OptimizeGEO Prompt & Topic Development Process

Purpose: Understand who they want to target, not the whole market. Key inputs extracted: * Priority regions * Competitor list * Target customer segments * Explicit non-targets to avoid (important filter) Why this matters: This defines the scope so we don’t generate prompts for irrelevant verticals (example, pharma, R&D, clinical labs in this case).

We avoid using the client’s internal terminology because it usually does not reflect how people actually search. We instead look at: * The jobs-to-be-done * The pain points that drive software adoption Typical categories to explore: * Compliance / documentation friction * Operational bottlenecks * Data integrity issues * Reporting requirements * Staffing + workload constraints

We search how practitioners talk in the wild. Use LLMs, Reddit, discussion boards, Q&A threads: * Search using workflow pain questions, not product names (example, “how to maintain chain of custody logs” instead of “LIMS software”). Outcome: * A list of real, verbatim questions showing how people actually describe their problems.

We don’t invent themes. We observe patterns: Example resulting themes: 1. Chain-of-custody & sample tracking 3. Instrument data-transfer issues 4. QA/QC review workflows 5. State/municipal reporting requirements 6. Turnaround-time bottlenecks These themes reflect recurring real-world pain not marketing categories.

Goal: Convert messy human phrasing → clean prompts that AI search engines respond to consistently. Normalization rules: * Keep the pain + workflow + constraint * Remove local context, emotion, anecdotes * Avoid vendor names, solution bias, or insider jargon This step produces the final prompt list for OptimizeGEO scanning.

A structured topics + prompts set, ready for measurement scans. This set: * Reflects actual market pain * Avoids branding bias * Aligns with target customer segments * Is generalized enough for AI platform visibility analysis

This approach avoids: * Guessing * Overfitting to client worldview * Feature-based prompts that have no real search demand Instead, we anchor prompts in: - Real conversations - Real workflows - Real regulatory + operational friction Which produces signal-rich visibility measurement data.