The story changes a little at each handoff between writer, designer, illustrator and developer, and every change is locally reasonable. Record constraints, not conclusions, and keep one person accountable for the argument at every stage.
The pattern
The narrative is agreed. Everyone signs off. Then it is written by one person, designed by another, illustrated by a third and built by a fourth.
At the end, the finished site says something slightly different from what was agreed, and nobody can point to the moment it changed.
Why it breaks
It changed at four handoffs, a little at each one, and every change was locally reasonable.
The writer needed a shorter heading, so a conditional claim became an absolute one. The designer needed a balanced three-column row, so a two-step sequence gained a third step that is not really a step. The illustrator needed the figure to read at slide size, so a mechanism with a causal direction became a symmetrical diagram. The developer needed the section to work on a phone, so a sequence that revealed in order became four blocks that arrive at once.
None of those people made an error in their own craft. Each of them made a decision with consequences one discipline downstream, and no one in that chain was accountable for the argument end to end.
What changes
Two things.
First, the structure has to record decisions, not just conclusions. “The platform section leads” is a conclusion. “The platform section leads, the three steps are causal and must be read in order, and step two cannot be shown before step one” is a decision that survives translation. Constraints travel; summaries do not.
Second, someone has to be accountable for the argument at every stage, with enough craft to see the consequence coming. That is a different requirement from good communication between specialists. A person who can see that a symmetrical layout will destroy a causal claim has to be looking at the layout.
The practical result
The finished work says what was agreed, and the review conversation changes character. Instead of “this feels off”, stakeholders can say “step two is appearing before step one”, because the constraint was written down and can be checked.
Where the constraint genuinely cannot survive a format, that becomes an explicit decision rather than a silent loss. Sometimes the right answer is that the format changes.
The distance between what a company decided to say and what its finished materials actually say is one of the least examined costs in investor communication. It is almost never one large error. It is four small ones that nobody owned.
