When growth slows, the most visible problem is not always the limiting one. A full inbox can hide an evaluation backlog; a quiet sales team can hide weak demand. Find the constraint by following work through the business and identifying the step that limits completed, useful outcomes.
Define The Outcome And Draw The Path
Choose one business outcome before inspecting the process. For a service company, that might be completed work within its chosen service area. For a law firm, it might be accepted engagements the practice has capacity to serve. Write down what counts and what does not, then map the steps from first interest to that outcome.
Use the actual workflow, including waiting, reassignment and exceptions. A simple sequence of inquiry, contact, evaluation, commitment and delivery is enough to start. Ask the people doing the work to identify where they need information or decisions from someone else.
Look At Flow And Waiting Together
For each step, record how much work arrives, how much leaves and how much remains open. Add the age of waiting items. A queue that grows while the next step stays fully occupied is a useful signal, but investigate whether missing information, rework or batching causes the wait.
Do not confuse activity with capacity. A team can make many contact attempts while few conversations occur. Another can process many records by repeatedly returning them for missing information. Count completed transitions and the effort behind them, not just tasks touched.
Test Competing Explanations
Take a small, representative sample of completed and stalled work. Trace each item and ask what prevented the next useful action. Compare sources, days and teams where appropriate. Separate a lack of demand from poor fit, a contact problem from a review problem, and a review problem from insufficient delivery capacity.
Illustrative example: a service company receives 60 suitable requests in a week, completes 40 estimates and can schedule only 20 jobs. If scheduling capacity is the limiting factor, increasing inquiries may grow the waiting list. The team still needs to check whether the scheduled jobs are the right mix and whether estimates are accurate before drawing a final conclusion.
Run One Bounded Experiment
State the hypothesis in plain language: “If we remove this specific delay, more suitable work should reach the next step without reducing quality.” Assign an owner, a time window and a measure of success. Include a guardrail such as error rate, customer wait or staff workload.
For the illustrative service company, a test might protect scheduling time or remove an approval step that repeatedly delays confirmed jobs. Change one important variable where practical. If demand and staffing both change at the same time, record that limitation rather than claiming the experiment proved causation.
Recheck The System After The Change
A successful repair can move the constraint elsewhere. Review completed outcomes, queue age and quality after the test, then decide whether to continue, revise or stop. Preserve the before-and-after definitions so the comparison remains meaningful.
The output of this exercise should be a specific investment question: add qualified demand, improve contact, increase evaluation capacity, simplify a handoff or expand delivery. That is a stronger starting point for a growth discussion than a general request for more marketing.
Common Questions
How Much Data Do We Need To Begin?
Enough to trace actual examples and see whether the same delay recurs. Start with a bounded period and disclose its limitations. You can fix an obvious unowned handoff before you have a sophisticated forecasting system.