Discovery meetings are slow when clients don’t know what they need. So we built an AI agent that does the legwork first: it researches the company, runs a conversational interview with the client via chatbot, and delivers a pain-point summary before you’ve even shaken hands. Less fumbling, more progress.
The discovery meeting problem
Let me paint a picture for you. You are in a discovery meeting with a new client. It’s going well, everyone is being jovial, they’re all on their second coffee of the day, the room is full of untapped potential. The beginning of a beautiful friendship, as it were.
But the meeting catches a snag. You and your company have all the capabilities to help this client, but the client doesn’t know exactly what they need help with. Ah. This could take all day. This could take multiple days to really get to the bottom of how this relationship will work, and what you could really offer them. You don’t sell a “one size fits all” solution, your business offerings are tailor made. The client doesn’t need an off-the-shelf product either, they just know they need support, but without knowing what the problem really is it’s going to take time to find it.
Unless, of course, you have an AI agent that has already debriefed the client, discovered all the pain points, marked any missing information, and now you are going to have a productive meeting in which progress is already being made.
How our AI discovery agent works
Well, look, I don’t mean to brag but we do have an agent that can do all of that. It was borne out of that exact issue. Here’s how it works:
Step 1: Automated company research
We start by giving the company’s name, website URL, and contact name to the agent which then conducts targeted research. The prompt for this step provides best practices for research, as well as how to use that research to generate productive interview questions.
Once it has completed the research, the agent provides a report with all the information found as well as generating a list of questions that could be asked in a debriefing interview. These questions, and the report, are then sent to our interviewing agent which contains context about Activate Intelligence and what we do as a company.
Step 2: Conversational client interview
This is where the client comes in. They are sent a unique link to a page with a chatbot. “Hi there,” the chatbot says, “Thanks for taking the time to debrief. First, can you confirm your name?” The client confirms their name and the interview begins.
But this is not a static interview using the list of questions from the research one by one. This is a dynamic conversation. The AI interview agent uses the list of questions and research as a guide, but it responds intelligently to the client. Does the client need to expand on something? The chatbot will ask. Has the client mentioned something interesting not in the research? The chatbot will follow up. Is the client busy and has mentioned the interview needs to wrap up soon? The chatbot will end when the client wants to end.
Although the interview is finite, the more complete and detailed the information, the better the discovery report and the easier it is to jump straight into the development phase.
Step 3: Pain point analysis and summary
Once the interview is complete, a summary is emailed to us detailing how the interview went and what pain points the client seemed to have, along with where we could potentially get started with AI solutions. We bypass a huge pain point ourselves by doing this.
Faster client onboarding, better outcomes
We all spend less time fumbling around looking for the issue, and we use this agent to get straight to the point.
It’s called the Debriefing Agent, it’s brand new, and we are all very excited about it. How about you give it a try? Leave a comment below or send us an email at intelligence@activate.co.uk
Leave a Reply