If only I could figure out how to replicate every business task in software, I would have a company capable of running itself. I’ll be on an island in the Caribbean with money automagically flowing in to my bank account. Living the dream.
Ok. Let’s return to reality. I don’t have everything automated.
When I find myself asking the same questions over and over again, this is a good opportunity to investigate if I can repeat the solution in software. I extended advice from a friend who calls his method the “rule of 3’s.” This is a form of DRY (don’t repeat yourself.) If I repeat the same lines of code three times, I should refactor the code into a function to call.
I’ve applied this to business processes.
Here’s a few examples of things my team has automated:
- Recruiting – We get flooded with resumes everyday, and we were wasting time asking each other, “Have you talked to this person? Are we going to continue with the process?”
One of the engineers took the initiative to write a script to fill out a spreadsheet with a link to the PDF resume and the email address. Then we schedule interviews from the same spreadsheet.
- Monitoring – We used to have engineers watching over systems (manual QA.) I keep eliminating scripts we have to manually run so they can use their time for other things like development or getting coffee. We replicated common behaviors a user would do and added a system to send alerts when certain conditions trigger.
Will this payoff?
The payoffs of automation isn’t always clear. Let’s say an engineer needs 100 hours to automate a certain process and the drain on an engineer manually executing a task is 1 hour / week. The break-even point is the payoff occurs after 100 weeks.
If you factor in the 1 hour / week drain which includes context switching – automation might appear not so great. Your engineer might not even be around after 100 weeks, which is a little less than two years.
I would look at automation this way – the engineer will never do this task again if the automation succeeds. The time can be invested into advancing other projects. This engineer will not be interrupted during their week.
Here are a few questions to think about:
- What happens if the automator leaves? My boss often mentioned systems tend to break in catastrophic ways when the automator leaves. Nothing is truly autonomous.
- What do you notice your team repeating? Like recruiting, this doesn’t have to be purely technical work. Could you automate that away?
- Is there a software solution you can pay for? If you can save more money and time by purchasing a SaaS solution, then you don’t need to build the automation yourself.
As a manager, look at automating parts of your team’s workflow. If you can find areas, you’ll free up time to work on more productive things.
Leave a Reply