TinySeed Tales: how Colleen found her ICP after pivoting to marketers

Executive overview

Colleen Schneetler is 22 months into her startup and still searching for the right market. Her original product — AI-powered reporting for engineering managers — generated initial excitement but no real engagement, because it wasn't solving a genuine pain point.

She pivoted to helping marketing data analysts query their databases using SQL, landed her first paying customer, and is now focused on finding a specific pain point within that audience to build a foothold.

The bar for product quality has risen so sharply that finding a niche with weak competition is now harder than ever — the only path forward is relentless customer discovery and continuous product improvement.

Abandoning the engineering manager idea

• Early users found the product interesting but never integrated it into their workflows
• Core problem: AI isn't yet good enough for fully non-technical users to operate independently
• The embedded reporting space had too many well-funded, polished competitors
• No unique angle emerged — Colleen decided to stop rather than consult her way into a dead end

Pivoting to marketing data analysts

• A marketing coach (formerly a marketing analyst) immediately said he would have bought the tool
• Key ICP distinction: marketing data analysts are marketing-first, not pure data people — they want to use data to do better marketing
• Colleen's strategy: build in public, run a newsletter, send cold LinkedIn DMs, teach marketers SQL
• First paying customer converted two weeks before recording — despite openly disliking the UI

What the product does today

• Chat-with-your-database: type a question, get a SQL query, run it, get charts or exportable tables
• Editable SQL was the most-requested feature — but almost no one actually edits it; they just want to see it
• Dashboard and shareable live links are gaining early interest
• Next potential feature: database triggers to push reports automatically (top request from paying customer)
• Pricing set at $59/month — chosen arbitrarily, expected to rise as value is proven

The crowded market problem

• Every niche Colleen investigated — insurance, environmental surveys, niche RFP markets — already had multiple polished competitors
• Ten years ago a rough MVP could win on functionality; today users expect a finished, polished experience
• "Chat with your database" is not a differentiator — many companies are doing it
• Positioning as "the data analyst for your company" is too vague; users still need to know what to ask

What comes next

• Six months of runway to find a signal worth pursuing
• Goal: identify a specific pain point where existing tools are either grossly overpriced or genuinely broken
• Rob's framing: success comes from finding where you fit in the market, not from inventing a new category
• Colleen sees potential in aggregating data sources into BigQuery and sitting on top as a cheaper Looker alternative
• "Chat with your database build reports" is not yet a problem statement — finding that problem is the next job