The Person Whose Job This Shift Threatened

I have written the Product Builder thesis a dozen ways: the role is changing, building got cheap, the artifact is being devalued, judgment is the scarce thing now. All of it is true and all of it is abstract until it has a face. This is the face. It is about the best PM on my team, and it is the ship story I am least proud of the handling of, which is exactly why it belongs here.

The short version

The best document writer on my team was threatened by the Product Builder shift, not because she was bad but because she was excellent at the one thing AI made cheap. Her craft was the clearest spec and the most thorough strategy doc in the org, and that craft was the currency her reputation and her sense of her own value were built on. When a model started producing clean documents for free, the thing she was best at stopped being scarce. I handled it badly the first time by softening the message to protect the relationship, which really protected my comfort and stole two quarters of her runway. When I finally said it straight, her first question was why I had not told her sooner, and I did not have a good answer. The lesson is not about tooling. It is that the transition is a people problem, the people are owed the truth early, and kindness that delays the truth is just comfort for the manager, paid for by the person.

The best PM I had

She was, on the old scoreboard, my strongest PM. When a decision needed to be made legible, she made it legible better than anyone. Her specs were the ones other PMs copied. Her strategy docs were the ones that survived exec review without a rewrite. If you needed a messy, contested problem turned into a clear document that aligned six stakeholders, she was who you gave it to, and she knew it, and she had every right to. A decade of being excellent had earned her that identity.

That identity is what the shift came for. Not her competence, her currency. The thing I wrote about artifacts being promotion packets was not a theory to her. She had built a career, correctly, on producing the best version of the exact artifact that was now being generated for free. The skill was real and the skill was suddenly abundant, and abundance is what deflates value. She was collateralized in a currency the market was devaluing in real time, and she could feel it before she could name it.

The tell

The tell was that she worked harder at the thing that was deflating. As the ground shifted, she did not pull back on documents. She doubled down. The specs got longer and more thorough and more polished, because that was the lever that had always worked, and when a smart person feels the floor moving, the instinct is to do more of what made them valuable. It is the most human response and it is exactly the wrong one, because you cannot out-effort a change in what is scarce. A more perfect version of a deflating artifact is still a deflating artifact.

I could see it, and here is my failure: I did not say it clearly. I told myself I was being kind. I gave feedback around the edges, praised the judgment inside the documents, gently suggested she try more prototyping, and let her keep running toward the thing that was disappearing. I optimized the conversation for how comfortable it was for me to have, and I called that respect for her.

The conversation I stole from her

Two quarters later, when the shift had gone from a breeze to a wind, I finally had the conversation straight. I told her the truth: the thing she was best at was no longer the scarce thing, the role was moving to judgment expressed through prototypes and evals and adoption, and I believed she could make the move but only if she started now and stopped pouring herself into documents.

She was not surprised by the content. She was hurt by the timing. Her first question was why I had not told her this two quarters ago, when she had more runway and less panic. I did not have an answer that survived being said out loud. I had known. I had softened it. And the softening had cost her the one thing I could not give back, which was time to make the change with dignity instead of under pressure.

That question, why didn't you tell me sooner, is the one I carry from this. I had confused kindness with comfort. The honest conversation was uncomfortable for me, so I delayed it, and I dressed the delay up as consideration for her. It was not. It was consideration for me, and she paid for it. The cruelty in these situations is almost never the honest conversation. It is the one you do not have in time.

What actually worked

The part I did get right, eventually, is the part worth passing on.

I separated her worth from the deflating skill, out loud and specifically. The thing that made her the best spec writer was not the specs. It was that she thought clearly, and clear thinking is the scarce input in the new role too. The judgment was always the valuable part; the document was just the medium it used to travel in. That reframe mattered, because a threatened expert's worst fear is that they are being told they are worthless, and she was not. She was being told her medium changed, and her actual talent, the judgment, transferred.

Then I gave her real support, not a pep talk. Time carved out to learn to prototype, cover to be bad at it while she learned, a smaller scope for a quarter so she could rebuild her evidence base on the new currency, and a concrete picture of what the scarce skills actually were so she was not aiming in the dark. The reskilling was real work and it was not guaranteed, and I stopped pretending otherwise. Honesty about the odds turned out to be more respectful than false reassurance, because it treated her like someone who could handle the truth, which she could.

She made the move, for what it is worth. Not everyone does, and I want to be careful not to wrap this in a bow, because the version where the person does not make it is common and also not a moral failure on their part. But she made it, and the thing that made it possible was the judgment underneath the documents, which had been the real asset the whole time. What almost sank her was not a lack of ability. It was the identity built around the artifact, and the extra two quarters I let her spend defending it.

Why this is a leadership problem, not a tooling one

Every conversation about this transition wants to be about tools, because tools are easier. Which model, which harness, which prototyping stack. But the hard part of the Product Builder shift is not standing up Claude Code. It is sitting across from your best person and telling them that the thing they are best at is no longer the thing, in time for them to do something about it, without making them feel disposable in the process. That is the director-layer pressure and the reskilling challenge and the human cost, all in one room, and no tool touches it.

The leaders who handle this transition well are not the ones with the best stack. They are the ones willing to have the uncomfortable conversation early, and to hold the person's dignity and the hard truth at the same time, which is harder than either one alone.

Try this week

Find the person on your team whose craft is the thing AI just made cheap. You know who it is. It is often your strongest performer on the old scoreboard, which is what makes it hard.

Have the honest conversation now, while they still have runway. Separate their worth from the deflating skill, name the scarce skills plainly, and offer real support to bridge the two. Do not soften it into uselessness to protect your own comfort, and do not mistake the delay for kindness. The person is owed the truth early enough to use it. That is the whole job this week, and it is worth more than any tool you could roll out instead.