The Best Process Improvement Is Deletion

Most operators improve a process by adding to it. A step gets added because something went wrong once. A form gets added because a step got skipped. An approval gets added because a form got filled out badly. Nobody ever calls a meeting to remove the approval, so the process only grows in one direction, and after a few years you are running a company whose real product is internal compliance.

I have run this brokerage since 2002, across two states and two very different markets, and the single highest-return operational move I know is subtraction. Not automation. Not optimization. Removal.

Why adding feels safer than it is

Adding a step is emotionally cheap. It looks like responsibility. When something breaks, adding a checkpoint is the move that reads as taking it seriously, and it costs nothing today because the cost is paid later, in other people's time, forever.

Removing a step feels like risk, because you have to name the thing you are willing to let go wrong. That is the actual work. If you cannot say out loud what failure you are accepting, you have not thought about the step hard enough to keep it either.

The asymmetry is what compounds. A step that adds four minutes and runs eighty times a month costs you a work week a year. Nobody expenses it, so nobody sees it.

The three questions

Before I keep any step in a process, it has to survive three questions.

What breaks if this disappears tomorrow? Not what could theoretically break. What breaks, specifically, and how would we know. If the honest answer is "we would find out eventually and fix it," the step is insurance against something cheap.

Who is this actually protecting? Half of the steps in a mature company protect a decision-maker's comfort rather than the client's outcome. Those are the expensive ones, because they look like diligence.

Was this built for a problem we still have? Most process is a scar. It was the right response to a real incident, and the incident has not recurred in years, partly because everyone learned from it and partly because the business changed shape. The scar stays because removing it feels like forgetting the lesson.

What it looks like in practice

Three examples from our own operation, all of them subtractions, all of them boring.

We had an approval gate sitting in the middle of our content operation: work would get produced, then wait for a sign-off, then publish. The sign-off almost never changed the work. What it reliably changed was the calendar, turning a piece finished on Tuesday into a Friday, or the following week, or nothing at all if the week got busy. We deleted the gate and moved the standard upstream into the brief. Quality held. The calendar stopped slipping, which was the actual problem.

We had duplicate scheduled jobs doing overlapping versions of the same weekly work, each one created at a moment when the previous one seemed insufficient. Three schedules is not three times the output. It is one output and two sources of confusion about which is authoritative. We paused two and kept one.

We had a duplicate business listing on Google under an old company name, splitting our reviews and our search authority between two identities. Deleting it was not an improvement to our marketing. It was the removal of an obstacle we had been marketing around for years.

None of those three required a new tool, a new hire, or a budget. All three took less than a day.

Where subtraction is wrong

It is not a universal solvent. Steps that exist because a regulator or a fiduciary duty requires them are not candidates, and you do not get credit for speed on the transaction you have to unwind. Neither are the steps that prevent a failure that is catastrophic and irreversible. Those are cheap insurance no matter how rarely they pay out.

The test is reversibility. Delete freely where the failure is recoverable and visible. Leave it alone where the failure is permanent or silent. Most process fails the first test and gets protected as if it passed the second.

The Monday version

Take your most-run internal process, the one that touches the most transactions, and write down every step in it. Not the version in the manual. The version people actually perform, which you will only get by asking the person who performs it.

Then find the step that exists because of one incident nobody can date, and delete it. One step, this week. Watch what happens for thirty days. In my experience you will discover two things: nothing breaks, and the person who performs that process has been carrying a list of four more.

The bottleneck is rarely capacity. It is usually the accumulated residue of every reasonable decision you have ever made.

Carlos Alex Rozwadowski, who runs CENTURY 21 Premier Real Estate as Alex Rozwadowski, is the CEO of the brokerage, with offices in Columbus, Georgia and Auburn, Alabama. A U.S. Army Ranger veteran and former Rothschild and Co analyst, he has led the company since 2002.