During a systematic sweep of professional report-writing workflows, I found two gaps. One was clean. One was complicated. The difference between them had nothing to do with market size or the existence of demand.

It had to do with prerequisites.

The Clean Gap

A clean opportunity has one workflow, one buyer, one output format, and no technical prerequisite to delivering value.

The professional does field work, comes back to the office, and writes a structured report. The report follows a defined standard. Every report is roughly the same format, same sections, same required content. The professional spends hours doing this on every engagement.

The product story is simple: input the field data, generate the report. AI fills the prose sections with appropriate language. Professional reviews and edits. Hours saved per report, dozens of reports per year per user.

No engineering calculation required before you can start writing. No domain-specific computation that must precede the output. Just structured observation translated into structured documentation.

That’s a clean gap.

The Complicated Gap

A complicated opportunity has the same surface characteristics — structured workflow, defined output format, professional community — but adds a prerequisite that changes the product story.

In the energy audit space, the workflow has two distinct phases. Phase one: perform engineering calculations to determine what efficiency measures are cost-effective. Phase two: write narrative descriptions of those measures, organized by category, with financial justifications.

The narrative writing (phase two) is where the time goes and where AI could help. Professionals spend hours writing essentially similar descriptions of similar measures across different projects.

But you can’t skip phase one. The AI can’t help with phase two without knowing the output of phase one. And phase one is an engineering calculation that professionals do in specialized tools.

This isn’t insurmountable. You could integrate with those tools. Or require data entry of the phase-one results before generating phase-two content. But suddenly the product story has conditions. “AI writes your audit narrative — if you import your EEM results first.” The buyer has to do work before they get value.

Why Prerequisites Change Everything

The clean gap has zero conditions on value delivery. You open the product, enter field observations, generate a draft. The value is immediate and the product story fits in a sentence.

The complicated gap has conditional value. The product helps you — after you’ve done the prerequisite work, using the prerequisite tools, in the prerequisite format. Each condition is a reason to not bother.

This matters more in some markets than others. Enterprise buyers with established workflows and dedicated staff can handle integration overhead. Small professional services firms — two to ten practitioners, generalist staff, no dedicated software administrator — cannot. They make purchasing decisions based on whether the product solves a problem, not whether it solves a problem if they also change their other workflows.

The Other Complication: Incumbent Count

The clean gap I found had one aging incumbent. Form-based. No AI. One product, one company, minimal recent development.

The complicated gap had three. Each slightly different in scope, each with an existing user base, each requiring a new entrant to displace a familiar tool rather than fill empty space.

Displacing one old tool is tractable. The incumbent’s users aren’t loyal — they’re stuck with what exists. If you build something demonstrably better, the upgrade path is straightforward.

Displacing three tools that serve overlapping but distinct segments requires either building for all three segments or choosing one and accepting that two-thirds of the market will have a “good enough” alternative. The market fragmentation is a moat, not for the incumbents, but against new entrants.

What Makes Something Tier-1

Tier-1 opportunities have:

  • One workflow with one output
  • No prerequisite computation before value delivery
  • One or zero incumbents (ideally one old one with no AI capability)
  • A buyer who can make a purchase decision without involving IT, procurement, or executive buy-in
  • A problem description that fits in two sentences

The test: can you explain why someone would buy this product in the time it takes to walk from the elevator to your floor?

If the explanation requires “and then” more than once, the product story is complicated. Complicated product stories take longer to validate, longer to sell, and longer to realize whether they were worth building.

The gap is real in both cases. The question is whether the gap is accessible enough to be worth entering.