Most got replaced by the next thing that showed up with the same urgency.

The pattern repeats often enough that I’ve stopped moving at its pace. What I pay attention to instead is what doesn’t change when the platform does.

What I’ve watched go wrong

I’ve watched teams rebuild their entire online presence three times in a decade, chasing the platform instead of building the foundation underneath it. Each rebuild started from scratch. Each one repeated the same conversations about voice, about hierarchy, about what the content was actually trying to do. Each one cost more than it should have, because none of the thinking carried forward from the last one.

The platforms weren’t the problem. Nothing underneath them was solid enough to survive the move.

What actually holds up

Every platform shift moves fast — new tools show up promising to outrun the last ones, and most teams feel pressure to keep pace or fall behind. Through all of that churn, four things have stayed constant for me:

  • Clear design guidelines. Not a style guide that lives in someone’s drawer. A working reference with documented tokens, type rules, component logic, and voice principles — written down clearly enough that the next person in can use it without guessing.
  • Sensible information architecture. Content organized around what people are actually trying to do, not around how the org chart is structured or how the previous CMS happened to work.
  • Content that knows what job it’s doing. Every page, every section, every block earning its place instead of filling space because the template had a slot for it.
  • Brand voice documented well enough that anyone can use it without breaking the system — a new hire, a freelancer, a custom GPT. Specific enough to be usable. Defined enough to hold up when the person who built it isn’t in the room.

These four things hold up regardless of where the work gets built next. They’re what let you change platforms without starting over, what makes a redesign an evolution instead of a rebuild, and what plug into AI tools once those become part of how the team works. Same foundation, more layers on top.

What this means for how I work

The work I do today runs on the same fundamentals it ran on twenty years ago. The job they have to do has expanded — but the fundamentals themselves haven’t moved, even as everything around them accelerates.

When I take on a new client, the conversation starts there: not with what platform we’ll use or which tools to consider, but with what foundation needs to be in place so whatever we build is still useful in five years. Sometimes that means rebuilding the design guidelines before the site. Sometimes it means restructuring the information architecture before any new pages get added. Sometimes it means documenting the voice properly so the team can stop guessing every time they write something. The work looks different from one client to the next; the reasoning underneath it stays the same.

The part worth investing in

If you’re trying to decide where to put your budget, put it underneath the platform, not on top of it. Not the platform of the moment, not the tool getting attention this quarter — the foundation that carries forward when the platform changes again, because it will, probably faster than last time.

That’s the part of the work that holds up. That’s the part worth investing in.