A Conversation with Lalit Wangikar

We spoke with the winner of this year’s BPI Challenge over Skype1 and recorded the interview for you. The BPI Challenge is an annual process mining challenge, where a real dataset is anonymized and put up for anyone to analyze. This year’s winning team consists of Lalit Wangikar, Arjel Bautista, and Syed Kumail Akbar from CKM Advisors. Lalit and his team used a combination of our process mining software Disco and other data analysis tools for their winning analysis.

We spoke with Lalit about the BPI Challenge, the benefits of process mining, and the process mining interest group he is planning to start in the USA.

To see what he has to say, watch the video above or read the transcript below!


Transcript of the interview

Anne: Hi Lalit. Thank you so much for coming on here.

Lalit: Thank you, Anne, it’s a pleasure to be talking to you about our participation in the business process intelligence challenge.

Anne: Exactly. Lalit, you’re a partner of CKM Advisors, which is a New York based boutique consulting firm, and your team is the winner of this year’s BPI challenge. For those of you who don’t know what the BPI challenge is: The BPI challenge is a yearly process mining challenge, where–based on a real data set–participants can hand in their analysis results, and the submissions are judged by a jury.

Now Lalit, can you tell us a little bit about your background, where you heard the first time about process mining, and why you decided to participate in the BPI challenge?

Lalit: Sure, that’s a great question. I have been an analytics practitioner for more than ten years, before joining CKM. I helped build and grow an analytics company that did mostly predictive analytics work in the areas of marketing, operations, risk, largely for financial services, including banks and insurance companies. While doing that work, we also were solving problems in the operations performance improvement area, and that necessarily takes you to improving process performance. While looking at that, I came to realize that this was an area that was not as well addressed in the main industrial domain as it should have been, and it was in that context that, in fact, I joined CKM advisors to build a consulting firm that will focus on solving operational challenges using big data, if you will, right?

The way we look at operational performance, it consists of three or four areas. One, you have to first understand the kind of work you will receive and the amount of work you will be receiving in these operations, which is generally on the forecasting side. And there’s a lot of work that has happened on the forecasting side. Once you know what work you’re going to get, you have to intelligently allocate you resources to meet that work, to deliver around that work, and you have to have processes that effectively satisfy the needs of work items that come into your operations.

When we looked at these three or four components, I realized that getting to a process design, or understanding the process in detail, was something that was not addressed very well, and we began researching, and that’s when we stumbled upon process mining. That’s when we came across all the work that is being done by the process mining task force, which a lot of your colleagues, and yourself, have been involved in and a part of, and we got interested and excited about it. We tried a couple of different routes as well as different tools, including ProM, Disco, and a couple of others, and while researching on it, we came across the Business Process Intelligence Challenge, and we said that if we have learned something, let’s try and put it to test and see how well we end up doing. It proved to be an extremely powerful method and tool.

Anne: Excellent. You participated as a team, so there were three people who submitted the challenge contribution that you had. How did you go about it? Did you work on it as a team, or was one person taking the lead and the other ones gave feedback? Can you talk a little bit about how you did it?

Lalit: Sure, the team consisted of myself and Arjel, who’s part of my team, and Syed, who is also part of the team. Arjel and Syed are consultants who joined us over the last one and a half years, but there are many ways in which other people got into the depths of the problem, who started working with Disco in a lot of detail, and other process mining tools in a lot of detail. So we’ve worked on it as a team. We all ended up doing a lot of the analysis that we presented in the paper. We also acted as each others’ quality checks, if you will.

It was a process where we all learned together what process mining means and given our consulting background we brought a heavy applied bent to it. In the sense of our focus throughout the analysis was coming up with insights that the bank could potentially use as they look at the process and try to improve the performance of those resources or outcomes. It was a team effort where we challenged each other in getting to these answers.

Anne: Right, you just mentioned it, it was a bank, so the process that was analyzed was a loan approval process, and actually, that was – I have a quote here from the jury – that was one of the things that was highlighted about your contribution, what you just mentioned: That it was a very results-driven method of analyzing, a hypothesis-driven method of analyzing. One of the highlights was that you really managed to convert analysis results into business level results and recommendations. And like you hinted a little bit, you have quite some experience: You have been head of analytics in two different companies before starting as a partner at your current firm.

So, given this background in analytics, can you try to explain again a little bit more: Where do you see process mining, how does it compare to other analytics techniques, where are the strengths, and what kind of gap does it fill?

Lalit: Sure, I think that’s also a good thing to talk about. In my past career, I had worked with several process improvement experts, and mostly we saw the use of techniques such as six sigma, lean, or traditional processes engineering methods being deployed, for both understanding and improving processes. When we compare that to process mining, the big difference we see is that process mining takes you to a whole new level of granularity in understanding how the process works and also understanding performance of resources who are assigned to that process. I think it truly helps you get to opportunities that are not found by using traditional means of process improvement.

Now, we have actually done some work for a UK-based commercial bank, and their small and medium enterprise loan origination process. We deployed some of these techniques there, and what we found was that we could get to two to three times the level of improvement that you could find using traditional means. That speaks to the power of this technique. It’s that you really are able to drive understanding of the process to a level of detail that you cannot in traditional means.

There is a very clear and straightforward reason for that: Most of the traditional methods will rely on established process documents, interviews with people, or following through a small number of work items by way of time and motion studies, just following through them, then through end-to-end process that gets executed. What that ends up leaving on the side is the exception handling that goes on and all the hidden factories, if you will, that exist in an organization, right?

What we find is that through traditional means you get a great understanding of the “average process”, so maybe 60 percent to 70 percent of work items go through the average process, but the challenge is the exceptions or the things that do not follow that average path. Inevitably, that’s where you will get you the most improvement opportunities. That’s where there are a large number of exceptions. That’s where a lot of the work effort goes in. That’s where you see quality challenges. That’s where you see turnaround time challenges.

Process mining as a new tool to look at event level data for a large number of work items gets you to a level of detail that you can start covering these exceptions in a very meaningful manner, and you can show it very tangibly to a business manager or a process owner. That’s the benefit of using process mining.

Anne: Excellent, very interesting. Thanks a lot. The last thing I want is ask is: You told me that you’re about to start a processing mining interest group in the US. What exactly do you have in mind, and what would you like to achieve with it?

Lalit: The idea for the process mining group actually came from the process mining boot camp we saw Fluxicon organize in Europe. When we saw the videos and the participation there, we came to realize that the awareness about process mining as a discipline is not as high in the US or perhaps even in North America. We do not see industry practitioners having adopted, and we saw actually several examples of that in the presentations that were done both at the process mining boot camp as well as the BPM conference, and we believe there is a lot of opportunity to do that in the US.

So, the idea of having the interest group is bringing together the initial sort of resources who are working on it today and building greater awareness about it, together figure out ways of getting more and more business users to test it out. And if you are able to get some people to test it out, I’m fairly sure they’ll start adopting it as well as the benefits will be very, very clear and transparent.

The idea is to bring together researchers in this space, who are pushing the methodology forward and helping establish the ways of building this background better, and business practitioners, which will be also in our self interest as they are potential clients, and we want them to be aware of this new way of looking at how processes are being executed in the organization and so on and so forth. The idea is to bring that group together and drive greater awareness, greater adoption, and just drive more trials, if you will, by the business community that we are a part of. So, the hope is we may find good traction through that activity.

Anne: How can someone who would be interested in participating, based in the US perhaps, how would they get in touch with you?

Lalit: I can be found on LinkedIn, and if anybody sends me a message about, I’ll be sure to get that going and start an interaction with not just me but other practitioners here in this area. They can directly reach out to us at CKM advisors, my email address appears on the BPIC paper, and we can put that at the end of the interview as a link so people can connect me through that as well.

Anne: Okay, great. Thanks a lot for your time, Lalit. Again, congratulations for winning the award, and it was a pleasure to have you on. Thank you very much!

Lalit: Thank you, Anne, and thanks for making the Disco tool available with all the data for this challenge for participants on a free basis, so we highly appreciate that, and we’re looking forward to using the 30-day trial on a real business problem some time soon.

Anne: Excellent. Thank you! Bye-bye.


  1. The quality is not optimal, which was our fault. This was our first Skype-based interview and we learned quite a bit about the technicalities involved. But Lalit was great! Just watch the interview and see for yourself…