Selected work.

More and more buying journeys start with a question to an AI assistant rather than a search engine. When someone asks ChatGPT, Gemini or Perplexity for a recommendation, the answer they get shapes the decision — but brands have almost no visibility into how, or whether, they show up in those answers.

Traditional SEO tools measure rankings on search results pages. They say nothing about what an AI model actually says when a prospective customer asks about your category, your competitors or you by name.

Klarity monitors how brands are represented across the major AI answer engines. It runs realistic prompts on a schedule, captures the answers, and turns them into a clear picture of visibility, sentiment and share of voice over time.

• Automated prompt tracking across ChatGPT, Gemini and Perplexity
• Visibility and share-of-voice scoring against competitors
• Sentiment and accuracy monitoring on brand mentions
• Trends over time, with alerts when things change

Klarity is built as a custom web app on foundations designed to scale: a scheduled pipeline that queries multiple AI providers, normalises very different response formats into a consistent structure, and stores the results for analysis and comparison.

A tight scope and weekly iterations took us from concept to a working product quickly, while senior product judgement kept the data model, provider abstractions and reliability solid — the parts that are painful to get wrong once real customers depend on them.