When something goes wrong in a company, the instinct is to blame the person. The warehouse shipped the wrong product. The salesperson forgot to follow up. The accountant entered the wrong number.
But after running Suraelec and watching hundreds of these failures, I've come to a conclusion: most operational problems are actually design problems.
The Pattern
Every recurring error follows the same pattern:
- A human is expected to remember something
- The process allows the wrong action to succeed
- There is no system to catch the mistake before it matters
When you see it this way, blaming the individual is pointless. The system allowed the error. Fix the system.
Trash In, Trash Out
This is the oldest rule in computing, and it applies perfectly to business operations. If your product names are inconsistent, your inventory counts will be wrong. If your pricing is spread across spreadsheets, your quotes will have errors.
The solution isn't more careful employees. It's:
- Structured data with enforced naming conventions
- Single sources of truth instead of duplicated spreadsheets
- Validation at the point of entry instead of correction after the fact
Systems Over Heroics
The best companies don't depend on heroes. They depend on well-designed systems that make the right behavior the default.
When I started building automation at Suraelec using tools like n8n, Power Automate, and custom APIs, the goal wasn't to replace people. It was to remove the burden of remembering, the friction of manual processes, and the possibility of the wrong path being the easy path.
A Concrete Example
We had a recurring problem: customers would request quotes, and salespeople would manually search our catalog, check prices, and build the quote in Excel. Errors were common. Response time was slow.
The fix wasn't training. It was an automated quoting system that:
- Pulls live product data from SAP Business One
- Applies customer-specific pricing rules
- Generates a formatted quote document automatically
The error rate dropped to near zero because the system doesn't make typos.
Boring Operations Are Good Operations
There's a temptation to celebrate firefighting. The person who stayed late to fix a crisis gets praised. But that crisis shouldn't have happened.
The best operations are predictable, stable, and uneventful. When a system works properly, it should feel almost invisible. That's the goal.
What This Means in Practice
If you're running a company and you see the same type of error more than twice:
- Stop blaming people. The system failed, not the individual.
- Map the process. Where does the error become possible?
- Design it out. Can you automate the step? Add validation? Remove the manual handoff?
- Measure. Did the error rate actually drop?
This is how you build a company that scales without proportionally scaling chaos.