A roadmap built to hit a sales number isn't a roadmap. It's a quota with a Gantt chart. And most technical roadmaps are built exactly that way, even when nobody says it out loud. The thinking underneath is lifecycle-refresh logic: the server is old, so replace it; the contract is up, so renew it; the budget exists, so spend it. Every item points back to revenue, because that's what the role is quietly rewarded for.
I don't build them that way. I optimize for one thing: what the client actually needs to build a foundation they can grow on. Sometimes that's a six-figure migration. Sometimes it's a no-cost change that quietly removes a risk they didn't know they were carrying. The dollar amount is an output, not a target.
Two real examples of what that looks like. A client was midway through migrating to a new ERP, run by a different vendor. There was nothing in it for us to sell — not our project, not our platform. But I could see that once that migration finished, several on-prem servers would have no reason to exist anymore. So the roadmap item was simple: track the migration, and when it lands, decommission the servers it makes redundant. Cost to the client: zero. Result: a smaller attack surface, less hardware to fail, and a lower support burden every month after.
The second one: managers at another client were typing their own admin credentials every time a staff member needed a software update approved. It was a daily friction point and a real security hole at the same time. We deployed a privileged-access tool that already fit inside what they had — no new licensing cost — that took the manager out of the loop and closed the gap. Better workflow, better security, no invoice.
Both of those come from a principle I drill into my team: we don't replace old things with new things by default. Swapping an old server for a new server doesn't reduce anything. It just resets the clock on the same liability. We optimize for elimination — increase scalability, decrease liability, increase security, and create value, ideally all in the same move. Eliminate servers whose services can live in SaaS. Eliminate on-prem identity where the business doesn't need it. Move line-of-business apps to the vendor's cloud. The best outcome usually isn't a newer version of the problem. It's the absence of the problem.
Here's why the $0 items matter more than they look. They're how you earn the right to make the expensive recommendation. When a client watches you put forward the path that bills them nothing — repeatedly, on purpose — they believe you when the six-figure project genuinely is the right call. That trust can't be bought or pitched. You earn it by occasionally costing yourself the sale and being visibly fine with it.
A roadmap should be measured by what it does for the business behind it, not by the size of the invoice stapled to it. Build it that way and the revenue still comes. It just comes after the trust, instead of trying to stand in for it.