Following a legislative review of state purchases in fiscal year 2022, the state of Oklahoma discovered that its agencies had procured more than $3 billion worth of goods and services outside the oversight of its Office of Management and Enterprise (OMES) Central Purchasing division.
“Understandably, state leaders were concerned,” says CIO Joe McIntosh. “The legislative review identified issues in the procurement process that created gaps in transparency and accountability. As a result, taxpayer dollars were being spent without appropriate checks and balances.”
To address the problem, the Legislative Office of Fiscal Transparency and the state governor tasked OMES with increasing oversight of state purchasing to maintain proper compliance with state laws and regulations.
“One of our key directives was to audit procurement for all state agencies by the end of the calendar year 2023,” McIntosh says. “To achieve that, we needed a cost-effective solution that could streamline workflows, provide a robust view of procurement processes in real-time, identify rogue spending, and enable us to audit all agencies by the Dec. 31 deadline.”
To do that, McIntosh and OMES turned to process mining, a technique for analyzing event data to better understand and improve operational processes. Process mining encompasses both data science and process management, providing a way to analyze digital footprints and event logs so that organizations can identify bottlenecks and other issues in core business processes.
Process mining on the rise
Operational efficiency remains a top CIO concern, and with the rise of digitization, redesigning business processes is increasingly becoming a key strategic initiatives for CIOs. One in three respondents to the 2024 State of the CIO survey said they anticipated spending more time revamping business processes compared to other IT activities within the next three years, up from 25% today. That figure polled No. 2 behind driving business innovation as CIOs’ most anticipated focus beginning next year.
To tackle that issue, some early movers, like Oklahoma’s McIntosh, are turning to process mining platforms to help achieve their goals.
Process mining provides the potential to enable organizations make quicker, more informed decisions when overhauling business processes by leveraging data for insights. By using the information gleaned from process mining, companies can better streamline workflows, enhance resource allocation, and automate repetitive tasks.
And the market for process mining tools is growing. According to a November 2023 report by global market research and management consulting company Global Market Insights, the process mining market is estimated to expand at a compound annual growth rate of more than 40% over the next decade, reaching $31.52 billion by 2032.
Among the key growth factors are increased demand for process efficiency and the integration of artificial intelligence into process mining solutions. AI enhances process mining by automating complex data analysis, uncovering intricate patterns, and predicting process behavior, according to the report.
“The push for efficiency, automation, and insight are driving leading companies to process mining,” says Ray Wang, principal analyst and founder of Constellation Research. “The demand is coming from companies [that] have been told by strategy firms that they must cut costs to preserve margins. Once these projects are in place, there is a realization that process mining is the foundation for AI-driven process orchestration and automation.”
Organizations, like OMES, that have deployed process mining are seeing significant gains.
Modernizing with purpose
In June 2023, Oklahoma began using process mining tools from Celonis to analyze procurement processes based on purchasing laws and massive amounts of financial data.
“As a government agency, it’s our responsibility to identify where things are slipping through the cracks and correct them as quickly and effectively as possible,” McIntosh says.
Among the goals were to gain greater insight into procurement, slash maverick spending, cut audit times, and reduce legal and financial risks to the state.
“With these tools, we’ve gained real-time insights into how taxpayer dollars are being spent, specifically through reviewing purchase orders and purchase card transactions,” McIntosh says. “This approach offers a comprehensive view of purchasing activities and notifies us of deviations, critical to ensuring compliance.”
Process mining has helped the state address decades-old problems that many government agencies are familiar with, including outdated approaches with data scattered across various systems, emails, spreadsheets, and written documents; limited resources for reviewing purchase orders; and complex processes, with 122 agencies using various purchasing procedures and approvals.
With the Celonis process mining platform in place, OMES has a consolidated dashboard of financial and procurement data that can map the trajectory of purchases to ensure everything is on track and in compliance, McIntosh says. “OMES was able to review more than 16,000 purchase orders valued at over $2.26 billion in a matter of weeks,” he says.
Previously, processing this amount of data would be tedious and require the state to contract about 45 people. “Now, OMES can do much more in a significantly shorter timeframe, saving us thousands of hours in manual labor and tens of thousands of tax dollars in wages,” McIntosh says. “Plus, this allows us to redirect our workforce to other areas where they’re needed.”
In less than 12 weeks, the platform enabled OMES to oversee 100% of spending, reviewing more than 24,000 purchase orders worth $4.58 billion.
“Process mining has been a great asset to the state, helping to drive efficiencies and — perhaps more importantly — transparency, which is crucial for accountability and compliance,” McIntosh says. “The state now has the tools to make data-informed decisions that ultimately benefit taxpayers, who want to see their money put toward improving their lives, like schools, parks, libraries, and police and fire departments. That isn’t possible without a comprehensive view of the state’s procurement activities.”
Streamlining freight operations
International freight forwarding and logistics company Albini & Pitigliani S.p.A. (ALPI) is also benefiting from process mining.
The company began using a platform from Apromore in late 2022, with the goals of identifying steps in processes that were taking the longest or that were causing the most errors; measuring the effectiveness of process changes implemented; gaining a competitive advantage by improving processes; and improving customer satisfaction by reducing process time and errors.
“Our initial approach to process mining focused on analyzing the end-to-end process within our main Transport Management System [TMS],” says Antonio Mortello, technology manager at ALPI. “The goal was to identify where our colleagues were encountering issues and delays, even if they were not consciously aware of them.”
The analysis posed several challenges. For one thing, there was more than one flow across many stages of the process. In addition, it was necessary to understand the unique characteristics of each specific route. And some operational staff were hesitant to adopt changes.
“To tackle these issues, we began by examining the TMS logs to map and categorize all the stored information, aiming to add a process mining layer to our activities,” Mortello says. “We analyzed traffic patterns to identify areas with a more linear process flow, which helped us focus our efforts.”
After these initial steps, the IT team designed a flow and a dashboard for compliance analysis across three traffic types. “We used these dashboards to track key performance indicators [KPIs] relevant to our area managers,” Mortello says. “This process provided valuable insights into our operations, allowing us to pinpoint the most time-consuming steps, whether due to waiting times from third parties or reworks.”
The next focus is on streamlining the process flow with external offices and identifying opportunities to automate steps in information exchange. “By addressing these areas, we hope to achieve a smoother operational flow and improve overall efficiency,” Mortello says.
Using the Apromore platform, ALPI’s IT team effectively identified the root causes of problems, created operational process digital twins to model and test simulations, and worked with subject matter experts to implement changes that resulted in a 75% reduction in errors in fulfillment by improving communication between different departments, a 15% increase in sales through reducing the number of steps in the sales process, and a 20% increase in customer satisfaction, among other benefits.
Process mining tips
Successful deployment and maintenance of process mining requires a clear vision from the management team and board, Mortello says, as well as commitment and persistence.
“Process mining doesn’t usually yield immediate, tangible results, but it can offer unique insights into how a company operates,” he says. “A leadership team with a long-term vision is crucial to ensure the technology is utilized to its full potential.”
It’s also important to thoroughly analyze processes prior to “fixing” them. “Make sure you have a good handle on the process you think you have and the ones you really have,” Constellation Research’s Wang says. “What we see across the board is a quick realization that what’s assumed and what’s done is very different.”
Another good practice is to create a bridge between the technical team implementing process mining and the operational team being analyzed, Mortello says. “This ‘middleware’ role helps ensure that both sides speak the same language and understand the purpose of process mining,” he says. “It’s essential to communicate to the operational team that process mining is a tool designed to help them improve processes, not to monitor their daily activities.”
It’s also important to foster a culture of continuous improvement. “This means encouraging the operational teams to provide feedback and suggest enhancements,” Mortello says. “By creating an environment where the target team feels empowered to contribute ideas, you can turn process mining from a purely analytical tool into a collaborative solution.”
This continuous feedback loop helps ensure that the technology evolves to meet the needs of the people using it, leading to more effective and sustainable outcomes, Mortello says.
One of the most important factors for success is to work closely with the process mining vendor, even after the platform has been deployed, OMES’s McIntosh says.
“In general, I believe it’s very important to work with a process mining supplier that doesn’t just sell you software,” he adds. Organizations need to work with technology partners that are invested in their success, understand their goals, and are willing to work with them to make corrections as needed, he says.
“Also, when deploying process mining, public sector agencies and companies should have a clear vision of what they want to achieve,” McIntosh says. “Our goal was to increase transparency and save taxpayer dollars wherever possible. From there, we had our priorities mapped out: get better insight into procurement, reduce wasteful spending, and accelerate audit times.”
Process mining is like many other technology tools in that organizations can get more out of it if they are more strategic about how they use it, McIntosh says. “It isn’t a problem-solving silver bullet, so to speak, for all things,” he says. “However, it’s a great tool that helps to simplify things in ways that couldn’t be done otherwise.”
When selecting a process mining platform, look for one that offers flexibility in end-to-end process mining discovery, from preparing the data to visualizing it, Mortello says. “This flexibility was crucial for us, as it allowed our team to tailor the process according to our unique needs and challenges.”
ALPI was able to customize and control every step of the process. “We wanted a tool that provided us with the licenses and rights to manage the process independently, without being restricted by external constraints or additional costs for customization,” Mortello says.
