Along with Robotics, Machine Learning is right up there with the latest trends technology companies are shouting about. Some of the commentary can be very abstract though. So what does Machine Learning really mean in a practical sense?
Let’s take a look at some Digital Process Automation real-life business examples and see how Machine Learning can be applied to these solutions in a practical fashion:
1. Recruitment Strategies
How do we know our recruitment processes are working appropriately? Are we recruiting the right people? When we bring them through our interview process, are we selecting the right candidates? Most of the time we don’t know this until 6-12 months later but are we analyzing this information in the greater scheme of things?
That is, linking candidate selection processes, interview processes, training & certification processes, onboarding processes, performance reviews and exit interview processes all together to ascertain our success/failures in a holistic manner? And how can this information be gathered over time and presented in an instant comparison to inform a decision on the next recruit?
The answer is: Machine Learning.
Driving innovation within a new organization can involve many interconnecting processes, though many see it as a New Product Development process, where ideas are generated and people compete for the mindshare of the senior leadership team through “Dragons’ Den” scenarios.
In reality, there is is a lot more to it. New Product Development is one process but what about tracking incremental innovation within existing products? As you build up a portfolio of products, are you tracking the success of these? Are you learning from decisions on new product ideas/investments in existing products? How do these compare to sales over time?
A huge focus on organizations is placed on capturing the ideas and deciding on them – equal focus should be spent post-decision and learning on these decisions to inform future new product decisions, with Machine Learning prompting recommendations based on previous trends.
3. Project Decisions
Deciding on whether to proceed with a project or not will involve many factors to consider initially. Depending on the type and size of project, this may involve many decision council members or a select few.
When we are making a decision on a project, could we compare the voting results of the decision council to previous successful projects and see how these votes compare to help inform us? Can we compare the post-implementation review results, from a metrics perspective, to report on the success of a project and determine trends of successful projects in their totality?
These are just a few examples to help illustrate the power of Machine Learning, but I believe the common theme and the key to these technologies, is that they don’t replace the human decision making process - they aid it.
This, I believe, will be fundamental to its success.