Imagine a four lane highway. Traffic is moving at seventy miles per hour. Suddenly, the road narrows to a single lane. It does not matter how fast the cars were going before the merge or how wide the road becomes after it. The speed of the entire system is dictated solely by that one narrow point.
This is a Bottleneck. In a business context, a bottleneck is a point of congestion in a production system. It is the single step in a process that has the lowest capacity and therefore limits the total output of the entire organization.
For a founder, understanding bottlenecks is the difference between being busy and being productive. You can work eighty hours a week optimizing your marketing, but if your engineering team cannot ship the product, your marketing efforts are wasted. The bottleneck determines the speed of the company.
The Theory of Constraints
#This concept was popularized by Eliyahu Goldratt in his book “The Goal.” The core premise is simple but profound. Any improvement made anywhere other than the bottleneck is an illusion.
If you have a factory line with three machines, and Machine B is the slow one, making Machine A faster does not help. It actually hurts. It just creates a bigger pile of inventory in front of Machine B.
In a startup, this manifests in “work in progress” (WIP). If your sales team is closing ten deals a week, but your onboarding team can only set up five accounts a week, you do not need more salespeople. You need to fix the onboarding process. Hiring more salespeople in this scenario is burning cash to create customer dissatisfaction.
Finding the Pile
#Identifying the bottleneck is usually easier than fixing it. You just have to look for the pile.
Where is the work waiting?
- Is it a stack of unreviewed code waiting for the CTO?
- Is it a queue of support tickets waiting for an agent?
- Is it a list of expense reports waiting for the founder’s signature?
The resource with the pile in front of it is the bottleneck. Startups often try to solve this by yelling at the people managing the pile to work harder. This rarely works. You cannot outwork a structural constraint.
The Founder Trap
#In early stage companies, the bottleneck is almost always the founder.
This happens because you insist on approving every decision. You want to see every social media post. You want to interview every candidate. You want to review every line of code.
While this ensures quality, it destroys velocity. You become the single lane on the highway. To fix this, you must engage in what Goldratt calls “Elevating the Constraint.”
First, you exploit the bottleneck. This means you ensure the founder is not doing low value work. You stop scheduling your own meetings. You stop fixing the printer.
Second, you subordinate everything else. The team stops asking you for approval on minor items. You give them a budget and autonomy.
The Moving Target
#Fixing a bottleneck is satisfying, but it comes with a catch. As soon as you fix one bottleneck, another one appears elsewhere.
If you fix the engineering bottleneck and start shipping features twice as fast, the bottleneck will likely move to the QA team who cannot test that fast. If you fix QA, it might move to Customer Support.
This is the endless game of operations. You are constantly hunting for the new constraint. The goal is not to eliminate all bottlenecks, as that is impossible. The goal is to ensure the bottleneck is never in a place that threatens the survival of the business.