A South African shopkeeper gets a new poster every morning: the company behind Magic Poster AI is an Indian firm with over ₹500M in annual revenue

An AI poster app listed less than three months ago, with 167k installs. The developer name reads "P Labs," but scroll to the bottom of the store page and the legal entity is Primetrace in Bangalore, the company that makes Crafto. It hit the top five of the category chart in South Africa and is now falling. Here's what it sells, how the volume was generated, how well it collects money, and what you can take from it and what you can't.

A comment saying "there's a new poster every morning"

In 2026 September 12, a South African user named Brent Williams gave Magic Poster AI five stars on Google Play: "A few sentences and it made professional ads for my coaching business. Every morning it tells me a new poster is ready, and they're all great." The review has 13 "helpful" votes and was still on the first screen of the South Africa store page as of 11-06.

That sentence is basically the whole product: you set up your shop once, and from then on it proactively makes a poster for you every day and pushes it to your phone.

What it is

The store page's first line is blunt: "Magic Poster turns a sentence into a finished promo poster for your business." Usage takes three steps:

1. Fill in your logo, shop name, and a one-sentence business description, once;
2. Say in plain language what you want to promote today: a weekend discount, a new dish, changed opening hours. Attach a photo if you want it built around a product;
3. The poster appears right in the chat window, ready to save to your camera roll or share.

It specifically highlights "no canvas, no layers panel, no design tools to learn," and lists a long roster of target users: cafés, food trucks, barbershops, fitness coaches, repair shops, tutors, real estate agents. The last line: "For shop owners who need to post something today, have no designer on hand, and no time to learn design software."

Compared with editors like Canva or Adobe Express, what it saves is the step of picking templates, editing text, and aligning elements. Compared with older template-poster apps, it adds "remembers who you are" and "sends you one automatically every day."

Who's behind it

The developer name in the store is "P Labs," but the legal entity in the developer info is PRIMETRACE TECHNOLOGIES PRIVATE LIMITED (Bangalore). The support email is [email protected], and the privacy policy is hosted on a kutumbapp page (Google Play store page, 11-06).

Primetrace was formerly Kutumb. Its Crafto app in India does exactly this: makes holiday greetings and promo posters for small businesses and everyday people, in Hindi and other local languages. Crafto has roughly 157M installs on Google Play. The company says it has about 15M monthly active users and about 5M paid subscribers, and Crafto contributes about 80% of company revenue. The founder's framing: they noticed "users were paying local designers to make social posts," so they "productized the graphic designer" (ET Entrepreneur, 2026-03-17). The same article puts annualized revenue at about ₹550 crore and annualized EBITDA at about ₹200 crore.

So Magic Poster isn't a new idea. It's a playbook already collecting money in India, translated into English, turned into a chat format, and tested overseas. The same company has another nearly identical app on Google Play, Business Poster AI (listed Aug 3, with copy that also says "new posters will keep appearing on your home screen"), plus an earlier app with the same name. That's multi-app parallel testing.

Where the volume came from: 37,000 in two days

The install curve makes the source obvious. Cumulative installs were about 13.9k on Aug 31. On Sep 1 and 2 they rose by about 17k and 20k respectively, reaching 51k cumulative by Sep 2 (third-party estimate, AppstoreSpy). From mid-to-late September it ran two to six thousand a day, October one to two thousand a day, and early November five hundred to fifteen hundred a day. On 11-06 it added 587 that day.

Organic word of mouth doesn't produce this shape. It's a concentrated paid push that then decays. Reviewers say so directly: "The ads say it's free, then after a while you have to pay" (9-26, South Africa, 2 stars). The parent company is an experienced advertiser: Google's Ads Transparency Center shows the PrimeTrace account with 348 creatives in total, 322 of them video, and 7 active in the last 117 days, though these point to Crafto and are mainly aimed at India. What name Magic Poster ran under and on which platforms can't be determined from outside.

The spend pushed it onto the South Africa Google Play "Art & Design" free chart: from 10-29 to 11-04 it held #4–#5, dropping to #11-05 and then #11,11-06 by 21. Nearly all of its reviews come from South Africa.

Apart from paid acquisition, it had no other channel. Searching TikTok for "magic poster ai" turns up no relevant videos (11-06), nobody mentioned it on X in 30 days, Reddit only has generic "AI for posters" discussion, and no media outlet has covered the app. The only thing that got press was the parent company's revenue.

Why South Africa

Inference: mostly Android, English works out of the box, small shops rely heavily on WhatsApp and Facebook for promotions, and ads are cheap. The reviewers are the target users: "This makes advertising and marketing so easy" (9-23), "Thank you Magic Poster team, my business is booming" (9-26), "You made me a star among my peers" (10-10, all caps). The South Africa store page shows 11-06 stars with 4.2 ratings and 269 reviews.

How well it collects money

The model is a hard paywall after a free trial. Starting late September, one-star reviews piled up: "Was a good app at first, now it says I have to subscribe and I can't use it, so I deleted it" (9-22), "Now it costs money" (9-26), "Won't open, forces me to subscribe" (9-26), "I don't want to subscribe, the limited free use was fine" (9-28). One user mentioned pricing: "Two weeks ago I paid 99, it said one month, now it won't work and wants me to pay again" (9-30). The South Africa subscription is about R99 a month, roughly $5+ (est.). The store lists in-app purchases at $29.00–$48.99 per item (US store view), and nobody outside can tell exactly how that maps to prices in each country.

Revenue signals are weak. 11-06 It appeared on the Art & Design Top Grossing chart in 15 countries, but always ranked between #132 and #149, and only on that one day. These charts are full-length lists of 200 apps, and the bar is low past rank 130. Inference: revenue is small, and at least for now it doesn't prove the Crafto model works overseas.

Costs are tightening too. 10-10 One two-star review said: "The system-generated posters used to be 4 a day, now it's only 2. There's a daily cap on posters, and I can't reach the developer." 10-27 Another said they get two automatic poster pushes a day, but tapping them shows nothing.

Who it's competing with

The category isn't new and there's no shortage of giants. Magic Poster got in by making "one sentence to finished product" the shortest path in a cheap, empty market, backed by a parent company that buys user acquisition.

The clearest evidence is the sister bundle from the same company: nearly identical product, with Business Poster AI at 15k installs and Magic Poster at 167k. The difference is in the paid-acquisition pacing, not the product.

What you can take, and what you can't

What you can take:

• "Set up the shop once, push one poster a day". This turns a one-off tool into a daily habit, which is what the Brent Williams review praises. For a subscription, that matters more than "how good the output looks."
• Chat instead of an editor. It has the lowest bar for small shop owners who can't design. The most common complaints in reviews are "I typed one word wrong and can't change it" and "it doesn't understand delete." That's exactly where you can beat it.
• Bring a category proven in the home market to a cheaper one. Daily marketing posters for small businesses have already been shown to make money in India. The question is whether they will in other countries.

What you can't take:

• The ad budget, creative team, and multi-bundle testing capacity of a company with over ₹500 million in annualized revenue.
• The templates, festival calendar, and localization know-how Crafto built up over several years.

If it were you

C: worth learning from, but not worth copying step by step. This is a large company's overseas experiment, its volume is bought and already fading, and the weak revenue signal suggests even it hasn't made the business work abroad yet.

Who it suits: people who already have small-shop users or a channel in an emerging market. For example, teams building POS, bookkeeping, food-delivery, or WhatsApp merchant tools who could add "one automatic poster a day" as a feature to lift retention and payments, instead of building a standalone app and buying installs. Or people from a specific country or language (Indonesian, Brazilian Portuguese, Latin American Spanish, Arabic) who know how local shops run promotions on WhatsApp.

Who it doesn't suit: anyone who wants to build a standalone English AI poster app and grow it from organic store traffic (Canva and Adobe hold the top two spots in this category, and Magic Poster itself has no organic content spread); anyone without ad-buying experience (paid acquisition is its only growth lever); anyone targeting developed markets (Canva and ChatGPT's free tiers cover about 80% of that).

How to build it: chat-based, record the logo, shop name, opening hours, and common products once; proactively push a poster each day tailored to local holidays and the day of the week (Friday pushes the weekend special, Ramadan pushes opening hours), one tap to share to WhatsApp status and Facebook; support local languages; get "edit one word" and "swap the image" right, since that's the biggest pitfall in its reviews. Don't paywall as hard as it does. Nearly all one-star reviews say "suddenly costs money." Give free users one or two posters a week, and sell daily posts and watermark removal to paying users.

How to acquire users: Google UAC and Meta ads in cheap markets, with creatives showing "a small shop owner gets a finished poster in the morning and posts it straight to WhatsApp" as vertical screen recordings. Test in one country with a small budget first, and see whether conversion can support a per-user price of a few dollars a month. (Inference: in these markets the price point is low, so acquisition cost has to be very low too.) A cheaper route is to partner with tools or payment channels that already have lots of small merchants and distribute through them.

Timing: no rush, but move before Primetrace spreads this playbook to more countries. It's currently testing in South Africa using several sister bundles. Once it finds a version that makes money, it will likely push the same approach into Nigeria, the Philippines, and Indonesia.

What would show we got it wrong: (1) Magic Poster or Business Poster AI scales up again in early 2027, and reaches the top 50 and stays there on the Art & Design Top Grossing chart in South Africa or another country. That would mean overseas payments work and the opportunity is bigger than written here, but the competitor is Primetrace itself. (2) It stops updating or gets pulled within a month or two, meaning the test failed and small overseas shops won't pay a monthly fee for this. Then don't build it on its own. (3) Canva or Google launches "set up your brand once on mobile, get a poster automatically every day" on phones. Then this gap is closed.