Working Harder, Shipping Less: Find the One Constraint Choking Your Business
When effort is high but output is flat, one step is holding back the whole system. Here is how to find it and stop feeding the wrong parts.
The stuck moment. You run an HVAC company. Calls are up, techs are working ten-hour days, and revenue is stuck at $180K a month. You hire two more techs. Output barely moves. You add a marketing push. Still flat. Everyone is busy, nothing is faster, and you cannot see where the money is leaking. That is a bottleneck: one step is capping the entire system, and every dollar you spend anywhere else just piles up in front of it.
The recipe. Find the one constraint before you spend another dime.
- Map the flow. Write the path a job takes end to end: lead comes in, gets booked, gets dispatched, work gets done, invoice goes out, cash lands. Five to eight steps, no more.
- Find the pile-up. At each step, ask where work waits. The bottleneck is the step with a queue in front of it and idle capacity behind it. In HVAC it is often dispatch or booking, not the techs.
- Confirm with one number. Measure throughput at the suspected step (jobs completed per day). If lifting only that step would raise total output, you found it. If not, keep looking.
- Feed it, do not starve it. Protect the constraint. Cut anything that makes it wait. Do not add capacity anywhere else until this step can keep up.
The AI move. Paste this into Claude or ChatGPT:
Role: You are an operations analyst who finds throughput bottlenecks.
Context: My business is [type]. Monthly revenue is stuck at [$X] despite [effort you added: hires, ads, hours]. Here is how a job flows from lead to cash: [list your 5-8 steps]. Rough numbers at each step: [any counts, wait times, or capacity you know].
Task: Identify the single step most likely constraining total output. Explain what pile-up or idle capacity points to it.
Format: (1) The one bottleneck, (2) the evidence, (3) the cheapest test to confirm it in one week, (4) three moves to relieve it that cost little or nothing, (5) what NOT to spend on until it is fixed.
Constraints: Assume I want throughput up without new hires first. Be specific to my numbers. If my data is too thin to call it, tell me the one metric to pull.
What it is worth. Finding the real constraint turns a $20K hire that does nothing into a scheduling fix that lifts output 15 to 30 percent next month, from capacity you already pay for.