Finding the Bottleneck in Your Process: Where to Measure to Know What Is Stopping Things
Back to blog
product·September 11, 2026·4 min read·By Yehonatan Saadia

Finding the Bottleneck in Your Process: Where to Measure to Know What Is Stopping Things

A bottleneck is not the hardest step but the one work piles up in front of. How to find it without tools, why it moves after you fix it, and what not to do.

Key takeaways

  • Look for a pile, not for effort. Wherever things wait, that is the bottleneck.
  • Improving any other step changes nothing; it only grows the pile.
  • After you fix it, the bottleneck moves somewhere else. That is success, not failure.
  • The most common bottleneck in a small business is one person who alone can approve.

A bottleneck is not the step that feels heaviest but the step work piles up in front of. That is a practical difference: someone searching by feel usually fixes the noisiest step, while the real delay sits one step earlier and says nothing.

Finding it without tools

Five checks you can run this week:

The checkWhat you are looking for
Where there is a pilePhysical or digital - a queue, a folder, a lengthening list
Who gets asked the most questionsA person everyone is waiting on
Where most items say "waiting for..."Statuses that stall in the same place
What is always running flat outA step that is never idle
Where waiting time exceeds working timeThat gap is the definition itself

The last row is the sharpest test: when work that takes ten minutes waits two days, the problem is not the ten minutes.

Why improving the wrong step does not help

A process produces output at the pace of its slowest step, so speeding up any other step only grows the pile in front of it. It sounds theoretical and is very tangible: doubling sales speed in a business where installation is the bottleneck produces customers waiting longer, not higher revenue.

The practical consequence: before investing in any improvement, ask whether it touches the bottleneck. If not, the saving will be real in hours and invisible in output - which is exactly the gap that produces the feeling that "we improved and nothing changed".

The four common bottlenecks in a small business

  • One person who approves everything. The owner, in most cases.
  • A step needing information that comes from outside - a customer, a supplier, an authority.
  • A single resource - a vehicle, a machine, a room, a software seat.
  • A decision that was never defined, so every case needs a discussion.

The first is the most common and usually the easiest to fix - not by hiring but by defining authority: what can be approved without asking, up to what amount, under what conditions. The expansion is in approval limits and controls.

What to do once you have found it

Not necessarily add people. Four options, in order of cost:

  1. Take load off the bottleneck - move anything through it that does not have to go through it.
  2. Prepare the work before it arrives, so it wastes no time on completions.
  3. Split - separate simple cases from complex ones and give them a short route.
  4. Expand - a person, equipment, or a licence.

Point two is cheap and surprising: in most human bottlenecks a large share of the time goes on clarifications - missing information, an unclear file, a question that could have been answered in advance. Improving the input form into the bottleneck is often the highest-return change in the entire process.

The bottleneck moves - and that is fine

After a successful fix, the delay relocates. That does not mean the fix failed but that it worked: the whole process is faster, and now a different step is the slowest.

So repeat the check quarterly rather than assuming what was true six months ago is true now. Growing businesses see this especially: the bottleneck in a three-person business is entirely different from the one in the same business with seven - which is the same logic as the wider cycle described in a 90-day efficiency plan.

What not to do

  • Do not speed everyone up hoping it helps - it grows piles.
  • Do not attribute the bottleneck to a person. It is nearly always a process, not a person.
  • Do not solve it with overtime. That hides the problem and makes it more expensive.
  • Do not measure load instead of measuring waiting - they are not the same thing.

The second matters especially in a small business: when the bottleneck is the owner, it is easy to read as personal criticism. In reality it signals that the process is built around one person, and that is exactly what limits growth.

What does it look like when the owner is the bottleneck?

It is the most common situation and it is almost never deliberate. It forms gradually: at first the owner did everything, then handed parts over - but the final approval stayed with them, and so did pricing, and so did the difficult customer call.

The test is simple: for one week, record every time somebody is waiting on the owner for an answer, and for how long. The resulting number is almost always a surprise - and in many businesses it turns out half the waiting is on decisions that could have been defined in advance: approving a discount up to a threshold, approving a credit, or approving a small purchase.

The fix is not giving up control but moving the decision into a rule rather than a conversation. A written rule lets somebody else decide the way the owner would have decided, and it also produces a short list of the exceptions that genuinely need to reach them.

Measuring waiting without a system

No tools required. Three methods that work in a small business: mark on each item or task the date it arrived at the step; record in one spreadsheet every time somebody says "I'm waiting on"; or simply photograph the pile at the end of each day for a week and see whether it grows.

The third sounds primitive and is particularly effective, because a pile that grew five days running is evidence nobody can argue with. And when the pile is digital - an email folder, a task list, an order queue - just count the items at the same time every day.

Sources

#bottleneck#process#measurement#efficiency#operations#אוטומציה לעסקים

Frequently asked questions

How long do you need to measure?

A week is enough for most businesses, sometimes two days. A genuine bottleneck is visible quickly, because the pile in front of it exists constantly rather than only on a busy day. If nothing shows in a week, you may have seasonal load rather than a permanent bottleneck.

What if there are two bottlenecks?

Usually there is only one real one, and the second is a consequence. Fix the obvious one and measure again - the second often disappears on its own, because it came from the pile created upstream.

Does automation solve a bottleneck?

Sometimes, depending on what the bottleneck is. Automating a technical step helps a great deal; automating a step requiring judgement usually does not, though preparing its input does. That is also why understanding the bottleneck should precede choosing a tool.

How do you know it worked?

With exactly the same measurement: how long from the start of the process to the end, and how large the pile in front of the step is. Two numbers measured before and after say more than any impression, and they also tell you where the next bottleneck has moved to.

Keep reading

Related service

MVP Development

Turn an idea into a validated product in weeks, not months.

Learn more

About the author

Yehonatan Saadia

Freelance automation, web & MVP developer

I'm Yehonatan Saadia, a senior developer who builds business automation, custom websites, and MVPs for small and mid-sized companies across the US, Europe, and Israel. These guides come from real client work, not theory.

Work with me

Have a project like this?

Tell me what you're trying to automate or build and I'll tell you the fastest reliable way to ship it.