I have seen countless examples of companies filling their roadmaps with AI projects. A list of initiatives: automate this workflow, add a copilot here, reduce headcount in that function. The results are real. The efficiency gains show up in the numbers. And within 12 months, every one of your competitors will have the same list.
That's the trap. AI projects are not a strategy. They're table stakes. The companies that pull ahead won't be the ones that adopted the most tools. They'll be the ones that asked a harder question: if AI changes what work is, what does our company need to become?
That question demands a structural answer, and most companies aren't asking it.
What the Structural Answer Looks Like
We stopped asking "which workflows can we automate?" and started asking "what does the software company of the future actually look like?" Who works there? How is it organized? What do the people do that the AI doesn't?
Once you ask that question, the org chart looks different.
AI projects change what your tools do. Structural change changes who owns what, how work flows, and what you are asking your people to do. You can execute a project list well and end up with a more efficient version of the same organization, but it is not a durable outcome.
Process change runs into the same ceiling. Kill your standups, shorten your planning cycles, move faster. All reasonable first moves. But if the underlying structure doesn't change (who is accountable for what, how teams are composed, how decisions get made), you end up executing the wrong structure faster.
The companies worth studying aren't the ones with the most AI projects. They're the ones that look fundamentally different than they did three years ago, and are growing because of it.
Where to Start
If you run a software company and you're building an AI roadmap, the useful question isn't "what should be on the list?" It's: what would our company look like if we built it from scratch today, knowing what AI can do?
Start there. The projects will follow. But the structure has to come first, or the projects are just expensive table stakes.