Marcin Michalak

Built 8 products. Shipped 3. Killed the rest early. Helped take a B2B SaaS from pre-seed → Series B → exit. Now I build simple tools for offline businesses and help others find the simplest solution to complex problems. Philosophy: less is more. Kill early. Ship what matters. Follow for honest product lessons.

this OpenClaw bot finds hotels with ugly listing photos, redrafts them as IG posts, and mails the owner a postcard...on autopilot. here's how social-media agencies can use this system and land hotel clients: - scrapes every hotel in a city in real time - filters by review count + rating + last post date + photo quality - pulls the strongest interior shot from the Google Maps listing - samples the brand palette from the hotel's actual visual identity - AI-redrafts the photo into a 1:1 brand-matched Instagram post - writes a postcard quoting a real reviewer + the exact space - mails it to the owner hand-addressed with a preview QR every step from discovery to brand-matching to outreach is automated. reply "HOTEL" + RT and I'll send you a free guide so you can build this too

@everestchris6 Hotel!

@SoraiaDev Look at session count, not just total actions. 300 actions in one sitting tells you something different than 300 actions across a few weeks. The users who convert are the ones who came back; if they only showed up once, no paywall was going to save it.

do you think some users will just never convert? no matter what you do? look at how much these free miromiro users got out of it: 1. 361 actions 2. 302 actions 3. 261 actions 4. 111 actions trial ends, they're gone. a hard paywall wouldn't convert them either are these users even reachable?

it's Monday. May I wish you all super high conversion rates? What's going on? 🤣

@JustenEcom Still WIP 😇

@marcin_knows bro built a lifestyle not a routine 💪🏻

Thanks to: - running 3x a week - bouldering / climbing twice a week - biking to my office every day - eating healthy - enjoying all of the above 😇

@nestymee Bouldering! 🤜🤛

aesty.ai founder Sunday workout Not AI btw 😄😎

Are you okay @claudeai ? 🙄

Paraphrasing the best reframe I've heard for indie builders: "Ask yourself if this problem genuinely bothers you. If it does, you'll keep showing up to fix it - quietly, consistently - long after the initial excitement fades."

The builds that actually matter are the ones that break you a little. Where you find out if the idea holds - and if you do too.

@eliana_jordan follow and get the guide

I made $30,250 in March 2026 From 6 apps Here is how🧵

@kapilansh_twt Exactly. Simple solutions, no bs and people love it.

@marcin_knows classic jobs-to-be-done in SEO form, people don't want your product, they want their problem solved, the tool is just what shows up at the end of that search

I spent a weekend analyzing competitors. Combined they get ~190K organic visits/month. Their top pages aren't product pages. They're how-to guides. "How to add email signature in Outlook." "How to change signature in Gmail." People don't search for email signature tools. They search for help with their email client. Then they find the tool. That's the entire content strategy.

I run 6 side projects and none of them are "successful"-here's why I don't care. None of them are "successful" by tech Twitter standards. But here's the thing: they all work. They solve real problems for real people. Some of them make money. All of them taught me something. I don't need a unicorn. I need a portfolio of small, useful things that compound over time.

@mal_shaik Amazing breakdown, thank you for sharing. Would you recommend any good guide on how to setup good Claude.md file?

People always ask about the stack. Here's what actually matters for a small tool like this: The signature is just table-based HTML with inline styles. That's it. No fancy framework needed. The hard part wasn't the code. It was making it render the same in Gmail, Outlook desktop, Outlook web, and Apple Mail. Four different rendering engines. Zero consistency.

@ideabrowser in case you work a lot with .csv/.xls — csvtosheets.com

I want to spend $1,000 on new apps and tools today. Show me what you are selling. SaaS, Mobile Apps, Mac Apps, Agents, etc. Drop your link. If I like it, I'm buying.  If I don't buy, I'll give feedback on why I didn't convert. Who wants some customers?

Turning an agency internal tool into a product is weird. When it was just for our clients at @hehe_studio , I didn't care about onboarding. I'd just hop on a call and walk them through it. Productizing forced me to rethink everything: → The setup flow had to work without me explaining anything → Outlook rendering bugs I used to fix manually needed automated solutions → "Just email me if something breaks" doesn't scale past 5 users The hardest part isn't building features. It's removing yourself from the process. Still working on it.

@SoraiaDev Dark mode only! Don't forget about adding tooltips on hover for the icons, so it's newbie friendly 😇

this toolbar might be the cleanest UI I've ever built wrapping up core features on PocketUI before I record the demo video for the landing page should I add a light mode or keep it dark only?

@audiencon brandfooter.com

drop your project i’m boosting builders tonight 👇