From Affiliate Site to SaaS: How I Built 14 Free AI Tools in 3 Weeks
Posted on IndieHackers by aifreestack
I run FreeAIStack — a site that started as an affiliate/review site for free AI tools. You know the type: "Top 10 AI Writing Tools of 2026," comparison articles with affiliate links, hoping to make a few bucks on repeat traffic.
It worked, barely. I was getting about 400 visitors/day but making maybe $50/month in affiliate commissions. Most readers clicked through my Amazon links, landed on a paid tool, and bounced. The conversion funnel was just too long.
The Pivot
Last month I made a decision: instead of recommending other people's tools, I would build my own.
The idea was simple — take the most commonly recommended free AI tools in each category and build a free version myself. Resume optimization, email writing, content generation, image creation, video scripts, background removal. 14 tools total, each with a generous free tier (5-10 uses/day).
I spent 3 weeks building the MVP using Flask + DeepSeek API. The whole thing runs on a $20/month VPS. No VC funding, no fancy infra, just me and the code.
The Numbers So Far
It's been 3 weeks since the relaunch. Here's where we are:
136 unique visitors total (since switching to Umami analytics)
2,053 pageviews
~10-15 visitors/day (still very early)
0 paying customers (ouch)
The good news: Google has already started sending organic traffic. Today I saw 5 visits from Google search. That tells me the SEO pages are working — they just need time to rank.
What I Built
The hardest part was deciding what NOT to build. I kept scope tight:
5 core AI tools (Resume, Email, Content, Image, Video) with Pro tiers
9 smaller tools (Background remover, translator, code checker, etc.)
46 programmatic SEO landing pages (e.g., "AI Resume Optimizer for Software Engineers")
20 manual blog articles targeting high-volume keywords
Everything runs on DeepSeek's API, which costs me about
0.50
/dayininference.Hostingis
0.50/dayininference.Hostingis20/month. Total monthly burn: ~$35.
What I'm Doing for Growth
Traffic is the bottleneck. I can't optimize conversions when I only get 15 visitors/day. Current strategy:
Programmatic SEO — 46 role-specific resume pages + 10 tool-specific landing pages
Manual blog content — 20 hand-written articles, no AI slop
Directory submissions — Submitted to ~10 AI tool directories
Sitemap with 131 URLs — Submitted to Google Search Console
Google natural traffic is starting to come in
What I'd Do Differently
If I could start over, I would have submitted to Google Search Console on day one instead of waiting. The 20 blog articles and 56 SEO pages are sitting unindexed because I didn't submit the sitemap early enough.
Questions for the Community
For those of you who have bootstrapped SaaS products:
How long did it take to get your first paying customer?
What growth channels worked best for a free AI tool?
Should I focus on Product Hunt relaunch or organic growth?
Would love any feedback on the site: https://aifreeaistack.com
Built with Flask, DeepSeek API, Umami analytics, and more coffee than I'd like to admit.
The SEO play is the right long game but 15 visitors/day means you can’t wait for Google. The fastest path to your first paying customer right now is going where your exact user already is Reddit threads where people are asking about free AI resume tools, AI writing tools, etc. Answer genuinely, mention the tool only if it fits. That’s faster than any directory submission.
The affiliate-to-SaaS pivot makes a lot of sense. Shorter funnel, more control. On your growth question — directory submissions worked well early on, but the channel that actually converted was showing up in conversations where founders were already frustrated with their current tool stack. Not promoting, just being genuinely useful in the thread. The organic tail from that compounds quietly over weeks.
We did the same at AmpPilot (amppilot.com) AI marketing platform— built it because managing content across Claude, Canva, Buffer was too much glue work for early-stage founders. Scratching our own itch basically. The founders who found us through old community threads converted way better than anyone from a launch-day spike.
Fourteen tools in three weeks is ambitious.
Which one surprised you the most in terms of actual usage?
I had a free trial customer but he ended up leaving . Your website looks really good and I love the route your taking. Think you got more experience than most creators who are not used to marketing.
I liked that you separated the bottleneck from the symptoms.
It's tempting to keep tweaking onboarding or pricing when the real limitation is simply not having enough people coming through to learn anything meaningful.
At what point would you decide traffic isn't the bottleneck anymore and start optimizing the product instead?