by Serguey Shinder
We had a fortnight of work to scope and a meeting booked for it on the Wednesday. On the Tuesday evening, wanting to arrive prepared, I described the problem to a model and asked it for an implementation plan. It gave me one in about the time it took me to sit back in my chair. Six steps, sensibly ordered, with a rough shape of the interfaces and a note about migration.
It was good. That is not a complaint about the plan. I read it twice, changed two things, and brought it to the meeting as a starting point, and I said so plainly.
The meeting took nineteen minutes. Everyone read the document, agreed it looked reasonable, argued mildly about the naming in step four, and we left. I remember feeling that we had been efficient.
Three weeks later we hit the thing that was always going to be the hard part, which involved a downstream consumer nobody in the room owned and a data shape that could not be changed without a conversation with another department. The plan had not mentioned it, because I had not mentioned it, because I did not know. But two of the people in that meeting did know, and neither of them said so.
I have asked both of them about it since, and their answers were almost identical and entirely reasonable. They arrived expecting an empty room and a whiteboard. They found a coherent six-step document. The only way to introduce what they knew was to object to a plan that already existed, and objecting is a different and more expensive social act than contributing. So they read it, found nothing obviously wrong, and let the meeting end early, which everyone was pleased about.
The plan was not the problem. What the plan did was skip the part of planning that was actually doing the work. Those meetings are slow and a bit uncomfortable and full of half-formed suggestions, and I had always experienced that as waste. It turns out that the discomfort is the mechanism. People contribute what they know while the shape of the thing is still soft. Once there is a document, the same knowledge has to arrive as criticism, and most people, quite sensibly, will not pay that price for something they are only slightly worried about.
I still use the model for this. I have changed one thing. I do not bring the plan to the meeting. I bring the questions it raised, and the list of things I am unsure about, and the room fills the gaps in the way it always did.
Speed at the wrong moment is not speed. Arriving with an answer saves the hour the meeting would have taken and costs the three weeks that the hour would have prevented.
– Serguey Asael Shinder
Leave a Reply