How and why to automate and optimize work processes
Workflow automation is a powerful and very useful tool for increasing efficiency and productivity, reducing the time and resources spent on manual actions. And in general, it’s just nice when robots and triggers start doing your routine work.
I am one of those people who so adore automated and optimized processes that even in my personal life I am ready to spend hours delving into auto-actions and commands on a Mac in order to open an application or connect a working token with not two clicks, but one.
In this post I want to talk about automation and optimization of work processes from the point of view of a PM (project manager) in a product team. In general, I consider the initiation and implementation of new optimization ideas to be a separate KPI item that should be included in the goals of each project.
Scope of application
It is clear that automation cannot be applied everywhere, so first I decided to highlight those processes in which this should not be done (or should not always be done):
Processes requiring human participation, expert opinion or judgment. It is logical that you should not automatically transfer a problem in which a designer drew an interface or an analyst compiled an analysis directly into development, without preliminary review and evaluation.
Processes or tasks that have a high degree of uncertaintydepending on other incompletely controlled processes/results/indicators.
Processes that simply don't need automation. Of course, there are also tasks where it is easier to do something manually, without going deep into automation and without spending additional resources on it.
Automation should be used in the following processes:
Regular tasks or processes that have a low degree of uncertainty (routine). A simple principle works here – if you regularly do the same work, which is completely understandable in execution and has low complexity, then it’s urgently time to automate it (or at least try :)).
Processes that do not require human participation/expertise. Here, too, everything is clear – if performing some action does not require any expert knowledge or assessment, then it’s time to automate the action.
Tasks and processes that constantly consume a large amount of time and human resources. In such things, optimization is more applicable. For example, if your team is constantly spending a huge amount of effort reporting to the product, why not discuss this with the relevant managers and executives and reduce the amount of reporting, or for example, increase the time frame so as not to take up a third of the sprint with constant reporting?
Finding a solution and implementation
To successfully automate processes, you must first find these same processes or their weak points that would be nice to improve.
Usually I didn’t have any problems finding applicants, because they constantly reminded themselves of themselves during the work process and sometimes became a headache (when every month you spend several hours integrating dashboards and metrics into a presentation for a report, sooner or later your head will come the thought “why shouldn’t a robot do this?”).
Let's say that we have already found such a process or a weak point and figured out how to improve it. What's next?
And then, if the process is large and complex, requires the involvement of developers/technicians for implementation or managers for approval, then we need provide a clear rationale, purpose and profit our project to adjust/optimize/automate the workflow.
Of course, in small teams and companies you can simply agree with the manager to make adjustments to the process, OK everything in the chat and be happy. However, in large companies with a hierarchy and a lot of people responsible, it is necessary to justify your future changes in order to coordinate and implement them.
At this stage we must:
Analyze what profit optimization or automation of the work process will bring. Just take it and calculate in numbers what will affect us right now, and how the load/costs/deadlines will decrease after the implementation of our idea.
Calculate how much time and resources will be required for implementation.
Describe exactly how we plan to implement this or that feature.
If everything is agreed upon and OK, congratulations, the only thing left to do is to implement the necessary automation/optimize the process according to a pre-drawn plan. I don’t see any point in describing the implementation stage itself, since it can and will vary greatly depending on the chosen process, the company’s technology stack, etc.
Some useful tips
Never be afraid or embarrassed to bring all your ideas to your manager. Optimization is cool, it saves the company resources and money, and your merits will definitely be appreciated if you are successful.
Always try to think bigger in your work tasks. Find places that can be improved. Do not tolerate processes that bother you and burden you for objective reasons. Look for ways to improve them.
Be sure to test the solution before launching it into production so as not to break important processes.
Always try to describe and document your project and the automation itself, write the necessary instructions and calculate its profit. This will help not only to use the process more conveniently, but also to scale it to other teams, and also add it to the list of your professional achievements.
I hope that the reading turned out to be simple and understandable, suitable for the first post 🙂
