To my fellow Process Mining enthusiasts,
For each ServiceNow release, I explore the city it is named after and attempt to draw parallels to our Process Mining enhancements. With Washington D.C., the connection is evident without research – being the epicenter of Democracy, it aligns seamlessly with our goal of empowering people in the decision-making process. The enhancements in the Washington D.C. release, such as the Guided Set-up, auto-generated Project Metrics dashboard, improved Process Map widget for dashboards, and new Automated Findings, aim to broaden the accessibility of Process Mining insights, enabling everyone in an organization to make more informed process improvement decisions at an accelerated pace, which is necessary in today's fast-paced competitive environments.
There is plenty to be excited about with the Washington D.C. release. Let’s dive in.
The first major enhancement to focus on is the introduction of a Guided Setup experience for Process Mining projects. As the number of people in an organization that look to leverage process mining to improve their operations increases there was the need to provide an experience that doesn’t require this new audience to know how to navigate core ServiceNow list and forms to generate Process Mining insights.
This new experience breaks down the simple steps needed to generate a Process Mining project into four areas (Set Objectives, Scope your Analysis, Set Improvement Opportunities, Review and Mine) and provides contextual guidance along the way for those who may be new to process mining and some of the core concepts.
Example of the Scope your analysis step in the Guided Setup experience
Prior to this new Guided Setup experience you could generate visualized process maps in a matter of minutes (see here), now with this new experience we can extend that same power to more people in the organization.
When embarking on a Process Mining exercise very often you are trying to analyze a specific process inefficiency or test a hypothesis to quantify the impact. In those situations, you may have specific reports, KPIs, and dashboards you could align and connect to the project to provide more context for those reviewing the results and insights. You would also take advantage of the rule based finding definitions to flag specific improvement opportunities.
There are other situations where “you don’t know, what you don’t know” and you are simply in discovery mode. For these situations there are two powerful enhancements in Washington to help someone get a better understanding of the process they just mined and the potential improvement opportunities.
The first is an auto-generated Project Metrics dashboard. In prior releases you had the option to add a dashboard to the Summary and Insights page of a product (you still do), but that required you to have one available. In Washington, Process Mining will automatically generate a dashboard that has two visualizations. The first will be a time series (trend) based analysis of the average closure time of the records you mined to give you a sense of whether closure times are rising or falling over time. The second will be a histogram showing the count of records bucketed into closure time duration buckets, so you understand the distribution of records by closure time duration.
Example of the new Project Metrics dashboard on the Summary and Insights page in the Process Mining Workspace
The length of the time series and the duration buckets will automatically be calculated based on the data mined in the project.
For those of you who still want to create custom dashboards to align to your projects, have no fear you will still have the option to add those dashboards and toggle to them.
Example of the ability to toggle between a Project Metrics dashboard and a custom dashboard in the Summary and Insights page
The second enhancement to help accelerate your ability to discover potential process improvement opportunities quickly are some new automated findings. For most of organizations, there is no shortage of process improvement opportunities. But identifying them all and then prioritizing which ones to focus on first can be a daunting task. With automated findings we help streamline that process.
In Vancouver, we introduced automated findings to help identify rework scenarios, ping-ponging records, and process variants that have extra steps. You can learn more about them in this Process Mining Academy session.
In Washington, we extend the library of automated findings by four:
- Pattern repetition – will detect work that is going through a series of repeating steps more than once. For example, there may be a quality issue or training opportunity where a certain type of ticket is moving from Work in Progress to Awaiting Caller to Resolved back to Work In Progress to Awaiting Caller to Resolved multiple times or passing through the same three Assignment Groups over and over again.
- Extreme arc duration – will detect outliers that are taking significantly longer than the median time it takes for a piece of work to move through a certain process transition. For example, it may typically take a certain type of tickets 3 days to move from Work in Progress to Closed. But there are 15 tickets that took greater that 24 days, these would be flagged and surfaced on the Summary and Insights page.
- Extreme arc repetitions – will detect outliers that are repeating a certain process transition multiple times. For example, you may have a set of tickets that are going from Group A to Group B more than 7 times before closure. The difference between this automated finding and the Ping-Pong automated finding is based on the number of times the repetition happens. (these thresholds are configurable).
- Slow transitions – will detect clusters of records whose average time to make a certain process transition is longer than the average time it takes another cluster of records to make the same transition. For example, there is a cluster of 250 customer service cases that are taking on average 24 hours to move from Work in Progress to Closed, but there is another cluster of 75 that are taking 4 days on average to move from Work in Progress to Closed. The automated finding will flag those 75 and surface them on the Summary and Insights page.
All these automated findings get surfaced and categorized along with the rule-based findings in the Improvement Opportunities section of the Summary and Insights page of a Process Mining project.
Example of the automated findings being surface in the Improvement Opportunities section of the Summary and Insights page in the Process Mining Workspace.
Then the person (process or business analyst, team leader or process owner, inquisitive platform owner, etc) doing the analysis can further analyze that opportunity, add some notes, or follow up activities, or archive it. Not all situations will represent an opportunity to improve but simply surfacing them helps provide some directional guidance to where you might want to focus first because it presents the largest potential opportunity.
Once you are on the workbench and using the visualized process map to do further analysis, there is another change and an enhancement to be aware of. When you click on an activity node (the ovals on the map that represent States, Groups, etc.) or when you click on an arc (transitions between States, Groups, etc.) there is a panel the pops up to give you additional details about what you clicked on.
The change you need to be aware of is we’ve grouped the Show Records, Requestor intent (clustering), Root Cause Analysis, and Filter on transition options into a collapsible “Investigate” menu. (Note: You may have to scroll sometimes to see it)
Example of the new “Investigate” menu in a Node details panel in the Analyst Workbench
The enhancement you need to be aware of is that for nodes (States, Groups, etc.) you now can get a breakdown of where the records are coming into that node from, and where they are being sent to. So, for example a manager of a certain group wants to get an understanding of where their team is getting their work from and where they are transferring it out to most often.
In the example screenshot below, your IT Support Services has received the most incidents from IT Support – Americas (204) and has sent the most incidents to IT – Support Americas (1500).
Example of the incoming and outgoing work analysis on a Node panel in the Analyst Workbench
This is a fantastic level of visibility when trying to understand workload.
Which leads to the next enhancement – the ability to make this incoming and outgoing work analysis available on dashboards used by different stakeholders throughout the organization. Not just those using the Process Mining workspace.
The process map dashboard widget that was introduced in the Vancouver release allows you to put these powerful process insights side by side with process KPIs and operational reports to give stakeholders that holistic view of work they are responsible for.
Example of a Process Mining process map widget embedded in a Platform Analytics dashboard
Then as the stakeholder identifies a potential process bottleneck, they now can drilldown into that specific step of the process and get a better understanding of the volume and velocity of work moving in and out of that step.
Example of the Incoming and Outgoing work view in the process map widget after someone has drilled into a specific node.
In addition to the enhancements to the process map widget, you can now start surfacing the improvement opportunities that are powered by the rule-based and automated findings in your Platform Analytics dashboards. The Insights panel can be configured to show Process Mining improvement opportunities based on the KPIs that are on the dashboard and the Process Mining projects you choose to associate to that dashboard. This connection helps to answer the “Why” behind the KPIs on the dashboard.
Example of Process Mining improvement opportunities powered by rule-based and automated findings available via the Platform Analytics Insights panel in a dashboard
Tremendous level of visibility available on the dashboard your process stakeholders use to manage their day-to-day activities.
As most of you know, one of ServiceNow Process Mining goals is to provide End-to-End visibility into all critical business processes. With every release we move closer to that goal. In Tokyo we released our multidimensional process map capability to provide visibility into processes that touched multiple tables (Check out the Process Mining Academy session on Multidimensional Process Maps). In Vancouver, we introduced the ability to apply Process Mining to external process data generated by applications other than ServiceNow (Check out the Process Mining Academy session on Process Mining for External Data).
In the Washington release we are bringing the two together. You can now get End-to-End visibility into your processes that combine both ServiceNow and external process data. For example, you might have an HR Onboarding process that has many steps and subprocesses involved in it. One of the subprocesses may be to request and purchase a laptop. The purchase and invoicing process may live in SAP/Ariba. You are now able to create a single process map that combines the main onboarding process steps that are flowing though ServiceNow and the purchasing process living in SAP/Ariba.
Example of a multidimensional process map that includes both ServiceNow process data and process data that has been sourced from an External application
These are some of the highlights of what is new for Process Mining in the Washington release.
Of course, there are other enhancements that are a bit more use case specific like the ability to set projects to retire after x number days to manage project sprawl, or the ability to group multiple field changes into a single compound activity, or the ability to mine archived data.
In an effort to keep this blog post from turning into a document longer than the Constitution of the United States I’m going to stop here.
We spend time on all of these enhancements and share some live demonstrations during the “What’s new in the Washington D.C. release” Process Mining Academy session.
Washington D.C. is the epicenter of Democracy, a style of government whose main characteristic is active participation of its citizens which drives accountability and responsiveness.
Those outcomes are exactly what we are trying to drive with the Process Mining enhancements in this release.
Thank you and may God bless Process Mining enthusiasts everywhere.
Other important Process Mining content:
Process Optimization NowLearning Course
Process Mining Use Case Series
Interested in additional blog posts on Process Mining, Performance Analytics, Predictive Intelligence and the Virtual Agent? Check out this Now Intelligence blog carnival.
1 Comment
