profitlab360

AI-powered niche research and SEO tag generation...

Visit Website
July 18, 2026 I built a niche research tool for Redbubble sellers — here's what 3 weeks of building taught me about the POD market

## The problem I tried to solve I've been selling on Redbubble on the side for ~18 months. The hardest part isn't designing — it's deciding **what to design for**. Most "niche research" advice online is either: - Generic ("try evergreen niches!") - Outdated (2021 lists that are now saturated) - Locked behind $97 courses that just resell free data So I built the tool I wished existed. 3 weeks, solo, shipped. ## What I built - Scrapes Redbubble search results for any keyword - Cross-checks Etsy + Google Trends to estimate real demand - Scores each niche on 4 axes: demand, competition, design difficulty, evergreen potential - Generates SEO-optimized tags (Redbubble allows 50 tags per design — most sellers waste them) - 502 indexable content pages for SEO (10 blog posts, 50+ niche analyses, tools) Stack: React + Supabase Edge Functions + Gemini for the AI scoring + Firecrawl for scraping. 24h cache to keep API costs sane. ## 3 things I learned the hard way **1. The POD audience hates "AI" branding.** First landing copy said "AI-powered niche discovery". Conversion was awful. Changed to "find underserved niches before everyone else" — same product, 3x signups. POD sellers are burned out on AI slop. **2. SEO content is the only sustainable acquisition channel for a tool like this.** Paid ads to POD sellers = expensive and they don't trust new tools. Reddit = banned my first 3 posts (cue sad trombone). Twitter/X = great for devs, useless for the actual ICP. So I went all-in on 502 indexable pages targeting long-tail queries. Long game. **3. Redbubble's algorithm rewards consistency over quality.** Talked to 5 sellers doing $1k+/month. Common pattern: 500+ designs uploaded over 12+ months. None of them are "great" designers. They just kept shipping. This insight is now baked into the tool's recommendations — it doesn't tell you to design 1 masterpiece, it tells you to find 20 viable niches and ship simple designs to all of them. ## What's next - Indexing — sitemap has 502 URLs, Google has discovered all of them. Now waiting for them to crawl. - Distribution — exploring Product Hunt launch, guest posts on POD blogs. - Paid tier — currently 100% free. Will add a tier for higher-volume users once I have signal that the free version actually delivers. ## Open question for IH community For low-CAC SaaS targeting non-dev audiences (POD sellers, Etsy shops, hobbyists) — what's worked for you for distribution beyond SEO? I'm trying to avoid the "build great product, get 12 signups" trap. Happy to share more numbers in the comments. Roast the landing page if you want — [profitlab360.com](https://profitlab360.com)

1 Comment

  1. 1

    The interesting opportunity isn't finding profitable Redbubble niches—it's helping creators repeatedly make better portfolio decisions before they invest time designing. I'd keep validating whether customers stay because the tool discovers opportunities or because it consistently improves the quality of the bets they choose to make.

About

ProfitLab360 helps Print-on-Demand sellers (Redbubble, Etsy, Amazon Merch) find profitable niches before they get saturated. Scrapes live marketplace data, cross-references demand signals from Etsy and Google Trends, and