Flow Audit

What the company appears to be: Flow Audit is a self-reported workflow data analysis tool that processes exported workflow data (CSV, XLSX, JSON) locally using deterministic rules. It optionally applies an AI layer for interpretation while preserving user privacy by not sending raw data to AI models.

What changed: The project was built as a hackathon submission during OpenAI Build Week, with a focus on privacy-preserving deterministic analysis and optional AI interpretation.

Single most important open question: Does Flow Audit have any commercial traction or evidence of use beyond the author's own development?

The description states that Flow Audit:

• Analyzes exported workflow data locally using deterministic rules
• Supports CSV, XLSX, and JSON exports
• Normalizes data into a common structure
• Computes reproducible workflow metrics covering assignment coverage, workflow states, review distribution, activity dates, due dates, data quality, and lexical title similarity
• Optionally uses an AI layer (Google Gemini) for interpretation after validating findings
• Operates entirely in the browser with deterministic analysis
• Uses Cloudflare Workers to securely manage AI interactions

Evidence: The author's own write-up.

Confidence: Low. This is a self-reported description of a hackathon project, not evidence of a product in use or commercial adoption.

The description states that Flow Audit:

• Respects the distinction between data collection and interpretation
• Focuses on deterministic measurement before AI interpretation
• Is built with privacy as a core principle
• Separates local analysis from optional AI interpretation to ensure reproducibility and trust
• Was developed during OpenAI Build Week, suggesting it is in early-stage development

Evidence: The author's own write-up.

Confidence: Low. This is a self-positioning statement, not evidence of market traction or customer validation.

The description does not state:

• Who the target customer is
• What specific user personas are intended
• Whether there is an identified ICP (Ideal Customer Profile)

Evidence: Not evidenced.

Confidence: Very low. No mention of customers, user types, or buyer personas.

The description does not state:

• How the product would be monetized
• What pricing structure exists or is planned
• Whether there are any paid features or tiers

Evidence: Not evidenced.

Confidence: Very low. No indication of business model or pricing.

The description states that Flow Audit:

• Was built with Next.js, TypeScript, Cloudflare Workers, and the Google Gemini API
• Uses a two-layer architecture: deterministic analysis in the browser and optional AI interpretation via Cloudflare Worker
• Operates entirely in the browser for deterministic analysis
• Sends only validated findings to AI models
• Uses Papa Parse, SheetJS, and IndexedDB for data handling
• Was deployed on Vercel with Cloudflare Workers
• Used Codex and GPT-5.6 as development collaborators (not part of the deployed product)
• Is designed to be privacy-conscious

Evidence: The author's own write-up.

Confidence: Medium. Technical details are provided, but no evidence of production use or scalability.

The description states:

• Flow Audit was built during OpenAI Build Week (a hackathon)
• It is currently an MVP
• No mention of users, customers, revenue, or adoption
• The author notes that the next steps include generating downloadable PDF reports, supporting additional platforms, and improving comparisons across exports

Evidence: The author's own write-up.

Confidence: Very low. No evidence of traction, customers, or revenue.

The description does not state:

• Who the competitors are
• What similar tools exist in the market
• How Flow Audit differentiates from existing solutions

Evidence: Not evidenced.

Confidence: Very low. No competitive analysis or positioning against other tools.

Inferences based on the description:

• The project is a hackathon MVP with no commercial traction, suggesting it may not be ready for market
• The author is a single individual (team size: 1), which raises questions about scalability and long-term maintenance
• No evidence of revenue, customers, or product-market fit
• The focus on privacy-preserving AI interpretation may limit its appeal to users who don’t prioritize such features

Evidence: Inferred from the description.

Confidence: Medium. These are logical inferences, not stated facts.

1. What is the intended customer segment and how did you identify them?
2. How do you plan to monetize Flow Audit?
3. Have you validated demand for this tool with potential users?
4. What are your plans for scaling beyond a single-person development effort?
5. Are there any existing customers or pilot programs?
6. What is the roadmap for expanding support for workflow platforms beyond CSV/XLSX/JSON?