12,000 users came from meta's twitter clone...

intro

mat an ios widget app that turns your photos into cutouts on your home screen.

he started marketing on threads and in the last 90 days his account pulled 829,760 views from 502,179 people.

i wrote earlier about why threads is worth your time. this one is how we actually did it.

tldr

→ threads doesn’t cap your reach at your follower count.
→ communities are the whole game, profile posts are where your reach goes to swahha.
→ replies pulled as many views as original posts did.
→ never post without an image.
→ it took 90 days and a six week dead patch in the middle to reach 12000 users & $2000.

the main thing

on X, your reach is mostly decided by who already follows you. you build an audience first, then you get distribution.

threads works backwards. you post into a community and it shows your post to people who joined that community, whether or not they know you exist. 99.9% of the people who saw rahul’s posts didn’t follow him.

that means a brand new account with 700 followers can reach half a million people in three months. we didn’t grow an audience and then launch. we launched into rooms that were already full of people with that interest.

why this is better than the alternatives

on july 12 rahul posted mat’s launch on X. 600 users, 1 paid.

three days later, same app, on threads. 1,400 users and $178 in the first 28 days.

X is still the best place on the internet to find people who will try your product. everyone there is building something. they’ll download, poke around, send you a feature request and disappear. awesome for feedback.

threads is where people actually paid for mat.

we tried reddit too. one post, about a thousand impressions, a genuinely global audience. attribution was a mess and it never became a channel. fine as a probe, never worked out as a strategy.

what should you do to achieve the same?

1. post in communities, not on your profile.
   for mat that was build in public, apple, art threads and stickers. stickers looks like a strange fit for an app until you remember mat literally makes cutouts.

   
   think sideways about your own product. if you’re building for football fans, football and team communities are obvious. ios apps and app design are the ones that’ll surprise you.
   
   → one community for what the product is about
   → one for how it’s built
   → one for what it looks like

2. treat replies like posts.
   rahul’s top 14 replies pulled roughly 190K views. his top 4 posts pulled roughly 180K. replies weren’t a supplement, they were half the engine.

   
   the method: find people posting home screens, aesthetic setups, cutout edits. anything mat belongs in. then say one relevant thing like a normal person.
   
   “you make your phone feel personal. you might like Mat, a Home Screen thing i made” did 25K views.
   “Missing mat here” did 13K.

   no pitch just share & media for them to see.

   

3. always attach media.
   almost nothing went out without an image. product on a real phone, the numbers, a before and after. carousels worked well, product first slide, revenue dashboard second.

   
   if you don’t have a visual, you don’t have a post yet.

4. write like a person.
   rahul’s launch post opened with “I am a designer with 12 years of experience and My first IOS app just went live”. random capital M. IOS instead of iOS. no hook formula.

   
   138 likes, from an account with no authority on the platform.

5. post in the evening.
   his audience is most active 8:30 to 11:30pm IST, heaviest on sunday, saturday and wednesday. check your own insights, but evenings won.

6. post through the dead patch.
   look at the graph. big spike at launch, then almost six weeks of nothing, then it picks up from mid august and holds.

that flat middle is where everyone quits. the spike makes it feel like it’s working. the silence after makes it feel like it stopped. neither is true.

what not to do

don’t use scheduling apps. you can’t pick a community through them, so everything lands on your profile. profile posts were the worst performing thing we tried, by a wide margin. post manually.

don’t let AI write your posts. we tested it. predictable AI copy didn’t flop, it just did nothing at all. use it to think if you want, don’t let it type.

don’t chase one viral post. the 95k view post was great. it didn’t build the business. 90 days of showing up did.

don’t post without knowing which room you’re in. a good post in the wrong community performs worse than a bad post in the right one.

ending notes

829,760 views. 12,000 users. $2,000 in revenue. one founder, one account almost nobody followed, most important is zero money spend.

none of this is a hack. it’s finding the rooms where your user already is, showing up in them daily, and being useful enough that a stranger taps through.

the platform everyone’s ignoring is the one where a nobody can still get reach. that closes eventually. right now it’s open.