Automation and Murphy’s law
Have you ever had something go wrong and found yourself uttering the words ‘typical!’ and ‘of course that happened’ at varying volumes? Well, according to Murphy’s law, you’re right. That thing was always going to go wrong.
Murphy’s law states that anything that can go wrong, will go wrong. It’s an age-old adage that applies to every walk of life. And automation software is just as susceptible.
Here, we explore how Murphy’s law applies to automation. What can you do to mitigate the chance (and impact) of this handy tech tool going wrong?
“Anything that can go wrong, will go wrong”
Murphy’s law outlines an old concept; one that still rings true today and will likely continue to do so until the end of time. The term hasn’t got a clear origin, in that precursors to the modern version of it are not hard to find.
We’ve likely all encountered Murphy’s law at some point. People have used it to refer to sailing. They’ve used it when practising magic tricks. And it rings true every time a traffic light turns red the moment you get to it.
A particularly old usage of the adage comes from Augustus DeMorgan’s 1872 book ‘A Budget of Paradoxes’. It adds a little titbit to the term:
“Whatever can happen will happen if we make trials enough.”– Augustus DeMorgan
The addition of ‘if we make trials enough’ clarifies the meaning a little more. Murphy’s law isn’t a message of cynicism or annoyance. Rather, it recognises that given enough time, anything that has a chance to fail will eventually fail.
How does Murphy’s law apply to automation software?
Automation software is a rule-based system. It works using rules that you specify. That is, you tell it exactly what you want it to do with step by step instructions. And then the automation tool does exactly what you’ve told it to do. It sits in the background and applies your rules until you tell it to stop.
So, what could go wrong?
For a start, if you inadvertently tell automation to do something incorrectly, it will do it incorrectly. It’s not smart or self-aware, and it won’t notice anything is wrong.
Moreover, if you tell it to complete a process that’s broken or flawed, it will either fail instantly or run until it hits the trigger that breaks it. It’s not a question of if a broken process will go wrong, it’s a question of when. Because, as Murphy’s law applies, it can go wrong, so it will.
And with automation software, when one process goes wrong, it opens the door for others to go wrong. A failed process can have a knock-on effect, meaning more can go wrong (and will) than you might think.
How to mitigate instances of Murphy’s law in your automation use
The key to avoiding the annoyance that is Murphy’s law is to reduce the number of things that can go wrong as much as possible. If nothing can go wrong, then it follows that nothing will go wrong.
To that end, automation best practices help you reduce the likelihood of a susceptible process.
- Check and refine your processes before automating.
Known as ‘business process improvement’ (BPI), you should take the time to analyse your processes. And, if they aren’t up to scratch, you should improve them, remove them or fix them. This way, you reduce the risk of automating a broken process.
- Double check your rules and test them.
Once you’ve written your rules and fed them to your automation software, take a moment to double-check that they are working. Do they produce the outcome you want? This helps you catch any incorrect processes before they cause a problem.
- Regularly check in and maintain your automated processes.
One of the biggest pitfalls of automation software is setting and forgetting. Make sure you do regular maintenance on your automated processes. This means checking that rules still work as you want, the processes are still needed, and everything is up to date. This prevents outdated processes from becoming another thing that can go wrong.
Automation and Murphy’s law
Murphy’s law is an unavoidable feature of life. But, when it comes to automation software, there are steps you can take to mitigate the chances of something going wrong.
So, that’s one less thing to worry about.