content library and resources, plus a low-threshold chart. The product itself has nothing you can borrow, and a dozen faster competitors are already in the space.

Who it suits: teams already building AI companionship or interactive stories with some paying users. Use it as a reference and learn three things: run campaigns in low-barrier countries and categories, test higher IAP tiers, and split audiences across several packages.

Who it doesn't suit: indie developers trying to build a "women-oriented AI romance chat" app from scratch. You'd be competing with Tipsy, Dulci and Whispy for user acquisition, with SeaArt and FlowGPT for content libraries, and you'd have to absorb the review risk of adult content.

How to do it (if you insist on entering this category): Don't build "yet another character card plaza". Reviews complain about storylines drifting, the app forgetting setup, no branching, and no deletion. Build something clearly better at "follows my storyline and remembers it", or pick a language that isn't saturated yet (for example, otome storylines in a local language rather than English). For IAP, use it and Tipsy's high-priced tiers as a reference, and don't rely on subscriptions alone.

How to acquire users: Learn how it picks markets: start with small-budget campaigns in countries with low-threshold Top Grossing charts, like the UK, Netherlands, Nordics and Poland, and use chart rankings to earn organic exposure. For creatives, open with a character's one-line setup ("You walk into your father's office and catch the mafia boss"), since there's a ready-made audience under BookTok and #aiboyfriend on TikTok.

When to wait: No rush. This isn't a window that's opening; it's a category that's already crowded.

What would mean you've got it wrong: ① SeaChat tops 500k installs within a month or two, or reaches the top 30 in US entertainment Top Grossing, which would mean the otome niche is bigger than it looks and worth a fresh look; ② SeaArt shuts down Sea Soul / SeaHaze and concentrates resources on SeaChat, which would mean women-focused romance is the winner they've picked; ③ third-party data shows its European monthly revenue passing $50k, which would mean the paying density per 50k installs is far higher than estimated, and the high-priced IAP route is worth more than it appears.