The primary objective of operations in an organization is to accelerate the speed at which processes are successfully completed. Faster throughput means lower costs, better cashflow, and more satisfied customers.

An obvious way to do this is to automate the tasks that makes up the process.

But not just any automation will increase process throughput.

For example, an end-to-end process that spans departments will only complete as fast as its slowest department:

All the localized improvement efforts that focus on non-weak links in the process diffuses the resources and energy available for improvement, and has minimal impact, if any, on the ultimate outcome. Instead of spreading limited time, energy, and resources across the whole process, it is much more effective to focus on the one part of the process that has the potential to produce immediate system improvement: the constraint.

If local change doesn’t move the needle at the enterprise level, it’s not really an improvement. And making a non-constraint more efficient is worse than doing nothing if it further overloads the constraint.

So automating non-constrained steps will not get you the results you expect.

The critical factor of context

Even automating the weakest link may not get you the better result you expect.

Let’s take a simple scenario: The process required to manage surgery at a hospital.

The most constrained resource in the scenario below is the surgeon. You can automate anything pre-op and post-op, but you will never be able to do more than 4 operations in an 8-hour single day (2 hours x 4 operations).

Conversely, if you are able to automate part of the surgeon’s job (like using a Bot) and cut it down to one hour per operation, you are still not going to be able to do 8 operations a day, because preop and post-op is set up to handle only 4 operations a day.

The point is that automation alone will not get you the results you need unless you understand what impact that automation will or will not have on other steps before or after it in the process.

So unless pre-steps and post-steps are able to keep up with the newly automated step, the overall benefit will be limited.

It is therefore critical that any automation project carefully considers these factors if you want the best results.