Lee McCabe on fixing the data before adding AI
Lee McCabe is Managing Partner at Claymore Partners, which builds commercial and digital infrastructure for private equity-backed portfolio companies. This conversation explores why reliable data should come before AI, which tasks AI can take on, and how customers are arriving with a view already formed.
AI is only useful when the underlying commercial data is reliable: a messy CRM can help a business be wrong faster. McCabe uses AI for research, data pulls and first drafts, while keeping the opinions and perspective that should come from the person.
What’s the quick origin story of your business, and what makes what you do genuinely different from the other options your customers are looking at?
I spent years as an Operating Partner at AEA Investors, working across about 45 portfolio companies. Before that, Facebook, Alibaba, Expedia and eBay. I watched the same film over and over. A sponsor buys a business, hires a consultant, gets a 90-page deck, and nothing changes. The CRM is a graveyard. Nobody knows what a lead is worth. Then everyone blames marketing.
So I started Claymore Partners to build the thing that’s actually missing: the commercial and digital plumbing. Data first. Marketing is an output of that, not the thesis.
The difference is simple. We don’t leave a deck. We leave a system that still works after we’ve gone home. Most consultants are paid to be needed. We’d rather be redundant.
Where does most of your new business come from today, and where do you wish it came from?
We practise what we preach. Building lead generation systems for PE-backed businesses and then sitting by the phone waiting for referrals would be a bit rich. So we built our own engine first, on data, the same way we build them for clients.
We know who we want to work with: the sponsors and operating partners who own the kinds of businesses we fix. We track who engages, score what that interest is worth and follow up when the timing is right, not when someone remembers to. Content, outbound and pipeline all run off the same data. No guesswork, no “brand awareness” line in the budget that nobody can explain.
Where do I wish it came from? More of the same, earlier in the deal. I’d rather be in the room at diligence than at the 100-day plan, when the sponsor has already promised the IC a number nobody has checked.
What’s one thing you’ve actually handed over to AI in the last year, and one thing you tried it on and went back to doing the human way?
Handed over: the first draft of almost everything. Research, data pulls, content drafts, tracking who has moved where across private equity. Work that used to need a junior analyst and a lot of coffee now runs before I’ve finished mine.
Took back: anything where the point is that it came from me. AI writes like a keen intern who has read every LinkedIn post ever published and learnt nothing from them. It loves a tidy list and a big finish. Clients can smell it. So the opinions stay mine. The machine gets the grunt work.
Have you noticed a change in how customers find you or what they already know before they reach out? Is anyone arriving through AI assistants like ChatGPT yet?
Yes. People arrive already knowing what we do, what I think, and which of my posts they disagree with. The first call used to be a pitch. Now it’s an argument. Much better.
ChatGPT is starting to send people too. Which is mildly terrifying, because it means a language model has an opinion about my firm and I had no say in it. Most PE-backed companies have no idea what these tools say about them. They’re still paying an agency to rank for keywords nobody types any more.
If you’re not deliberately feeding the machines, someone else is feeding them for you.
If you were advising someone in your industry on all of this for the next 12 months, what would you tell them to actually do, and what would you tell them to ignore?
Do: fix your data. It’s boring and unglamorous, and it’s the only thing that makes any AI tool worth the licence fee. If your CRM is a mess, AI will just help you be wrong faster. Work out what a lead is actually worth. Then decide what to automate.
Ignore: anyone selling you an “AI strategy” who can’t tell you your customer acquisition cost last quarter. Ignore the vendor demos. Ignore the board member who read one article on a plane. And ignore the idea that AI will replace your sales team. It will replace the bad half. The good half will just get more done.
Most of this year’s AI pilots will be binned by next summer. Make sure yours isn’t one of them.
Thank you to Lee McCabe and the team at Claymore Partners for sharing what actually worked with Leaders Perception readers.
Want to share your playbook?
Leaders Perception publishes founder and operator interviews every week. A few questions, about five minutes, no calls. It is free and everyone we accept gets published.