Growth Doesn’t Usually Break Everywhere
Growth rarely exposes a lack of effort first. It exposes a process that can no longer carry the load.
A lead waits too long for a response. A proposal needs rework. Jobs reach the schedule missing information. The team works harder, yet the customer experience and margin both tighten.
My response is process triage. It is a disciplined way to identify the one broken process creating the most drag right now, stabilize it, and make a permanent improvement instead of a temporary save.
That focus matters. The Theory of Constraints holds that every organization has at least one limiting factor, which may be a policy or a physical step. In a service business, it may be an approval rule, a handoff, a missing standard, or a decision trapped with one leader.
Triage is not an excuse to ignore everything else. It is how I keep a leadership team from spreading its attention across ten symptoms and solving none of them.
What Process Triage Is—and Is Not
Process triage begins with a simple question: Where is work getting stuck in a way that materially affects the business?
The answer is not always the loudest complaint. A scheduling backlog may look like a scheduling problem when incomplete job information is entering the schedule upstream.
I define a process as the repeatable path that turns an input into an outcome. It has a start, finish, handoffs, decisions, standards, and an accountable owner. When the same failure returns, the process deserves attention—not only the person caught in it.
Triage has three jobs:
- Protect the customer, team, and cash flow from immediate harm.
- Find the smallest number of causes that explain the recurring failure.
- Install a better way of working that is owned, measured, and checked.
It is not a giant transformation project or a blame exercise. And it is not a license to redesign work before we can describe the problem.
The Process-Triage Framework
Name the failure in operational terms
Start with facts, not labels. “Communication is bad” is a frustration, not a problem statement. “New web leads received after 3:00 p.m. are not contacted until the next business day” is a problem a team can investigate.
Write a one-sentence condition with the process, failure, affected outcome, and time window. Then gather a small sample of real tickets, jobs, orders, approvals, or requests. See where work entered, waited, returned, and ended.
Use a few practical measures rather than a dashboard full of noise:
- Volume entering the process and elapsed time to finish.
- Work waiting at handoffs, plus rework or exceptions.
- Customer impact, lost capacity, or delayed revenue.
The goal is visibility: enough to choose the bottleneck wisely.
Rank impact before choosing a project
Most companies have more than one broken process. Triage requires a choice.
I rank candidates by the size and frequency of harm, the people affected, the risk of waiting, and our ability to act this quarter. A daily failure that blocks revenue or delivery outranks an irritating internal inconvenience.
Ask these questions in the leadership conversation:
- Which fix would most improve flow in the next 90 days?
- Where does work accumulate or repeatedly bounce backward?
- Which failure creates the most customer pain, margin leakage, capacity loss, or risk?
- What do actual records show repeatedly?
Do not pick the executive team’s most interesting project. Pick the process whose improvement releases capacity or protects the most important outcome.
Stabilize the process before you perfect it
When a process is hurting customers or creating a costly backlog, put a temporary control in place. This is containment, not the final answer.
A containment step might be a daily review of unassigned work, an intake checklist, an escalation path for aging requests, or a short-term queue owner. Define the trigger, actor, and end date.
Containment buys time to investigate without normalizing heroics. If it requires a leader forever, it is a new bottleneck wearing a different shirt.
Map the work as it actually happens
Get the people who do the work together and walk one real item from start to finish. Map the trigger, steps, decisions, handoffs, waiting points, and finish.
I am looking for the gap between the documented and lived process. Where does someone chase information? Where does a handoff rely on memory? Where does work wait for an unavailable person?
This makes assumptions visible. A process map is a shared picture of the work, built from evidence.
Find causes, not convenient culprits
Once the failure is clear, ask why it happens, then why that condition exists. Keep going until the answer points to a changeable decision rule, system design, training gap, capacity limit, unclear standard, missing information, or accountability gap.
The American Society for Quality defines a root cause as a factor behind a nonconformance that should be permanently eliminated through process improvement. “Someone forgot” may describe an event, but seldom explains why the process allowed a repeat.
Use a small cross-functional group: people closest to the process and someone able to authorize the correction. ASQ notes that analysis is generally stronger when people work together and those removing the cause are part of the work.
Separate causes into three buckets:
- Direct causes are the conditions that produced this failure.
- Contributing causes make the failure more likely or more damaging.
- Root causes are the changeable conditions that, if removed, make recurrence less likely.
Do not demand one dramatic root cause. The test is whether a change addresses a real cause and can be verified in daily operations.
Assign one accountable owner and a clear countermeasure
Every triage effort needs a single owner for the outcome. That person owns the standard, metric, exceptions, follow-through, and escalation—not every task.
Define the countermeasure: what changes, who acts, where it happens, what “done” means, and which measure should move. Make the standard easy to find. Update the checklist, template, routing rule, training, and meeting rhythm that support it.
This is where EOS can be useful without becoming an EOS lecture. Keep the bottleneck visible on the EOS Issues List, then use IDS—Identify, Discuss, Solve to reach a decision and owner. Use the EOS Accountability Chart as a test: is accountability for the result clear and understood?
Verify, standardize, and watch for the next constraint
A solution is not complete when it is announced. It is complete when the process performs differently over time.
Set a review date and compare measures with the baseline. Talk to users and sample completed work. Look for new failure modes or a constraint that moved downstream. The FDA’s root-cause guidance makes the key distinction: finding a cause alone does not prevent recurrence; action and verification are necessary.
If the result holds, make the control normal work. If not, return to the evidence. Do not protect a favorite solution from facts.
A Practical 30-Minute Start
You do not need a retreat. Put the leadership team and process owner in a room for 30 minutes.
First, list recurring failures and choose one using impact, urgency, and evidence. Next, write the problem statement and trace one recent item. Finally, assign an owner, a containment action, the first data to collect, and the next review.
Leave with one sentence everyone can repeat: “We are fixing this process because it is causing this outcome, and this person owns the next action.”
Common Ways Triage Fails
Triage fails when leaders solve by volume. Adding meetings, software, or people before locating the constraint can increase cost and complexity while leaving flow unchanged.
It also fails when a team mistakes a person for the process. Hold people to standards, but build a process that makes the right action obvious and feasible. Accountability without a workable system is frustration disguised as management.
Finally, it fails when the scorecard disappears after launch. A recurrence check shows whether the countermeasure is real.
The Bottom Line
Growth does not require you to fix every imperfection at once. It requires the discipline to find the broken process doing the most damage, stabilize it, understand why it fails, give one person clear ownership, and verify that the new standard holds.
That is process triage. Focus on the constraint in front of you. Make the work easier to do right. Then move to the next bottleneck with a stronger operating system than you had before.
