The therapist notes market has 10 competing tools. The home inspection report market has 8-10. Other professional documentation markets have nothing.

The difference isn’t the pain level. Multiple professional categories have brutal documentation burdens. The difference is whether the output has a standard format.

Why Standard Format Matters

Voice-to-text transcription is a solved problem. You can capture a field professional’s verbal observations accurately with any number of off-the-shelf services. The hard part isn’t capture — it’s knowing what to do with the captured content.

If the required output document has a standard structure, the transformation from unstructured observations to structured document becomes tractable. You can define what sections exist, what each section should contain, what mandatory fields need to be populated, and what the professional conventions are. Given that structure, a language model can reliably populate it from transcribed field notes and photo observations.

If the required output document has no standard structure — every firm or jurisdiction uses a different format — the problem becomes dramatically harder. You’re not generating a report; you’re generating a document whose requirements you don’t know in advance. You can build a template system, but now you’re also in the enterprise customization business, which is a different (and much harder) company to run.

The Standards That Created Markets

Mental health documentation follows treatment note conventions established by clinical training standards: SOAP notes (Subjective, Objective, Assessment, Plan) or equivalent formats. Every therapist learned the same structure in training. The output format is consistent enough that a tool can generate it reliably. That’s why ten companies built in this space.

Home inspections follow InterNACHI and ASHI standards, implemented through state-specific regulations. Reports cover defined categories: foundation, roof, electrical, plumbing, HVAC, interior, exterior. Within each category, conditions are documented in standardized language. The output is consistent enough to build for. That’s why 8-10 companies built in this space.

ASTM E2018 defines the structure of property condition assessment reports. The standard specifies exactly what sections are required, what cost classification categories exist (Immediately Necessary, Short-term deferred 0-1yr, Long-term deferred 1-10yr), and what an Opinion of Probable Cost should include. Every PCA report produced by a reputable firm follows this structure. That’s why this might be the next market to develop.

The Evaluation Question

When you’re looking for professional documentation opportunities with this template — unstructured field input, structured required output, bad time ratio, high recurrence — the standard format question is the first filter after validating that the pain exists.

The question isn’t “does this profession write reports?” Most knowledge workers write reports. The question is: “does this profession write reports that all follow the same structure?”

If the answer is yes and the structure is defined externally (by a standards body, regulatory requirement, professional association, or industry convention), you have a tractable engineering problem: ingest the standard, build the generation pipeline, validate the output against the standard.

If the answer is yes but every firm uses its own idiosyncratic template, the engineering problem is much messier and the sales problem is much harder (every customer needs customization, which means high onboarding cost and ongoing support burden).

What This Means in Practice

Before the build decision on any field report automation tool, the right technical due diligence step is to read the relevant standard. Download ASTM E2018. Read the InterNACHI Standards of Practice. Find the clinical documentation guidelines from the relevant mental health association.

If you can read the standard and see a clear report structure with defined sections and mandatory elements — the output format is tractable. If the standard is vague about format, or varies significantly by jurisdiction, or doesn’t exist — the problem is harder than it looks.

The AI in these tools isn’t doing anything exotic. It’s filling in a template from unstructured input. The template is the hard part, and the template is only well-defined when someone external — a standards body, a regulator, a professional association — did the work of defining it.

Find the standards that already exist. Build the template that implements them. The AI is the last mile, not the main event.