What it actually takes to become an AI-first business.

I sat down with Sara Maldon, who leads internal AI transformation at Make, at the last SaaStock USA in Austin. Make is the visual automation platform, the kind of company you'd expect to have a head start on this. They do. But what's interesting isn't that they're ahead, it's how they got there and what it cost them to figure it out.

She joined Make at a point where the brief was essentially: go figure out what AI means for us. No roadmap, no benchmark, no comparable. Just a mandate and a company to work with. That was three years ago. Most companies are only having that conversation now.

The thing that stayed with me from the conversation is the Samurai model. Make eventually landed on embedding one dedicated AI owner into every department, not a central AI team dictating from above, but people who understand both the business function and the tooling, sitting inside the teams that need to change. Seventy percent of their time goes on building high-impact projects tied to quantitative metrics, not time savings. Twenty percent on coaching others. The rest on maintenance.

It sounds straightforward. It took a year of Sara going department to department, finding quick wins, understanding what each function actually needed, before they had enough to build that structure properly. Most companies want to skip to the model without doing the groundwork first.

The framing she used that I found most useful: this isn't an AI project. It's a business transformation that happens to run on AI. For Make it's coupled with HR transformation, organisational design, how they hire, how they grow. The mandate they've been running with for the past year is doubling revenue with the same headcount. That's not an efficiency target, it's a restructuring of how the business operates.

On tools, they use what works: Cursor, Claude, Gamma, Whisper Flow, Synthesia. They use their own platform where it makes sense but aren't dogmatic about it. The objective is to become AI-first. The tooling serves that.

The most transferable advice for founders came when I asked about a SaaS company at five million ARR that hasn't made any moves yet. Sara’s answer: start with what annoys you. What part of your day feels like a waste of time? That's where AI tends to deliver immediate value. LinkedIn management, turning talks and interviews into assets your marketing team can actually use, getting context fast before important conversations. Not glamorous use cases, but they compound.

On leadership, her observation was simple. She hasn't seen a company succeed at this where leadership wasn't bought in, and she hasn't seen one succeed where the bottom-up piece was ignored. You need both. Most companies start with one and eventually figure out they need the other.

The full conversation is on The AI Revolution Show.

If you're not already subscribed, it's worth your time.