Standardizing a Bad Process Only Makes It Easier to Repeat | Observations from the Shop Floor
Michael Gardner, COO
Over the years, I have learned to be careful when calling a process bad, because most bad processes didn’t start that way.
They usually started because someone was trying to solve a real problem. Maybe the software couldn’t do something the business needed. Maybe a particular customer required special handling. Maybe the company adapted a process around the capabilities or limitations of an employee. Sometimes a workaround was simply the fastest way to keep orders moving when there wasn’t a better alternative.
Whatever the reason, it worked, so people kept doing it.
Years pass. Employees change, customers change, and technology changes. The original problem may disappear completely, but the process created to solve it remains. Eventually, the organization remembers how to perform the process but no longer remembers why the process exists.
That is when things get interesting.
When People Become Part of the System
I’ve spent much of my career around businesses where people are remarkably good at making things work.
Someone knows that a certain type of order needs to be handled differently. Someone else knows that information coming from one system has to be interpreted before it can be entered into another. Another person knows which field really matters even though the procedure says something different.
None of this necessarily appears on a flowchart – yet orders ship, customers get taken care of, invoices go out, and the business keeps moving. From the outside, the process appears to work, and technically, it does.
What can be difficult to see is how much human effort is required to make it work.
People fill the gaps between systems, correct incomplete information, recognize exceptions, and translate what one part of the business means into what another part needs. Over time, those accommodations can become so familiar that nobody thinks of them as accommodations anymore. They are simply the process.
Success Can Hide the Problem
This creates an interesting challenge for successful companies because the better their people become at working around a problem, the harder the underlying problem can be to see.
An experienced employee may resolve in minutes something that would stop someone unfamiliar with the process for hours. Another employee may recognize immediately that something doesn’t look right and correct it before anyone downstream ever knows there was a problem.
Those are valuable people doing valuable work. But their ability to compensate for weaknesses in a process can also hide the cost of the process itself. Take away the person who knows what to do and suddenly everyone realizes how much knowledge was living outside the system.
That is often when a company decides it needs to standardize, and that’s where we need to be careful.
Before We Standardize
Standardization is powerful because it creates consistency. However, consistency and improvement aren’t necessarily the same thing.
If we document every step of an existing workaround, train everyone to perform it the same way, and then build technology around it, we may have created a very consistent process without improving anything. We have simply made the workaround permanent.
I’ve seen this happen during software implementations. Everyone understandably wants the new system to accommodate the way the company operates today, so we document what people do, recreate forms, reproduce reports, configure workflows, and automate steps.
Then one day somebody asks the question that probably should have been asked at the beginning: Why are we doing this?
Sometimes there is a very good answer.
Sometimes the answer begins with the name of someone who retired years ago.
Technology Makes This Even More Important
This question becomes even more important as technology gets better.
For years, automation generally required programmers, consultants, project plans, and considerable investment. That friction naturally limited how much of a company’s processes could be automated. AI is beginning to remove some of that friction by allowing technology to read information, interpret documents, recognize patterns, move data between systems, recommend actions, and perform work that previously required people.
That creates enormous opportunity, but it also creates a new risk.
If we don’t understand why a process exists before asking technology to perform it, we may simply teach technology to execute an outdated workaround faster. The same principle applies whether we are implementing an ERP system, building an interface between two applications, creating a workflow, or introducing AI.
Technology is a tool for solving problems, but understanding the problem still comes first.
What Are Our People Actually Doing?
Before automating a manual process, I’ve learned that one of the most useful things we can do is spend time watching the people who actually perform it. We shouldn’t just document which buttons they press. We need to notice where they stop, what causes them to question something, what they routinely correct, and what information they look for when something doesn’t make sense.
Those moments often reveal something far more important than the documented procedure. They reveal the judgment that has been quietly holding the process together.
Understanding that judgment helps us decide what should be standardized, what should be automated, what knowledge needs to be preserved, and what should disappear altogether.
Final Observation
Organizations naturally preserve the things that help them succeed, and that’s usually a strength. But processes can outlive the problems they were created to solve, particularly when good people have spent years quietly making those processes work.
Ironically, their success can make an inefficient process look surprisingly healthy.
So before we document a process, standardize it, build it into the next system, or teach AI how to perform it, perhaps the most important question is also the simplest: Why does this process exist?
If we understand that, we can make an informed decision about what deserves to be preserved and what no longer does.
Because standardizing a bad process doesn’t make it better. It only makes it easier to repeat.