What is task mining?

This technology uses optical character recognition (OCR), natural language processing (NLP), and machine learning algorithms to interpret and analyze this data, which in turn enables analysts and stakeholders to identify operational inefficiencies.

Task mining solutions are considered part of process discovery, a subset of process mining, and according to Gartner’s “Market Guide for Process Mining” the market for this technology is rapidly growing.

As the COVID-19 pandemic continues to fuel digital transformation efforts, adoption of task mining technology is anticipated only to increase as the benefits of it are fully realized.

Process mining focuses on end-to-end process optimization, such as an overall procurement process. In contrast, task mining focuses on the individual tasks that ladder up to that larger process, such as budget approval for accounts payable. They also primarily differ in the types of data that they use for each analysis.

Process mining primarily relies on business metrics and event log data from information systems, such as enterprise resource planning (ERP) or customer relationship management (CRM) tools. In contrast, task mining can use user interaction data, which includes keystrokes, mouse clicks or data entries on a computer. It can also include user recordings and screenshots at different timestamp intervals.

These data points help analysts and researchers understand how individuals are interacting with a process and subprocess to complete a task. They both also use data science techniques to arrive at these insights to optimize processes; task mining just enables this process at a more granular level.

While task mining and RPA both focus on process automation, the two technologies are different but complement each other well. While task mining technology helps businesses identify bottlenecks in their process workflows, RPA tools implement and perform against the automation opportunities that are discovered through these analyses.

Task mining tools start by collecting data from users’ machines, which can include keystrokes, clicks, user inputs, recordings, screenshots and more. From there, optical character recognition capabilities can add more context about what the user is doing.

For example, it might look at the timestamp data to help assemble a general timeline of activities in a subprocess. Once that data is structured appropriately, machine learning algorithms can be used to cluster data into specific tasks in the subprocess, such as “submitting a purchase order.”

The data can then be combined with event log data to help contextualize performance. This data-driven insight helps businesses identify bottlenecks and take the necessary steps to resolve them.

Task mining techniques have been used to improve process flows across a wide variety of industries. Process maps can help businesses focus more on the key performance indicators (KPIs) that matter, spurring them to reexamine their operational inefficiencies through process mining and task mining.

Some use cases of task mining include:

• Task documentation: As new team members come onboard, documentation is frequently reviewed to close any knowledge gaps. However, depending on the project and the available resources, documentation might not be available or up to date. Task mining tools provide a way for teams to bring insight into a task in a larger process, creating alignment across the team. It also reduces the need for individual dependences, providing an easy way to build documentation and visualizations through process mapping and other automation tools.
• Governance and compliance: As businesses face stricter government regulations, task mining can help to hold companies accountable by identifying areas where compliance errors occur. This visibility can provide a pathway to resolving these issues more quickly, potentially saving companies on costs, such as legal fees and negative brand publicity.
• Task automation: Task mining produces a clearer view into specific subprocesses that can also enable program managers and people managers to understand which parts of the process can be automated through tools like RPA.

While task mining can yield many benefits, the most frequently realized benefits include:

• Increased efficiency: Task mining focuses on identifying operational bottlenecks to expedite process improvement. As those inefficiencies are found and rectified, companies experience increased velocity across tasks. If a task in a process has been over-resourced, it can also lead to staff reallocation to other priority work, potentially improving employee morale by finding them more meaningful work.
• Better compliance: Tasking mining tools collect data from users, enabling governance teams to determine breakdowns in compliance during specific tasks. This ability to pinpoint problems and resolve them quickly can help facilitates better governance and compliance across the business.
• Greater transparency: Task mining can provide workforce insights at the individual level, enabling managers to give valuable feedback during performance assessments and reward employees fairly for their work. It can also help them to reallocate employees to different work when they don't seem to be a great fit.

However, task mining is not without its challenges. Some key difficulties include: 

• Data privacy: Task mining can record and log user actions and it can also raise concerns around privacy. As a result, users should approve these tools before activation and they should also protect users’ personal data through appropriate anonymization.
• Missing context: Because task mining focuses on a subprocess within a larger one, the larger context of performance can sometimes get lost. It’s important to use task mining technology alongside process mining to get a fuller picture of performance across teams. Otherwise, companies risk prioritizing task optimizations that do not have the greatest impact to the business.
• Concept drift: As businesses move quickly to transform for the digital age, tasks and processes can change in real-time. Changes in tasks and processes can impact analyses, leading to concept drift.