There's a concept from operations management called the Theory of Constraints. It states that every system has one constraint—the bottleneck that limits throughput. No matter how much you improve everything else, the system output is determined by that one constraint.
Most founders ignore this. They see ten problems and try to fix all ten. They optimize the whole system. They spread resources across everything. Progress stalls because they're not actually addressing what's holding them back.
If you want to move fast, you need to find the constraint and ignore everything else.
The Constraint Is Usually Obvious
It's obvious once you look. The problem is most founders don't look. They're too busy solving secondary problems.
Example: A SaaS founder is stuck at $2M ARR. She comes to me saying she wants to optimize her retention (currently 85%), improve her product (feels slow), build more features, and scale her sales team. Four directions. All worth doing eventually. But they're not the constraint.
The constraint is that her average customer is worth $8K annually, and 80% of her revenue comes from five customers. Her entire growth is locked behind selling to large enterprises. Her constraint is not product or retention. Her constraint is: she can't sell to more large enterprises. Her small sales team can close maybe two per quarter. That's the ceiling.
Every other optimization is secondary. You could double her product quality and retention wouldn't change the constraint. You could add ten features and the constraint remains. She's limited by sales velocity, nothing else.
Once she understood that, everything changed. She didn't optimize retention. She didn't build new features. She hired a sales engineer, created better sales processes, and built more content for enterprise buyers. Sales velocity doubled. Revenue doubled. The constraint shifted.
How to Find Your Constraint
Look at your unit economics. Where does your profit come from? What's the most expensive part of your business? What's the slowest part? If customer acquisition takes three months and retention takes three weeks, acquisition is your constraint.
Ask yourself: what one thing, if it changed, would change everything? If I could double sales velocity, would that matter? Yes. If I could improve product quality by 50%, would that move the needle? Probably not. Sales velocity is the constraint.
Look at what's preventing growth right now. You want to 3x next year. What would have to happen? If the answer is "we'd need more customers," your constraint is acquisition. If the answer is "we'd need those customers to stay longer," your constraint is retention. If the answer is "we'd need bigger customers," your constraint is deal size.
Ask your team. Ask your salespeople what's stopping them from closing more deals. Ask your product team what's blocking progress. Ask your customer success team what's causing churn. The constraint is usually something they'll tell you if you ask directly.
The Wrong Way to Fix the Constraint
Most founders throw money at constraints. Your constraint is customer acquisition? Buy more ads. Hire more salespeople. Build a partnership. All reasonable, all temporary.
The problem is that these tactics hit diminishing returns. You can't scale acquisition linearly by spending more. At some point, the underlying system breaks. You need cheaper or better customer acquisition, not more of the same.
This is where most founders get stuck. They identify the constraint correctly but "fix" it by doing more volume. That works until it doesn't.
The Right Way to Fix the Constraint
Systematize the constraint. If acquisition is your constraint and you're dependent on founder relationships, you can't scale by hiring salespeople. You need to systematize how you acquire customers. Document the process. Build templates. Create qualification criteria. Once the process is systematic, then you can hire people to execute it.
Remove friction from the constraint. If your sales cycle is six months, that's your constraint. You're only dealing with a handful of prospects at any given time. Can you shorten the cycle? Remove discovery calls? Build self-serve qualification? Every week you shorten the cycle is more sales cycles per year.
Build leverage into the constraint. If support is your constraint—customer success has 80% of revenue concentrated in their hands and they're maxed out—the solution is not hiring more support. It's building support leverage. Can you build templates? Self-serve? Community? Cohort-based onboarding? Anything that generates leverage without linear hiring.
Change the constraint entirely. Sometimes the best move is to shift to a different constraint. If customer acquisition is your constraint and it's incredibly difficult and expensive, maybe your constraint is actually your positioning. If you repositioned, would you attract different customers? Could you acquire them more easily?
What Happens After You Fix It
Once you fix your constraint, something important happens: a different constraint emerges. This is by definition. The constraint moves. It's like squeezing a balloon—release pressure in one place and the bulge appears somewhere else.
This is actually good. It means you're making progress. After you double acquisition capacity, maybe your fulfillment process becomes the constraint. After you fix that, maybe your ability to support more customers becomes the constraint. The constraint migrates.
The question at each stage is the same: what's the constraint now? Fix that, only that, and ignore everything else.
Why This Matters
The founders who move fast are not the ones who are good at optimization. They're the ones who are ruthlessly focused on the constraint. They don't split attention across multiple initiatives. They don't try to improve the whole system. They identify one bottleneck and remove it.
This is how you get disproportionate returns on effort. While everyone else is spreading resources across ten projects, you're concentrating on one. While they're trying to improve everything incrementally, you're getting 10x improvement in the thing that actually matters.
The constraint is your leverage point. Find it and push it, and the system moves. Everything else is noise.