There are two ways to design a process: the old way and the new way. The old way is pre-AI. It's built around what a person can physically do: how long it takes someone to gather the information, spot the pattern, write it up, and get it in front of the right person. Every step is sized to a human's speed, whether anyone designed it that way on purpose or not. It also considers human and system interaction points, but that's about as technical as it gets. The new way flips who's actually running the process. AI-native process development is entirely different, and it's where the real ROI of AI shows up. In this new state, AI shouldn't be a tool bolted onto a human-built process. AI is the primary process holder, the one actually executing the sequence, and the human's role changes completely as a result.
If you want your commercial team relying on AI for more than the occasional task, actually having AI hold real pieces of how work gets done, you have to design the process itself differently from day one. This is where almost everyone gets it wrong on the first attempt. You cannot take the process you already have, the one built for a person to run, and hand pieces of it to AI, and expect the return you're picturing. The process itself was never built for AI to run. Giving a machine a piece of a human-shaped process doesn't change the shape of the process. It just makes the wrong shape faster.
Designing from the AI perspective means flipping who builds the first draft of the process, and the flip only works if you brief the AI properly, and you are clear on what needs to be human, vs what can be done by the AI tool. Before AI proposes anything, you have to tell it what the process is for: the outcome it has to produce, what separates a good result from a passable one, the constraints it can't design around, and the failure modes you already know from watching this work go wrong. That briefing is the first real step, and it's where the expertise moved. Do it well and AI will propose a sequence, decision points and handoffs built around what the outcome requires rather than what a person used to have to do by hand.
None of this means AI should run the whole thing untouched from day one. The goal isn't to hand AI a process and walk away; it's to decide, on purpose, exactly where humans stay in the loop for approval, and build those checkpoints into the process itself instead of bolting a review step on as an afterthought. That's not a sign a team isn't ready for AI-native process design. It's what AI-native process design looks like when it's done well. The starting question still can't be the old-way exercise with an extra question tacked on at the end. You don't map the current state, design a future state, and then ask where AI could speed up step four. It has to be different from the first minute: what value is this process going to create, what should it look like if AI is the one running most of it, and exactly where does a human have to stand.
This is the shift I want teams building now, not after their current process is running cleanly and they finally feel ready. Fourteen years inside Danaher's commercial organization taught me to always start with the value a process is supposed to create, and build backward from there. The old way starts with what a person can already do. The new way starts with the value you're creating, includes AI in building the process from day one, and puts humans exactly where judgment and approval are critical, deliberately, not because the team ran out of runway to automate further. They're not variations on the same question, and the teams building this muscle now, instead of waiting to feel ready, are the ones who'll actually get the return.
Curious to see what this looks like in practice? Part 2 of this issue walks you through a real case study of an AI native process, developed this way. Subscribe and follow along!
If this was useful, the rest of the Commercial Intelligence Report is one click away. Subscribe to the Commercial Intelligence Report, a weekly analysis on how industrial commercial teams are turning AI into measurable ROI, and the discipline that makes it stick.