Case study
AI for a membership team that researches every lead.
Our client is a US-based tech NGO. Its members are technology companies that work together on a shared online safety mission. The membership development team finds, researches and brings in new member companies. We are building AI into that work, inside the ChatGPT Business workspace the team already pays for.
- Client
- US-based tech NGO
- Team
- Membership development
- Tools
- ChatGPT Business, Google Drive, their CRM
The problem
Every prospective member needs research before anyone reaches out: what the company does, where it stands on online safety, and whether it meets the membership requirements. That research was done by hand, one company at a time.
Emails and posts also had to sound like the organisation, whoever on the team was writing them.
What we built, in two phases
We started with the team, not the tools. Three training sessions and two hands-on writing sessions came first, all built on the team's real tasks. The tools came after.
Phase 1
- Writing skills: Four ChatGPT skills, including a brand voice built from the organisation's own guide.
- Personal voice: Drafts that sound like the person sending them.
- Discovery agent: Researches a company from public sources and drafts a brief in Google Drive.
- Second brain: Holds the team's knowledge so ChatGPT answers from it.
Phase 2
- Company intelligence: A structured brief on each applicant, checked against a fit checklist.
- Enriched discovery: More data sources behind each prospect brief.
- Contact mapping: The right person to talk to at each company.
- CRM dashboard: The whole pipeline in one place.
How a new lead moves now
Each step of the membership pipeline has an AI helper. A person checks the work at every handover.
- Find the prospect (Discovery agent)
- Research the company (Company intelligence)
- Check the fit (Fit checklist)
- Find the right contact (Contact mapping)
- Write the first email (Voice skills)
- Track the pipeline (CRM dashboard)
A person reviews every brief and every email before it goes out.
How we worked
- No new subscriptions: Everything runs on the paid tools the team already had.
- Public data only: Company research uses public information only.
- People decide: The agents draft. The team reviews and sends.
- Tested on real work: Each skill was tried live: an email to a current member, a first email to a prospect, and a LinkedIn post.
What changed
From our conversations with the team, the biggest change is how much more they now see about each prospect: a much richer profile of the company, and a clearer idea of who to reach out to and how.
Each prospect starts from a drafted brief instead of a blank page, and first drafts come out in the organisation's voice.
Want something like this for your team?
Tell us about the work that takes your team the most time. We will tell you honestly whether AI can help. Email sales@chazrt.com or book a call.