Editor's note: new to process mining? We suggest starting with the introductory read What is Process Mining?.
“Life isn’t about finding. Life is about creating.”
– G.B. Shaw
Suppose you are in the following situation: you are managing a department that has performed all the steps of process analytics and discovered their real process using process mining. Now you have the process insights.
Eventually, you’ve got a beautiful process model lining out how your people work and what steps in general they take. You have validated all the data and everyone agrees that indeed, they are seeing a true reflection of their work in this process model.
Of course, the next logical question is “What’s next?”
What to do with the insights?
A process model is just that, a model. Unfortunately just having knowledge about your process will not create value from your process mining.
You need to connect the knowledge you have as a process owner/practitioner in your company with the knowledge gained from the process model.
This way you will really determine what parts of your process need improvement and what changes you can make in your organization.
You can take action and make the real difference.
Business Concepts that make the difference
Of course, every process is different and every company performs it differently, but many process mining projects share similar goals that profit from similar business concepts.
In here, you’ll find five concepts we often encounter in getting process mining projects from just a concept to bringing actual change based on process insights.
1. Process context
Every process has a context.
One of the most important goals in preparing your process project is to capture that context as complete as possible. The context information in your process model allows for a much more useful view on your data.
With context information, you will be able to relate process flows or timings that might seem unnatural at first, to the complete set of conditions of the case.
This kind of insight is invaluable in formulating the correct next action in your process improvement and making sure that the changes you are planning to make are correct.
So, what should you be looking for in terms of context?
In general, you want to capture all variables that can influence the process. With that in mind, these variables depend on the process but let me give you an example of what you could be looking for from practice.
In one project, simple supplier information helped process mining become a powerful means of communication towards these very suppliers.
As a result of process flow differences demonstration for those suppliers using a standard invoicing app and those who don’t use that app, the advantage of the app became immediately clear.
A huge decrease in the amount of rework as an effect of missing information on the invoice and with that, much faster payment time, a lot less manual work and a much more streamlined procure-to-pay process.
2. Process flow agreements
Every process owner has an idea of how his or her process should flow.
This idea can be an explicit agreement about a certain amount of checks that a team needs to perform. Of course, it also might be just a hunch about what departments should handle what steps in the process to be examined.
Adding this knowledge to your process mining toolkit is not only essential for any form of conformance checking but can also enrich your analysis with a lot of insights from the actual process practitioners.
Conformance checking is one of the most powerful uses that process mining facilitates. By actually comparing your expected process with your executed process, a process owner can see where his expectation is not met and act on those differences.
A much more in-depth look into conformance checking and its advantages has already been written by one of my colleagues and I can highly recommend reading that if you’re interested.
Sometimes the process insight of your flow analysis does not even have to be extremely complex to have business value.
Just knowing who executed a certain step might already provide valuable feedback on management expectations of process execution and the actual employees’ performance.
3. Timing agreements
Next to process flow, timing is one of the most commonly discussed topics in process mining projects.
Every process owner wants his process to be performed as efficiently as possible without cutting corners. Although it’s impossible to define what this means with just a process model.
To effectively analyze what parts of your process are taking more time than they should, knowledge about how much time each process step should take is essential.
From seeing how much time each step is taking and comparing that with our expectations about timing, we can detect bottlenecks, distinguish where there is a shortage of resources, analyze where external dependencies slow down the process, and much more.
Incorporating the terms found in contracts with suppliers or customers allows for an even more in-depth look in how your process is done.
By comparing the due date of, for instance, a payment with the real payment date on a complete scale in the accounts payable process, we can concretely get an overview on what kind of discounts we’re missing. It might be that we are paying too late or there is a negative effect of paying your invoices too early.
Another very natural example is service-level agreements (SLAs) for ticketing systems.
Almost every help desk has an agreement on timings such as first response time and solution time. By identifying what’s causing the cases that go overtime, we can improve our key performance indicators (KPIs) on these timings and increase customer happiness.
4. User-based information
Some of the most interesting results from process mining you will gain from looking at the way people work together.
In a lot of processes with more than one team involved, we can spot a trend of unnecessary case responsibility shifts. This can not only cause delays in process execution, but also annoyance in the day-to-day work of your employees.
By adding the team that has executed process steps in UiPath Process Mining, we can check whether not only our process flow is what we expect, but also how our cases flow through different teams.
Spotting the differences between the expected and as-is team flow allows us to spot where information might be missing and where we need to provide more structure.
Of course, in this context, teams are any groups of people, whether it is country, department, or facility. Viewing the interaction between different groups of people can be very interesting to detect, explore, and act on.
5. Monitoring your progress
Of course, for a lot of companies starting with process mining, process analytics and discovery seem the most interesting features the technology offers its users.
However, process discovery only allows for a first benchmark on how your process was performed in the past. The benchmark might allow for some actionable insights, but without following actions these insights continue empty.
After changing your process in any way, shape or form, the logical next use of process mining is to track what effect that change has on your process performance.
Are you actually achieving more cases that are following the first-time-right path?
Are your KPIs improving based on your change or is there extra action needed?
Without an up-to-date look on your current process, it is impossible to tell what effects your changes have made.
That’s why after having a successful process analytics project setting, the next item on your agenda is how are we going to achieve a better process and how are we going to measure that performance.
What KPIs really show a well-functioning process and how are we going to evaluate our changes on a regular basis using these KPIs?
With those questions answered, you’ll get much closer to a continuous improvement cycle and will be able to move towards an ideal process!
Process discovery is only the beginning of your exciting journey to business improvement.
The real difference comes with changes you implement in your organization, actions you take to prevent the risks you are now aware of, extra budget, and strategy twists that have to be performed according to process insights.
Remember, your organization is ready for process mining when it’s ready for big changes.
But this is your chance to grow and reach a sustainable, efficient point of business wellbeing.
This is a companion discussion topic for the original entry at https://www.uipath.com/blog/automation/process-insights