Podcast

Full Episode Transcript:

Jon: Pankaj Chowdhry, the founder and CEO at FortressIQ, joins us today on the hello, Human podcast, where we discuss the latest topics in artificial intelligence and how it’s being applied in the real world. I’m Jon Knisley, the host of hello, Human, and a longtime technologist helping companies adopt and utilize emerging digital solutions. A big thanks to FortressIQ for sponsoring the program. Be sure to hit the subscribe button wherever you listen to the podcast.

In this episode, we’re going to explore uncovering the mystery of your data. For a while now, companies have looked to better leverage their data. It’s not surprising because we know that data-driven organizations win in the market and outperform their peers. But companies are awash in data. Tons of information is available, from CRM systems regarding customers, to ERP applications providing insight on financial practices, to security sources helping us prevent cyber attacks.

Despite this wealth of data, most organizations still don’t understand how they operate, especially at a granular user activity level of detail. This process data is the last missing puzzle piece to having a comprehensive view of your organization. Fortunately, the situation is changing, and companies are starting to gain an understanding of their processes. Process intelligence technology from companies like FortressIQ is now there to help us gain that insight, and then employees and partners can use that insight to help improve how companies operate.

We are fortunate to have a true industry leader give us his perspective and insight on capturing and unlocking process data, and ultimately decoding work. Welcome to the program, Pankaj. I guess I should say welcome back. You’re our inaugural guest on season one and you are back to kick off season two. 

Since it’s been a year since we last talked, maybe you could start by reflecting a bit on the past year, how have you seen the technology industry shift in the last year, and how has FortressIQ been impacted or adjusted to those shifts in trends?

Pankaj: Thanks for having me on, Jon. That’s a great question. Last year, I didn’t even know where to start with reflecting on it. It’s been turbulent. An incredible amount of change has happened. We talked about decoding work. I think one of the most fascinating things is how the very nature of work has changed in the last year. I think the interesting view that we have, as we were able to look across different industries, different ones of our customers, is how the nature of their work changed and became inherently more distributed.

I think that was one of the most interesting aspects of what we saw. The idea that an office or work or a job was somehow location=dependent, was just completely blown apart. As these things started evolving—they aren’t dependent on the location and they aren’t dependent on time zones—they’re just dependent on the ability to understand the outcome that we’re looking for, whether it’s booking an order, shipping a package, or answering a customer’s question.

When we start to see that the defining characteristics of work are how it impacts your customers and your employees, then really trying to understand now that it’s happening everywhere. How do you get that through in view of it, and seeing how companies have tried to adapt to that. We talked about the industry changing, adapting to saying, look, the nature of my employees have changed, the nature of work has changed, the nature of an office has changed. I need to understand what that is. I think that’s been one of the most interesting things we’ve seen come out of this last year.

Jon: A common misconception or challenge with automation is that you may think that you’re doing it right, but in reality you’re missing a big part of it because you don’t have that full knowledge upfront during the planning period. Can you share an example or two when you had data that’s been counterintuitive to what you expected to see, and how you now have a better ability to resolve or aid that problem with that full insight?

Pankaj: I think the idea that oftentimes business processes revolve around a core application is one of the things that we’ve seen being challenged as organizations try and adapt. We have a customer, a global CPG company, and we were helping them to understand one of their finance processes. They said, look, this is an SAP-based process. Leveraging our technology and our platform, we were able to show them that on average, their workers are spending less than 24 minutes a day inside of SAP for this process.

It was really eye-opening for them to see how much work was going on outside of SAP that really had an incredible opportunity for improvement. This idea that you have almost a human ETL tool, the idea that there’s a person that’s extracting information on one system, transforming or massaging it then, loading it into another system. Just adapting themselves to work around the limitations of existing systems, and really being able to show how that work is needlessly complex presents a great opportunity for improvement. Those are some of the things that we see where people think that the process is already automated in an existing platform, yet really exists across a multitude of different systems.

Jon: I think we see that again and again through every industry and every business unit that we really look at, that a majority of work takes place outside of those core applications. I think that also comes to the idea of ultimately, that traditional process mining technology that looks at the data specifically within a system, and then couple it with our task discovery process intelligence discovery. Does that give you that more comprehensive view of the organization? Would you make that argument?

Pankaj: Yeah. You can take a look back (I think) probably close to 20 years ago with that seminal report—I believe it was from IDC—that talked about how the knowledge worker spends about 30% of their time just searching for information, just trying to get information out of A so that they can work on it in B. 

As we start to understand the nature of work, this idea that there’s a log file that shows what we do—there is in Outlook, there is in Word, there’s in PDF, there’s in Teams or Slack—being able to decode how people are adapting to go find the information that they need, asking people for help, that’s all part of the process that becomes invisible if all you’re doing is looking at log files. That’s really (I’d say) one of the largest ‘aha’ moments that our customers see when they start decoding work leveraging our platform.

Jon: Looking specifically at process automation, you have challenges, very well-documented around scaling RPA. Automation Anywhere just reported that only 49% of companies have deployed more than 10 bots. How does uncovering more of your data and operations help alleviate some of these challenges and help programs scale across the enterprise?

Pankaj: One of the interesting things that we’ve seen there is the ability to handle variation in a process. When you think about a loan process, a loan process isn’t just a loan process. It’s an approved process, a decline process, a KYC process, or a ‘request for more information’ process. There are all these different variations of a process that happened. Oftentimes, when you’re trying to ask people how work is done, they can very, very concisely lead you to that happy path to say this is how we process a loan. But getting all the different variations out there is really, really difficult.

We’ve got some great data around this where we can see that processes being checked, activities that were processed may only execute one or two times, and then there’s an additional step or this step is missing. When you start contextualising that into RPA-based automation, the idea that you may automate one path but then have all of these different variations that aren’t automated, those turn into exceptions. Exceptions—you don’t know that they’re going to be there—turn into a broken running of the process. The bottle will break. It’s actually not breaking. You never coded that handle this variation.

I think one of the things that we’re seeing by really being able to look at data at a concise and expanded level, the ability to uncover those different variations, measure them, and then plan for them. You may be able to say, look, I’m going to have this bot that’s going to handle 50% of the work. We’re going to have people that will handle the other 50%, which are exceptions, and know the level of staffing that’s necessary when you start automating these things.

Jon: I think along those same lines that you brought up the question of rework. There’s a tremendous number of bots and it gets 50% to get started then never get completed on end. Again, I think having this greater insight of the operation going into the process allows you to make smarter decisions. In some cases, deciding what not to automate is just as beneficial as deciding what to automate. 

We’ve discussed this concept of decoding work in the past, and it’s a fairly unique one. How would you define it? What is the value to an enterprise when they successfully decode their operations?

Pankaj: I think it’s important when you talk about decoding work to really define what is work, what are you decoding. This idea that you have individuals within your organization that are working on tasks and activities is usually for the benefit of some sort of outcome. When you start saying, look, I want to decode this, I want to understand that at a data-driven level, what impacts someone’s ability to successfully complete a task? Do those tasks have to be completed in a certain sequence? Are they dependent on other things? That’s all the data that goes into understanding that process.

When we talk about decoding work, it’s really trying to understand it at a very detailed level—all of the different components, parameters, data points, KPIs that go into actually impacting how that work occurs, how long it takes, what the quality level is, and who it impacts. When you think about the nature of data, there are these unintended consequences. 

Think about Uber. Uber could not exist if we did not have real time up-to-date information on cars across the country and across the world. This is something that when we say, okay, well we understand where every car is, what every car is doing, where it’s going, when it’s picked up. 

Now we can start scheduling these things better, and it really changed some incredible things. The impact that it had was not only on traffic and taxis, but really the very nature of ownership of a vehicle. If I could position a car wherever in a dynamic manner so that it would always be there, you didn’t necessarily need to own that asset. We could have a much, much higher utilisation of everything from streets to cars to people. 

The same thing occurs when you understand work at a very detailed level. You can see how different quantitative items are impacting that work, the quality of that work, then work towards improving outcomes, whether it’s for the employee or for the customer, or just a small specific task that might be improving quality for a regulatory process that’s going to help you out with audits. That understanding is intrinsic to the ability to improve.

Jon: We’ve been tackling digital transformation probably for nearly a generation now. Often, transformation gets defined as that combination of customer experience but also operational excellence as well. In many ways, I think we’ve already successfully tackled CX. We’ve been doing it for a long time. If we can tackle operational excellence by decoding work, with that do you see an end to digital transformation? Or do we have a cycle of just continuous transformation that goes on and on forever?

Pankaj: I don’t think it’s going on and on forever. I think our ability to understand work has been really hampered by our ability to do things quickly. If we take the example that you were using—once we decode work, does X end—I think that the first component of that, again, that we have to break down is work is constantly changing. That way the customers want to interact with us, we have new products that need to get rolled out, service level issues.

Your work changes on a day-to-day basis. There’s no point in time or stick in the ground that says this is work and I’ve now decoded it. The ability to constantly adapt and understand the changing nature of work is one of the reasons that process intelligence is so necessary. If you were thinking about web analytics, you don’t put web analytics up on a Friday, then take it down on a Monday. Your website is evolving, it’s changing, traffic sources are changing. The idea that decoding the work is not an end state, it’s a process. It’s really how we look at all of this. How do we leverage that data to improve the outcomes.

Jon: I love that example of the web analytics because that really takes the current state of process technology and is all very episodic. It takes it into much more of an ongoing, continuous improvement type model, which I think ultimately is where it needs to go to have that true transformational impact on industry. Looking forward to next year and the following year, what additional changes do you see in the industry? How do you see organizations try to keep up with this constantly evolving world we live in these days?

Pankaj: I think the rate at which they are going to be adopting new technologies to change with this evolving nature of work, is something that’s going to be really the defining characteristic of a successful organization. If you think about what we saw over the last year, this idea that if you could not adapt to a distributed workforce to a remote workforce, well, your company didn’t really exist anymore. If you had to have everyone in an office, one foot away from each other, you simply couldn’t deliver for your customers.

Over the last year, companies that could adapt were able to accelerate growth and impact. I think what we’re going to see is all of these notions that we used to hold as inalienable facts—the idea that you had to have people in an office, that this had to happen, that there were all these real hard and fast rules about work—that’s all been shattered. As people start saying, well, what does a remote worker look like? What is the security necessary for remote work? As all these things start to change and evolve, the organization that can adapt to that very quickly—adapting oftentimes means implementing new technology that helps with that—those are going to be the ones that are excelling.

The idea that capital or location, that all these things were important, I think really that fundamental characteristic of how quickly can your organization adapt, utilise, improve and emphasise technology throughout every distributed component of your operations, is going to be the defining characteristic of what a next-generation organization is going to look like. We measure NPS scores, CSATs, all these sorts of things. We’re going to figure out a way to measure how quickly you can adapt to a new technology and really start benefiting from it. I think that’s going to be what organizations are really focused on over the course of the next year or two.

Jon: That’s great insight and a great point to end on. To recap today’s conversation with Pankaj Chowdhry, the founder and CEO at FortressIQ, data allows organizations to more effectively determine the cause of their problems and ultimately improve how they operate. Yet many organizations fail to fully discover the whole of their data, creating the mystery of data. Most organizations don’t truly understand how they operate, especially at a granular level of detail. Fortunately, we’re starting to address this gap and get an understanding of how our companies operate. Process intelligence technology is now there to help us gain that insight.

This episode has been part of our second season of hello, Human. A big thanks to Elizabeth Middleman for spearheading the season. That’s a wrap on today’s show. Thank you, Pankaj for joining us and FortressIQ for sponsoring. If you enjoyed it, be sure to give us that like or five-star review on whatever platform you’re listening to. I’m Jon Knisley and this has been hello, Human.