An AI radar that turns one sentence into a meeting-ready target list

The situation

For a business development team, the target list is everything, and building a good one is slow, manual work. Finding the right companies in a market, then the actual decision-makers inside them, then working email addresses that will not bounce, then enough context to open a credible conversation, is days of an analyst’s time for a single region. By the time the list is ready, the market has moved and half the enthusiasm is gone.

Our client, a global management consultancy, wanted to compress that from weeks to minutes, without dropping to the quality of a bought contact list. The brief was demanding: the results had to be current, verifiable and specific enough to walk into a meeting with.

What we did

We built a lead-generation platform we call Radar. A user describes the client they want in plain English, for example “large developers and contractors building giga-projects in Saudi Arabia and the UAE”, and the system does the rest.

• Finds the right companies. It scans the market against the brief and shortlists real, recognisable organisations, backed by recent, dated evidence such as new funding, expansions and contract wins, not guesses.
• Finds the people to actually call. It identifies senior decision-makers inside each company, roughly eleven per company on average, so you reach the right person rather than a single generic name.
• Checks that they are real. Every profile is verified as a current person against public sources, and stale or wrong records are dropped automatically. In pilot, ninety per cent of people were confirmed real and current.
• Verifies the emails. Each address is checked before it is shown, so the list does not bounce.
• Writes the opening. For each company it drafts a short “why we can help, why now” brief, and a fuller, meeting-ready briefing pack on request.

Under the surface it is an AI system wrapped in careful engineering: a FastAPI and PostgreSQL backend, a modern web application, and a pipeline that grounds every claim in real, dated evidence and verifies contacts and emails before anything reaches the user. We designed it the way we design any system people stake decisions on, to show its working and fail visibly rather than confidently inventing a name.

The outcome

From a single sentence, the platform now returns a finished target list in about a minute rather than weeks. Across the first three regions switched on, it produced 98 target companies and 231 named decision-makers, with ninety per cent confirmed real and current and every shown email verified. One company went from a single request to a full, meeting-ready briefing pack, eighteen senior decision-makers identified, in minutes.

The result is business development that starts from evidence instead of a blank page: fresher leads, better-qualified contacts, and far less time lost to manual research. The same engine extends to new regions and industries by changing the brief, not the code.