Manifesto

What the years can tell you

What we call experience is often time served, treated as proof of judgment. The years tell you how long someone has worked. They don't tell you how often that person's calls were right.

My own years in marketing made confidence easy to mistake for skill. From inside that confidence, asking for a check feels needless. The person making the call already feels qualified to make it.

In my record, I write down a decision and how sure I am before knowing the outcome, then check the result. There, I marked 65 calls 90% sure; 38 were right.

That is why I won't accept years as the credential, including my own. Otherwise, I am asking you to trust the same feeling I trust. I need something I can compare with what happened, not an account made tidy afterward.

I need the expectation written down.

What I had to remove

Written down was also why I trusted numbers I should have checked. I audited ten frameworks I had written up for my own work. Nine leaned on numbers nobody could check: 91.8%, 97.1%, 169 out of 184.

None had a log, a scoring rule, or a second person checking. I hadn't meant to invent anything. I had written what sounded right, then treated the written version as a reason to believe it.

Those figures looked precise, but I couldn't show where that precision came from. I couldn't ask anyone else to check the work behind them. There was no work they could inspect.

I took the numbers out.

A number on a page isn't evidence just by being there. I have to be able to trace it back to something.

A long career can build real skill. Practice can improve judgment, and asking for evidence need not mean dismissing that skill. I am not asking anyone to start again as though their years taught them nothing.

But without a check, I could feel 90% sure while getting far fewer calls right. From inside, rising confidence feels like getting better at the job. My own certainty cannot settle which one is happening.

What I will answer to

Before the answer arrives, I write down what I expect and how sure I am.

My own calls stay private at first, because a watched record tends to fill with safe numbers.

Client work answers to a record the client can see. Before client work goes out, I put my expectation and my confidence on that record. Then I ship the work, and the client checks the result against what I wrote down first.

Those rules ask different things of me. In private, I need enough honesty to risk being wrong on paper. With a client, I owe them access to the evidence, not just my reading of it.

If tenure is all you want your authority to rest on, we won't fit. I trust a documented track record over a confident veteran, including when the veteran is me.

The values I work by have to bear that test, too. You should be able to ask what supports them without accepting my word first.

Even experience has to wait for the answer.