There's a seductive idea in process improvement: "Let's just automate it." The logic feels airtight — if a task is slow and repetitive, a machine or software should do it faster and more consistently. And sometimes that's exactly right.
But here's the trap: automating a broken process doesn't fix it. It makes it break faster.
The Amplification Problem
Automation is an amplifier. It takes whatever your process does and does it at scale, at speed, without judgment. If your process is standardized, consistent, and well-understood, automation amplifies efficiency. If your process is chaotic, inconsistent, and full of workarounds, automation amplifies chaos.
Consider a data entry process where three people do the same task three different ways. One uses a spreadsheet template, one copies from email, one types from memory. Each handles exceptions differently. Now automate it with RPA. Which version did you automate? And what happens when the bot hits an exception that one human handled intuitively but nobody documented?
The bot crashes. Or worse — it silently produces garbage at 100x the speed a human could.
Standardize First: The Prerequisite
Before any automation initiative, you need answers to three questions:
1. Is the process documented? Not in someone's head — actually written down. Every step, every decision point, every exception path. If you can't describe it completely, you can't automate it reliably.
2. Is the process consistent? Do different people perform it the same way? If there are five variations, which one is best? Find it, prove it, and make it the standard.
3. Is the process stable? A process that changes weekly isn't ready for automation. Automation encodes the current state. If the current state is a moving target, your automation will be obsolete before it's deployed.
The Standardization Sequence
Lean practitioners have known this for decades. The sequence is:
- Stabilize — Eliminate the biggest sources of variation and failure
- Standardize — Define the one best way and make it the default
- Simplify — Remove unnecessary steps from the standard process
- Automate — Now accelerate the simplified, standardized process
Skip to step 4 and you're building on sand.
What Standardization Looks Like
Standardization isn't bureaucracy. It's not a 50-page procedure manual that nobody reads. Good standardization is:
- Visual. A one-page process map on the wall beats a binder on the shelf.
- Practical. Written by the people who do the work, not by someone in an office.
- Living. Updated when a better way is discovered. The standard is the current best method, not the original method.
- Measurable. If you can't tell whether someone followed the standard, it's not a standard.
The Payoff
When you standardize first, automation becomes dramatically easier and more effective:
- Requirements are clear because the process is documented
- Edge cases are known because they've been identified during standardization
- ROI is real because you're automating an efficient process, not an inefficient one
- Maintenance is manageable because changes to a standard process are controlled
The Bottom Line
Automation is a powerful tool. But power without direction is just expensive noise. Standardize the process, prove it works, simplify it — then let automation take it to the next level.
The question isn't "Can we automate this?" It's "Should we automate this yet?"