Vidclean

A month ago I posted here about VidClean at $9 MRR after three months. Wanted to come back with an honest update rather than let that be the only snapshot out there.

The real headline: the product changed shape.

VidClean started as a set of standalone tools (28+ now, silence removal, audio repair, subtitle generation, and more), but over the past month it grew a real timeline editor, including something I haven't seen any competitor ship: it learns from a person's own past edits and drafts matching cuts on their next video. Not a template, actually learned from their own editing history. That's the actual product now, the tools are still there and still bring people in, but the editor is what I'm building toward.

Where things stand, numbers-wise:

• MRR is at $51 now, up from $9 a month ago. Small, but real, and every recent subscriber has paid full price, no discount codes.

• Traffic crossed 600 visitors in a day for the first time recently, fully organic, no ads, no paid acquisition, still compounding.

• Domain Rating went from roughly 2.8 to 12 in that same month, without running any active link-building or outreach campaign yet.

• Honest caveat on the MRR number: every dollar collected so far has been a first payment. The first real wave of renewals is happening right now, over the next few weeks. That's the actual test of whether this holds up, and I don't have that answer yet.
  
  Two things worth being honest about:

  1. The editor and style profiles took a month, and I had to scrap and rebuild it once. Style profiles, the feature that learns from someone's past edits, is the idea I actually believe in most here. It's not a small feature bolted onto the tools, I think it's the flagship thing VidClean becomes known for. But getting there wasn't a straight line. I built a version of it, realized the approach wasn't right, and started over from scratch. A month is a long time on a side project next to a day job, and there were points where I wondered if I was over-investing in something that might not pan out. It's live now, and I think it was worth it, but I want to be honest that "shipped a novel feature" doesn't mean "shipped it smoothly."

  2. I found and fixed a real billing bug where subscription expiry dates weren't being read correctly after Stripe changed where that data lives in their API. Turned out zero revenue had actually been lost, the cancellation path worked fine independently, but the safety net behind it had quietly been broken since the beginning. The scarier part: an early version of the fix's own test would have mass-revoked every paying subscriber's access if it had shipped. Caught in review, not in production.
     
     What's still unresolved:

     Retention, genuinely. The free tier is deliberately generous (self-hosted CPU models instead of paid APIs make that sustainable, no account required, no watermark, no monthly cap), and it's converting real subscribers without discounts, which is a good sign. Whether people actually stick around past month one is the real open question, and I'll know a lot more in a few weeks than I do right now.

     That's the honest version. Happy to answer anything.