Perspectives · The A. de Verteuil Framework · Axiom I · Environment

After the Specialist Leaves

Alex de Verteuil ·

A supported first run tells you what a team can do with a specialist in the room.

A lab team completes its first run on a new diagnostic instrument. An applications specialist is there to guide setup, answer questions, and help the team recover when something unexpected happens. The run succeeds. Training is marked complete.

A week later, the same team tries again during a normal shift. The person who attended training is covering another bench. A colleague can follow the setup instructions, but when the instrument displays an unfamiliar message, they’re unsure whether to continue or stop. The run waits while someone calls support.

Both observations matter. The first tells us the team can complete the workflow with guided support. The second tells us what the team has to work with once the rollout becomes part of its day.

Axiom I. The environment shapes what people can reliably do.

What did the first run depend on?

In Mission Precedes Operations, I started with a question. What does this process need to accomplish?

For this rollout, suppose the goal is for the lab to complete a reliable run during a normal shift, with staff able to recognize when an issue needs escalation. That goal gives us a way to evaluate the training, the instrument, and the workflow around it.

During the first run, the specialist may have provided more than instruction. They may have recognized a warning immediately, known where to find the relevant guidance, or made a judgment that kept the work moving. The lab benefited from all of that. The question for the rollout team is where those capabilities will live after the specialist leaves.

Some belong in training. Some belong in instructions staff can consult while doing the work. Some belong in the way the instrument communicates what happened. And some issues need a clear route to an expert. You can decide where to invest once you know what the specialist was supplying and what the lab needs during routine use.

The same handoff happens between teams

Consider a development team transferring an analytical method to a receiving QC lab. The people who developed the method may know which steps allow for judgment and which conditions require them to stop and escalate. They’ve had time to build that knowledge through repeated use.

The receiving lab starts with the method, the training, and the procedures available to it. If analysts repeatedly pause at the same step to ask what an instruction means, the sending and receiving teams can look at that moment together. What did the analyst need to know? Where would they reasonably look for it? Who had the authority to resolve the question? Which issues required formal review?

That inquiry can lead to a specific change. The team might clarify an instruction, revise how it trains receiving analysts, or make the escalation path easier to follow. Each choice addresses a different part of the environment in which the analyst has to work.

Measure what happens under normal conditions

For the diagnostics rollout, I’d start with three observations.

  • Time to first successful independent run. How long after installation does it take a site to complete the workflow without live guidance?
  • Where support is needed. Which steps generate requests, and what information or judgment is the user looking for?
  • Specialist hours per site after training. How much follow-up work goes into helping a site reach routine independent use, where it runs the workflow during normal work without ongoing specialist help?

I’d read these together. A decline in support requests could reflect clearer onboarding. It could also reflect sites postponing runs. Time to first successful independent run helps distinguish between those possibilities. The support requests show where to look next.

I built the first Hidden Work Simulator scenario around this question. Two hypothetical rollouts follow the same installation schedule. As you change how often sites need help, how much sites differ from one another, and how much specialist time is available, you can see how many sites reach routine independent use, how much time goes to follow-up, and which requests carry into the next week.

That distinction matters when deciding where to put time and money. If a recurring problem can be resolved at the point of use, specialist time can go toward the cases that need expert judgment, or toward helping the next set of customers get started.

The next time a rollout goes well with an expert in the room, ask one more question before calling it complete. What did the expert make possible, and how will the team do that work on a normal shift?