7 Signs Your Company Needs to Optimize Its Processes

No company decides on a Tuesday that it is going to optimize processes. What happens is that something breaks: a client complains about a deadline, an audit leaves an observation, someone leaves and with them goes the only person who knew how something was done.

The problem is that those signals get read wrong. They are interpreted as a failure of a person, or of an area, and the answer is a meeting and a reminder email. Six months later it happens again. These are the seven signs that, in my experience, tell you the process is what fails, not the people.

1. The same task is done differently depending on who does it

This is the most common sign and the easiest to check: ask three people on the same team how they perform a specific task and write down the answers.

If they differ, it is not indiscipline. It is that the process was never defined, and each person built their own version from whatever their predecessor taught them. It works while nothing changes and it fails the moment someone new joins or volume rises.

2. Everything depends on one person

There is someone about whom everyone says "so-and-so knows that". When they go on holiday, a process slows down or stops.

That knowledge lives nowhere, and its value is deceptive: the company believes it has a key employee when what it really has is concentrated operational risk. In the regulated sector this is explicit: Agreement 011-2018 of the Superintendency of Banks requires identifying, measuring, mitigating, monitoring and controlling operational risk, and dependence on a single person is exactly that.

3. Nobody knows how long anything actually takes

You ask how long it takes to approve a request and you get a range ("between two and ten days") or a defensive answer ("it depends").

It does depend, of course. But if nobody can say the average, nobody can say whether it improved either. Without measurement there is no possible optimization: only opinions about whether things are going better or worse.

4. Rework is routine and nobody mentions it any more

Files come back because a field was missing. Invoices get corrected. Reports get redone. And nobody treats it as a problem because "it has always been this way".

Normalized rework is the most expensive waste there is, and the most invisible: it appears in no accounting line. It is paid in hours of expensive people doing twice what should have come out right the first time.

5. Growing hurts more than it should

The company wins a large client and instead of celebrating, the team panics. Every increase in volume requires hiring in the same proportion.

A well-designed process scales better than linearly: it absorbs growth without doubling headcount. If your cost curve rises at the same rate as revenue, the process is the ceiling.

6. Systems do not talk to each other and somebody bridges the gap

There is a person who exports from one system, adjusts in a spreadsheet and loads into another. Sometimes several people.

That work adds no value: it is generated by the lack of integration. And it introduces human error at exactly the point where correcting it is most expensive. Understanding the process should come before automating it, but when this symptom appears the integration conversation is already ripe.

7. Documenting is perceived as bureaucracy

When you propose documenting a process, the reaction is defensive: "we do not have time for that", "this is not a multinational".

That resistance almost always comes from a bad previous experience: a two-hundred-page manual nobody read. Documenting well is the opposite: the minimum needed for another person to run the process without asking.

Telling the real problem from the apparent one

The seven signs share a trap: the symptom points at one area, but the cause usually sits on the boundary between two.

The file that comes back for a missing field is not a problem of whoever reviews it; it is that nobody defined what information completes the previous step. The invoice that gets corrected is not billing's fault; the data arrived wrong from sales.

That is why process mapping starts with the handoffs between areas. That is where most problems live, and where nobody looks because there is no clear owner.

There is a simple test to know whether you are looking at the symptom or the cause: ask "why?" three times in a row. If the third answer still points at a person, you have not reached the bottom. If it points at a design decision — who approves what, in what order, with what information — that is the cause. People are rarely the problem; they are usually the ones holding a badly designed process together through extra effort.

Checklist: where to start

1. Pick one process, the one that hurts most. Not three.
2. Gather the people who run it, not their managers.
3. Draw how it is actually done today, not how it should be done.
4. Mark the handoffs between areas and the points where something waits.
5. Measure two things: end-to-end duration and how often it gets redone.
6. Identify the bottleneck. There is usually only one.
7. Change that and measure again after four weeks.
8. Document the improved process, not the previous one.

If by the end of point 5 you cannot put numbers on it, that is your first finding.

Common mistakes

Starting with the tool. Buying software before understanding the process automates the mess and makes it harder to change.

Redesigning without the people who execute. A process designed in a room without the people who do it is broken from the first week.

Attacking everything at once. Three processes in parallel with the same team means none of them moves.

Documenting before improving. It freezes the mess in a manual and adds the cost of maintaining it.

Measuring only at the end. Without a baseline there is no way to prove the improvement, and the project loses support exactly when it starts delivering.

A note on company size

This is not exclusive to large organizations. The Authority for Micro, Small and Medium Enterprises (AMPYME) runs formalization and business development programmes precisely because the jump from informal to formal, or from small to mid-sized, is where improvised processes stop working.

In thirty-person companies I have seen all seven symptoms at once. The difference with a three-hundred-person one is that there you can still fix it in months rather than years.

Where to go next

If you recognized three or more signs, the starting point is not buying a system: it is mapping one process and measuring it.

You can see how we approach this in process optimization, and if the problem is that there are many processes and none with an owner, the route is business BPM. To understand the full methodology, we wrote a guide on what BPM is.

Want a quick read on your situation? Book a free 15-minute assessment.