The holes were on purpose.

I didn't find the deeper issue because I was asked to. I found it because the loud problem got fixed first, and something quieter kept surfacing underneath it.

I was part of a team brought in to recover a stalled website relaunch, on a hard 100-day deadline ahead of a high-visibility event. There was no time to look past the brief. We executed.

Portfolio mapping

Once the relaunch shipped, a smaller, unscoped thread was still nagging at me: whether the site's navigation, built around internal product names, actually matched how buyers searched. I built a data model to check it, months of search terms, on-site queries, and competitor navigation, scrubbed and weighted. The data confirmed a real mismatch. That got funded internally as portfolio mapping. But the data only got us halfway there.

The other half came from talking to people. A researcher colleague and I interviewed over a dozen engineers, some of whom worked for competitors, about how they actually found answers when a spec sheet didn't have one. A generation of experienced engineers, the ones with a shelf of product binders and years of trade-show handshakes, had largely left the industry after a dramatic drop in demand. The people who replaced them were younger, without the binders or the relationships, and when they hit a question a PDF didn't answer, assuming they even knew which PDF to look for, they went looking in forums, asking strangers who might have worked for the competition the kind of thing a rep used to answer face to face.

The company wasn't in the business of selling equipment. It was in the business of being the resource an engineer trusted for the length of their career, whether or not that engineer ever bought from them.

Nobody asked us to find that. We were mining for a taxonomy fix and came back with the actual argument for the year's roadmap.

The workshop

That finding earned a bigger mandate: more than twenty people across four divisions, brought together to build a shared roadmap. I built a customer journey map and left deliberate holes in it, printed it at wall size, and over three days asked each division to fill in the blanks for its piece of the customer's path.

None of them were wrong about their own piece. That was the problem. Sales was right about lead quality, support was right about repeat questions, marketing was right about content and navigation, and each of them, taken alone, hadn't seen how their piece connected to everyone else's.

I wasn't the subject matter expert at that table. My job was closer to interrogation than instruction: feeding what one group said back to another, asking questions that sometimes were mildly unsettling on purpose, and making sure the least confident voices still got their ideas onto the board. I ran the scoring so ideas were judged on their merit rather than on who'd said them, so a junior team member's read on the customer could outscore a senior stakeholder's, if it earned it.

Each day opened the same way, a strawman hypothesis for what the roadmap should prioritize, put up in front of everyone. Each day closed the same way too, the same hypothesis, back on the wall, with instructions to sleep on it. Every morning started with the same question: does this still hold, or does it need to change. By the end of the third day, the group had rewritten it enough times that the final version wasn't mine anymore. It was theirs, built on the trust argument the fieldwork had surfaced months earlier.

I was the primary author of the digital roadmap that came out of this work. The client told me later it made it to the CEO's desk.

A note on anonymization: client and market specifics in this case study have been removed deliberately. That's standard practice for protecting client confidentiality, not a comment on the sensitivity of the story itself.