Transparency in operations is vital to businesses today, not just for reining in inefficiencies and waste, but also for troubleshooting work models, identifying areas ripe for training, and generally developing opportunities for improvement. As companies look increasingly at areas to automate—and contemplate spending big sums to do so—the need for such transparency becomes ever more important so that leaders don’t make decisions blindfolded.
However, getting a clear, complete picture of service processes—measuring capacity, accuracy, and the time it takes to execute when they’re functioning, as well as diagnosing breakdowns and bottlenecks when they’re not—has long been a challenge for companies. The complexity of services, which often involve coordinating multiple functions in nonlinear ways, makes bad handoffs a perpetual problem. Add to these factors the burgeoning number of customer touchpoints and the accelerated move to remote working since the start of the COVID-19 pandemic, and the challenge looms even larger.
Yet the traditional approach falls woefully short. Manual observation and recording are time-consuming, labor intensive, and inflexible activities. The dependence on human observation to complete them makes it hard to filter out subjectivity; indeed, the very choice of what to examine can bias the process from the get-go. Logistical constraints often make a certain degree of extrapolation inevitable. In services, creating a sufficiently granular level of transparency has never been easy. The upshot is that companies may overlook the actual problem—or, conversely, an improvement that on its face seemed incremental but that could end up delivering major benefits.
How can companies arrive at this picture? An approach we call process insights, although still in its infancy, has shown promise. It marries technology tools and analytics in a disciplined, three-stage process that offers transparency, consistency, and objectivity. It can deliver insights in short order, allowing for far faster yet more informed decision making, both tactical and strategic.
Taking a process insights approach
High-stakes decisions require compelling evidence, and big data sets can deliver, offering agnostic, statistically significant evidence that can inform a robust analysis. Many new digital tools allow companies to monitor the way work is done. Along with big advancements in analytics, AI, machine learning, and computer vision, these new tools not only enable observation but also help companies analyze with great granularity. Companies can test historical assumptions, hunches, and hypotheses before committing resources to craft solutions. And they have the option to revisit these later to cull further insights or to inform other decisions (Exhibit 1).
Capture, diagnose, analyze, improve
Process insights involves capturing the activities that comprise a process—through tools that record it, from start to finish, while it is being performed. This data collection allows for rapid diagnosis and documentation. From here, a company can automate the start and end point for the given process to allow for large volumes of data (statistically significant enough to reveal task variations) for mining and analysis. With an accurate end-to-end picture, companies can derive insights and ultimately improve or reengineer the process.
Whether or not automation is the driving motivation, the process insights approach serves three basic purposes. It provides a proof point so that decision makers don’t act on gut feel alone (“We think claims processing is way too complicated and is taking way too long”). It can reveal process information that leaders lack (“We have no idea how long it’s really taking”). And it can be a litmus test to validate the expected gains of a new approach (“With the new process we’re rolling out, we’re banking on a 20 percent faster average time to process”).
Technology enabled, not technology driven
Many companies look to technology as the solution. In itself, it is not. But when integrated with a process insights approach, tech solutions enable information gathering and analysis on a whole different level. By gathering inputs digitally, insights can be quickly generated. The idea is to establish a baseline for future benchmarking; to focus on the end-to-end process, not just what happens in the functional silos; and to ensure that the process is minimally disruptive to employees. Companies have the flexibility to analyze them with other inputs, at different times, and for short-term tactical, as well as longer-term strategic, issues.
Contrary to common perception, advancements in digital technologies allow considerable fine-tuning in implementation. Monitoring and data gathering, running in the background, can be conducted in a circumscribed way: targeted at the specific activities (and not everything that is performed on the given device), isolated from networks and the cloud, and designed in a way that respects the user’s privacy. Process insights is not about adopting a single technology; rather, it’s about layering on technologies to work with the existing technologies that power operations.
Finally, it’s important to emphasize that the process insights approach is about augmenting, not supplanting, the intuition and knowledge of frontline personnel and subject matter experts. The whole point of it is to turbocharge insights, deepen the enterprise’s understanding of processes, and ensure that resources are being applied to create the most value.
Process insights at work
Although the process insights approach is still relatively new, the experiences of two organizations offer a good picture of how the process works and the kind of benefits it can deliver.
Asian telco aimed to automate beyond the production line
In addition to expanding its robotic process automation (RPA) program, one of Asia’s largest telcos wanted to automate a substantial percentage of process work. The company set a goal to equip every employee with a robot assistant within two years. In its initial analysis, however, the company could not find a way to capture more than $5 million out of a $40 million savings opportunity it had identified. Leaders couldn’t see how automating the company’s many smaller, fragmented processes would be possible.
The company established a digital office dedicated to scaling up the RPA program and to making more inroads with automation in other areas. First, the office conducted a test of the process insights methodology to learn how to accelerate RPA for those smaller, long-tail processes. Leaders then designed a systematic approach to quickly capture the benefits of automating the production line. Their process insights exercise showed that they could realize about 90 percent of the savings opportunity through RPA and by applying different technologies and RPA together. They also discovered twice as many processes with automation potential than they originally thought.
These findings showed that the company would be able to accelerate and expand its automation transformation two to three times faster than it could using a standard approach. The added visibility also enabled leaders to design a future operating model and governance structure and develop a tool that could measure and monitor end-to-end impacts.
US manufacturer streamlined financial reporting
A large US industrial manufacturer wanted to simplify and redesign its quarterly financial-reporting process, a process that involved hundreds of people across many silos. The company hoped to cut the unwieldy weeklong process to half the time.
Leaders chose to conduct a pilot first, to verify whether they could realize any savings compared with a control group. With the participation of about 30 employees (specialists throughout the company), the company captured more than 200 working sessions consisting of almost 300 hours of activity—thereby providing a high-level view of time spent on the overall process and component tasks, broken down by type of work performed (for example, financial analysis, procurement repricing, excel modeling, and chart generation).
In ten weeks, the company analyzed more than 50,000 separate steps, classifying outputs by product group and work function (finance, procurement, operations, and engineering), while maintaining individual users’ anonymity. With this large, statistically significant data set, the company was able to evaluate the benefits of a streamlined approach against five key variables: total process time, whether weekend hours were needed, which work tools were used, to what degree noncore work displaced core work, and how much time was spent on non-value-added activities.
The results were surprising. In the pilots, each employee spent 42 percent less time on the financial-reporting process. Weekend work was no longer necessary. The company found that more than half of reporting activities were performed in spreadsheets, suggesting an opportunity to gain more efficiency by expanding the use of accounting software modules. Noncore process work was drastically reduced to an average of 1.6 hours per participant. Finally, the company learned that one-third of the work was spent on non-value-added activities—suggesting big potential improvements through automation (Exhibit 2). And in addition to revealing which functions represented the biggest bottlenecks (the supply chain group), the process insights approach helped the company prioritize its automation initiatives and revamp governance for the process.
Process insights’ longer-term payoff
Beyond the short term, day-to-day benefits—addressing inefficiencies, troubleshooting bottlenecks and breakdowns, identifying people and areas in need of training, and facilitating the sharing of best practices—a process insights approach supports longer-term, more strategic benefits.
Companies can obtain a more precise view of work that has strategic value versus work that is more transactional in nature. Such discovery reveals the complexity of processes and thus has a bearing on decisions about outsourcing, automating, or reconfiguring processes or any component activities. Process insights can help uncover metrics for ways of working, which can help leaders make better decisions about how to manage teams. Moreover, the ability to visualize how work is changing over time can help companies evaluate the impact of process improvement and automation efforts. More broadly, it can help consolidate and disseminate best practices across functions.
The very process of process insights contributes to building a more tech-enabled workforce among those employees who are involved. Employees can move away from the more manual, repetitive, and non-value-added tasks and perform more productive work, such as identifying ways to improve their work or overall process.
To decide whether and how to adopt process insights, companies can start by considering three questions: What important capabilities do we want to build? What skills and resources do we have to support the process? And what practical matters must we settle—such as software requirements, the number of processes and users, and legal and security requirements—before launching? Beyond these questions, it is also important to approach this undertaking with the right assumptions and intent:
- Start with a hypothesis-driven approach regarding where to unlock value. Use the approach to validate (or disprove) a hypothesis: to pinpoint and demonstrate value in a certain area, rather than looking for value in an area that is little understood.
- Select the right tool for the job. Process insights is less about finding a single platform that will solve all your needs and more about identifying a suite of technologies that can help you understand and manage operations holistically, based on your unique requirements.
- Embed the technology in an overarching delivery mechanism. The point of process insights is not to showcase analytics. It is most effective, and most valuable, when embedded into an existing initiative, such as automation or continuous improvement.
- Use process insights to augment—not replace—subject matter expertise. There is only so much information contained in systems and data. The approach and technologies of process insights are no substitute for the knowledge and intuition of frontline employees and specialists. Findings should serve to substantiate (or invalidate) assumptions and help uncover new insights.
Transparency is indispensable for understanding business processes. In the era of big data and analytics, and with the advent of task-capturing technologies, companies can now truly achieve it. A process insights approach, in tandem with such technologies, can help organizations pinpoint and reduce process inefficiencies everywhere in the enterprise those processes are performed. When implemented as part of a broader management system, the process insights approach serves both in the short term (to help tweak process design) and the long term (to support continuous improvement). Think of it as a holistic management approach with flexibility: it can work with low code or no code, in virtual collaboration environments, and in many other organizational circumstances and arrangements. Above all, process insights positions companies to make improvements that enable their people to perform the more valuable work they were meant to do.