Where Should We Improve? Ask the Teams.
In the first four articles in this series, we widened the lens of Workforce Intelligence—from individuals, to teams, to the interactions between teams, and then to how organizational behavior changes over time.
That wider view raises a practical question:
When we want to improve how work gets done, where should we begin?
It can be tempting to look for a signal in the data: a delay, a growing queue, a repeated handoff, a change in communication. Those observations may be useful once we know what we are trying to understand. But they cannot reliably tell us where the most important improvement is waiting to be discovered.
The mistake is assuming the opportunity can be detected from the outside.
Ask the teams
A better starting point is a conversation with the people doing the work:
What is harder than it should be?
What takes too long?
Where does work get stuck or have to be redone?
What do you work around every day?
If you could change one thing about how this work gets done, what would it be?
And if the question is “Which team should we ask?”, the answer is all of them.
Every team has a view of the work it does and the conditions around it. Asking across the organization helps reveal where experiences connect, where they differ, and where one team’s difficulty may be another team’s unseen constraint. These conversations should not be a one-time survey or an occasional improvement exercise. They should be a continuous part of how the organization works.
The answers don’t assume the team is the problem. They invite the people closest to the work to describe what they experience.
That experience gives us a place to begin—not a complete diagnosis.
Follow the work to where the condition sits
Sometimes the difficulty is within the team. A decision may be unclear. Work may be started faster than it can be completed. A necessary skill or piece of information may be hard to access.
Sometimes the cause is upstream. The team may receive incomplete requests, changing priorities, or decisions later than the work requires.
Sometimes the difficulty is downstream. The team may complete its part, only for work to wait for another group’s capacity, approval, or process.
And sometimes the condition sits outside the organization entirely—with a customer, supplier, partner, regulator, or another organization.
A team’s experience tells us where to start asking questions. Following the work helps us understand where the condition actually sits. We need to look across the relevant boundaries and involve the people who understand each part.
That changes the question from “Which team needs to improve?” to:
What is making this work harder than it needs to be, and where does that condition actually sit?
What automation can—and cannot—do
Automation can help once the people involved have identified something worth understanding or changing. It can reduce repetitive steps, make agreed information easier to use, or help track an outcome the people involved have chosen to learn from.
But automation cannot reliably discover the most important improvement opportunity on its own. A delay does not explain why work is waiting. A pattern in system activity cannot tell us, by itself, whether a handoff is wasteful or necessary. More data does not remove the need to understand the work in context.
The useful role for automation is to support people as they investigate and improve—not to watch them in search of a problem, infer what they need, or decide what matters on their behalf.
Try a change and learn
Once the people involved have a shared understanding of the difficulty, they can decide what small change is worth trying.
Be clear about what you expect to improve and what else might be affected. Then observe what happens. Did the work become easier? Did it move faster? Did the change create a new burden somewhere else? What did the people doing and receiving the work learn?
The first explanation may be incomplete. The first change may not help. That is part of the process. Each attempt should improve the organization’s understanding of the work and help people decide what to try next.
Make improvement part of how the organization works
A continuous process cannot depend on a senior leader or specialist noticing a problem and asking the right questions. The organization needs to learn to ask them itself.
Teams should be able to raise difficulties, bring the right people together across boundaries, try changes they can make, and seek help when a constraint sits beyond their authority. Leaders still have an essential role: creating the conditions, time, and support for that work, and helping resolve constraints teams cannot address alone.
Over time, asking, learning, and improving can become part of the organization’s everyday practice—part of its DNA. The goal is not to launch another improvement program. It is to build an organization that keeps learning how its work can be better.
Keep measures in their place
Measures can help answer a question the people involved have chosen to investigate. But a measure is not the work itself—and it can become misleading when people are asked to optimize the number.
If the goal is fewer handoffs, necessary coordination may disappear from view. If the goal is shorter cycle time, people may close work before it is truly complete. If the goal is fewer interruptions, an important request may go unanswered.
This is the lesson commonly associated with Goodhart’s law: when a measure becomes a target, it can stop being a useful measure. Use measures to learn whether a change helped. Don’t turn them into quotas or scorecards for the people doing the work.
Improvement does not begin with detecting the right signal. It begins by asking all the teams what is too hard, what takes too long, and what gets in the way. Then follow the work far enough to understand where the condition sits, make a change with the people who can affect it, and learn from the result. Keep asking. Keep learning. Keep improving.
What would change if every team in your organization could raise a difficulty and help the organization learn from it—without waiting for someone else to ask?