Quick context for anyone new: VidClean (vidclean.net) is a free online video and audio tool suite. No account, no watermark, no time limits. I'm building it in public, modeled on iLovePDF's approach. One free tool at a time, compound organic traffic over time, monetize later.
What shipped since my last post
The product has grown significantly. When I last posted I had around 8 tools. Now at 18.
New tools added: silence remover (flagship), remove background noise, enhance speech (free Adobe Podcast alternative using DeepFilterNet3), transcription to txt/srt/vtt using OpenAI Whisper, video compression, crop, trim, merge, stabilize, and more.
New languages: full Spanish translation (18 pages), full Brazilian Portuguese translation (17 pages). Brazil is 9% of traffic with zero Portuguese content previously.
Infrastructure: upgraded to Railway Pro, added a dedicated Whisper transcription worker with full queue isolation so transcription never blocks the silence remover.
Current metrics (honest)
Daily visitors: 30-40, stuck here for about 3 weeks. Weekends drop to 20-25.
Google Search Console (3 months): 131 clicks, 2,580 impressions, average position 59. Homepage filters to position 7.3 for brand queries. Non-brand organic is nearly zero.
Tool completions: 414 total across all tools since launch. Silence remover at 267, background noise at 55, transcription just launched.
AI citations: ChatGPT is my second largest traffic source. Copilot has cited VidClean 68 times over 3 months. Gemini has cited it once. This is the channel that actually works right now.
DR: 0.2. One confirmed referring domain in Ahrefs. This is the main bottleneck.
Revenue: $0.
Infrastructure cost: $25-30 per month.
What is working
The AI citation channel. The formula is: ship a tool with a specific searchable use case, drop helpful Reddit and Quora answers, get indexed, ChatGPT starts citing within 24-48 hours. This worked for the silence remover and is starting to work for background noise removal.
The "no account, no watermark, no time limit" positioning. This is what gets cited. Every competitor either watermarks, requires signup, or caps the free tier. VidClean does none of those things and AI systems highlight that when recommending tools.
What is not working
Backlinks. This is the core problem. Getting quality dofollow links without paying is genuinely hard. AlternativeTo rejected the site. Most directories either charge, require a link exchange badge, or are dead. Dev.to posts have the right DR but are getting 1-5 views each. The only meaningful backlink so far is whatlaunchedtoday.com.
Breaking the 30-40 visitor ceiling. The AI citation channel works but has plateaued. Without backlinks, Google organic stays near zero. Without Google organic, daily visitors stay stuck.
Key decisions made along the way
Stop building new tools until distribution works. Then broke that rule when the audit showed transcription was the only new tool with a real citation vacuum. Every competitor caps the free tier and none can match "no time limit."
No paid ads. No monetization to calculate ROI against yet.
No Pro tier until 150 daily visitors. Still holding that line.
What is next
Breaking the DR problem is the priority. Just submitted to WeekHack (DR 50+ dofollow, launching next week). Writing more technical posts on Dev.to and Hashnode. Submitting to AI tool directories now eligible because of the Whisper transcription tool. Exploring HARO for editorial links.
On the product side: /add-subtitles with burned-in captions is the next build. Then Product Hunt launch once the differentiation story is stronger.
Happy to answer questions about the stack, the AI citation strategy, or the building-in-public process. Follow along at vidclean.net or @thebuciyo on X.
The AI citation strategy is smart and I think you're onto something most people miss. I've been tracking a similar pattern — we built a free SEO audit tool with 35+ checks and found ChatGPT started citing it within 48 hours of hitting certain niche technical queries. The key insight nobody talks about: AI citations compound their own DR over time because every citation backlink creates a crawlable path that Google then discovers. Your silence remover at 267 completions with Copilot citing it 68 times is exactly that loop starting. What's your strategy for getting more ChatGPT citations? Are you targeting specific query patterns?
The query targeting is the whole game. We focus on constraint-specific queries rather than head terms. Things like "free transcription no account no time limit" or "remove silence from video free no watermark" where every competitor fails at least one clause. When ChatGPT searches for those, VidClean is often the only answer that satisfies all constraints simultaneously.
The seeding formula that has worked: find a fresh Reddit or Quora thread where someone is asking that specific question, drop a genuinely helpful answer mentioning VidClean, and citations typically start within 24-48 hours of Google indexing that thread. The key is the thread has to be about the exact constrained query, not just the general topic.
The AI citation angle is the most interesting part here. Most free-tool sites are still thinking only in Google SEO terms, but your strongest channel is already ChatGPT/Copilot recommending specific use-case tools because the positioning is simple: no account, no watermark, no time limit.
That feels like the real wedge. Not just “video tools,” but a utility suite built to be discovered and recommended by AI engines when users ask for a very specific job to be done.
One thing I would pressure-test before Product Hunt and more backlinks is the brand frame. VidClean is clear for cleanup, but the product is already wider than cleaning: transcription, compression, crop, trim, merge, subtitles, speech enhancement, and possibly more AI-discovered media utilities over time. The name may start feeling narrower than the suite.
Exirra .com would fit the broader direction better because it feels more like a modern AI-discovered utility platform than a single video-cleaning tool, while still leaving room for transcription, enhancement, editing, subtitles, and future media workflows under one cleaner brand.
Yeah the vidclean name may be a little narrow but switching names now would be a big step back