FlowDesk
The privacy-first, back-endless command center for freelance
Hey everyone!
Kept my head down and kept shipping since the last update. Here's exactly what changed:
Pro Lab — added a new section with experimental tools that run on your actual data:
Client Health Predictor — gives each client a reliability score (0–100) based on real payment history
Cash Flow Simulator — sliders for target income + hourly rate that show exactly how many hours/week you need
Burnout Guardian — monitors your weekly tracked hours and locks the Pomodoro timer if you exceed your set limit. Hit the threshold → timer blocked until you click "End Recovery"
Global Payments — added Pix (Brazil) and WeChat Pay (Asia) in Settings. Saves the key and appends the payment instructions to relevant invoices automatically.
Localization — the language switcher now works across the full app dynamically. English, Romanian, Spanish, French, German.
Themes — added Cozy Terra Cotta and Ocean Blue. 3 options total now + custom accent color picker.
Demo Data — one-click button that populates the entire dashboard with realistic invoices, clients and expenses.
Fix — the monthly/annual pricing toggle on the landing page was broken. Fixed.
Still 100% client-side. Nothing goes to a server.
Hey everyone!
I just pushed a series of updates to FlowDesk based on some great feedback I received over the last 24 hours. As a solo dev, I really appreciate everyone who took the time to poke around and point out where things could be better.
Here’s what I’ve improved:
Design Cleanup: Toned down the landing page UI. I’ve removed some of the excessive emojis and simplified the header styles to make it feel more professional and less cluttered.
Pricing Transparency: Updated the pricing section to show the monthly equivalent for annual plans ($7.99/mo). It makes the 20% discount much easier to understand at a glance.
Tighter Spacing: Adjusted the layout spacing (especially in the hero section) to make the page more compact and focused on the content.
Copy Refinement: Removed some of the "marketing-heavy" claims and replaced them with straightforward facts about the privacy-first, client-side architecture.
Technical Fixes: Fixed a layout shift bug in the features section, added smooth hover transitions, and finally added a proper favicon for browser tabs.
Live Changelog: Added a permanent changelog section at the bottom of the site so you can track the status of new features and fixes as I ship them.
Thanks again for the support and the honest feedback. It’s helping me make FlowDesk a much better tool for everyone.
Check it out live: flowdesk-ecru.vercel.app
Hey Indie Hackers!
Just wanted to share a major milestone: FlowDesk, the privacy-first command center I’ve been building, is live on Product Hunt today!
It’s been an incredible journey building this "backend-less" tool, and I’d love to get the community's support to help it reach more freelancers who care about privacy and speed.
If you have a moment, I’d deeply appreciate you checking it out and sharing your feedback.
Product Hunt Link: https://www.producthunt.com/products/flowdesk-2?launch=flowdesk-2
Thanks for all the support so far!
-
2
Love the 24h iteration cycle — solo dev speed is superpower.
Question: When prioritizing these updates (design vs pricing vs tech fixes), how do you validate which one actually moves paid conversions vs just making free users happier?
I learned "users asking" ≠ "revenue impact" the hard way ($20K lesson). Now I test feature ROI with video prototypes before coding — validate which improvement drives upgrades.
Curious how you measure impact of these sprints on revenue vs engagement!
-
0
Really appreciate the question — and honestly, I don't have a serious measurement system yet. Too early for clear conversion data.
I mostly prioritize by 3 things: what more than one person mentions independently, what's broken and needs fixing anyway, and what I'd personally want as a user of my own tool.
Your $20K lesson is exactly the kind of thing I haven't learned firsthand yet — but probably will. For now I'm going on instinct and direct feedback.
Video prototypes before coding is smart, especially if you have an audience to test with. At my stage I don't really have that, so I just ship and see what happens.
How did you find your first people to test prototypes with before launching?
-
2
Hey Angoleyus — congrats again on the PH launch! 🚀
On finding the first people to test video prototypes (before you have an audience), I kept it super scrappy:
Start with “already-in-pain” posts
Search IH / X / Reddit for people who posted about the exact problem in the last 7–14 days. Those are the warmest.DM with a tiny, low-friction ask
“Saw your post about X — I’m testing a 30–60s walkthrough to see if it actually solves it. Can I send it? No signup. Just 2 questions.”Ask conversion questions, not feature feedback
“What would stop you from paying?”
“What’s unclear / missing for you to try it?”
“Would this replace what you do today?”Trade feedback with other builders
10 mins for 10 mins works surprisingly well early on.
If you want, I can share my exact 2-question script + how I structure the prototype video so it maps to paid conversion (not just ‘nice-to-have’ feedback).
-
-
I’ve always been fascinated by how complex and "heavy" modern software has become. Most tools today require an account, a database, and a constant connection to a server just to create a simple invoice.
I built FlowDesk because I wanted to see if I could create a truly professional, "stunning" command center for freelancers where the user is in total control.
No accounts, no servers, no middleman.
The challenge was to pack all the features—professional invoices, CRM with reliability scoring, and automated reports—directly into the browser's LocalStorage. I wanted to build something that feels "snappy" and follows a strict privacy-first philosophy.
I’m launching it today not just as a tool, but as a statement: you don't need a massive infrastructure to run a freelance business professionally. You just need a well-crafted space to focus on your work.
I’m excited to hear what other makers think about this "local-first" approach!
Comment