I wouldn't have a business without collaboration. Partnering with consultants has given me access to clients I'd never have reached alone. I'm genuinely grateful for those experiences. But here's what I've learned the hard way: it's like a marriage. You don't know who someone really is until the honeymoon phase ends. Everyone manages their worst instincts while things are smooth. The moment pressure hits - missed deadlines, scope questions, awkward client conversations - you find out what you're actually dealing with. As the builder, you're usually the one exposed. The consultant owns the client relationship. When loyalties are tested, they'll swing toward the client almost every time. That's rational - the client is their long-term income. But it means you can end up blindsided in meetings where the consultant is promising things you haven't costed, haven't agreed to and now can't push back on without looking difficult. It's happened to me more than once. I naively assumed partnership meant we were aligned. Instead I was left holding commitments I never made. Your only safety net is the contract. Roles. Scope. What happens when new work gets promised without agreement. Get all of it in writing before you start. Not after the first problem - BEFORE. I still regret the times I skipped this step. Don't repeat my mistakes.

Something I keep hearing from people getting into AI consulting... "Should I pick an industry or pick a problem?" Honestly, I remember having similar conversations years ago in consulting, long before AI was part of the discussion. The technology changes. The question doesn't. Over the last few months, I've spoken with quite a few people building AI consulting businesses. Most seem to fall into one of two camps. Some are trying to become "the AI person" for a specific industry. Others are trying to solve a specific operational problem across multiple industries. Document collection. Client onboarding. Lead qualification. Knowledge management. Follow-ups. After thinking about this quite a bit, I find myself leaning more toward the industry side. Not because problems are not important. They absolutely are. But I have learned that understanding the industry often helps you uncover the real problems much faster. When you spend enough time in a specific space, you start noticing things that are not obvious from the outside. You understand the language. You understand the workflows. You understand where delays are accepted as "normal" even though they create massive operational drag. And perhaps most importantly, clients feel understood. One practical lesson I've learned over the years: People rarely buy technology. They buy confidence. Confidence that you understand their world. Confidence that you have seen similar situations before. Confidence that the solution will fit how their business actually operates. The technology is just the delivery mechanism. That said, I can also see strong arguments for leading with a specific problem. Some operational challenges show up everywhere. A bookkeeping firm, a property management company, and a software services business may look completely different on paper. Yet all three might be losing hours every week chasing documents, following up on missing information, or manually moving data between systems. So I am not convinced there is a perfect answer.

I've been the technical half of several AI partnerships with consultants and marketing leads. Here's what I've learned about the role in practice. It's not what I assumed going in. I assumed I'd receive neatly packaged briefs. Build. Invoice. Move on. That almost never happens. In every meaningful partnership I've been part of, I've been in early meetings, contributing to audits, helping shape discovery and sometimes building proof of concepts unpaid just to help get the deal closed. The thing that surprised me most is how much value flows from having worked across multiple projects. I built a legal pack review tool - an AI solution for property legal packs. One of my consultant partners is now actively marketing that concept to their clients in the property auctions space. That idea came from me having already built it elsewhere. When you've delivered enough projects, you become a source of ideas your partners can sell. That's changed how I see my role entirely. I've also become the encourager in several of these relationships. Consultants face a lot of rejection. Deals take months. Self-doubt creeps in. I've found myself being the one saying "this is possible, I've done it before, keep pushing." That's not in any contract but it's kept deals alive. If you're doing all of this - generating sellable ideas, providing technical credibility, keeping morale up, building POCs - and still being treated like the hired hands waiting for instructions, something's wrong. I've been in that position and I let it slide for too long. Now I make sure the arrangement reflects the contribution. But I also hold my side - deliver well, document properly, make my partners look great in front of their clients. That's the balance. Also, not all consultants should be allowed the privilege of working with you.

He had a client meeting - bakery business looking at AI solutions. The night before, we had a call to talk through their requirements and plan the discovery. He was expecting to go into that meeting with notes and a conversation framework. I stayed up until 4am and sent him three working demos instead. - QR code 1: WhatsApp link where customers can place orders and ask questions - RAG system behind it trained on the client's products, pricing, and order process. - QR code 2: Voice agent - customer fills in details, gets an AI callback. - QR code 3: Live dashboard showing all interactions and orders in real time. By morning he had scannable, testable, working prototypes to put in front of the client. Not a pitch deck. Not a "here's what we could build." Actual proof. The client loved it. Hence "you are a beast" at 10:42am. I still don't know if this closes. I'm not getting ahead of myself. But this is what I've learned about being the builder in a partnership - if you can produce something tangible within reason, do it. Don't save your effort for after the contract is signed. A working demo removes doubt faster than any slide ever will. Granted, you are delivering but you are also making your partner look incredible in front of their client. That's what keeps deals moving and partnerships strong. Was 4am excessive? Maybe. But I wanted to leave nothing to chance and nothing to the imagination. That's my job in this.

A question that keeps coming up in AI consulting conversations... Especially from people who are just getting started. "If ChatGPT exists, tutorials are everywhere, and AI tools keep getting cheaper... why would anyone pay us for help?" Fair question. I think most of us have wrestled with it at some point. After spending the last several months building AI solutions and talking to business owners, I've noticed something interesting. Very few people are paying for access to AI. They're paying for clarity. Most business owners already know AI exists. Most have watched the videos. Many have tried ChatGPT. Some have even purchased AI tools. Yet they are still stuck. Not because they lack tools. Because they don't know: - which problem is worth solving first - what process should change - what should stay manual - where automation could actually create risk - how to connect all the moving pieces One lesson that surprised me: The projects that get traction are rarely the ones with the most sophisticated AI. They're usually the ones that remove a frustrating piece of work that people have quietly accepted as normal. A report that takes three hours. A client onboarding process with five unnecessary handoffs. A team member spending part of every day copying information between systems. The AI is often the easy part. Finding the right problem is usually the harder part. Curious how others here think about this. What do you believe clients are actually paying for when they hire someone for AI help?