Welcome to the 2019 Predictive Analytics World (PAW) Conference Series. Recorded live in Las Vegas, this series is bringing interviews straight to you from exhibitors and speakers at this year’s event. In this interview, host Jamin Brazil interviews Krishna Kallakuri, Founder and CEO of diwo.
Find Krishna Online:
Email: kkallakuri@getdiwo.com
[00:02]
You’re listening to the Happy Market Research Podcast. I’m Jamin Brazil, your host. We are live at Predictive Analytics World, Marketing Analytics World. There’s lots of worlds. I think there’s nine. Anyway, there’s health care.
[00:14]
Quite a few.
[00:15]
Yeah, quite a few. My guest today is Krishna. He is the CEO of diwo. Tell me a little bit about what diwo does.
[00:24]
Diwo is a cognitive decision-making platform. So, the word “cognitive” means the whole idea of how can you take the knowledge aspect of your data, teach it to a machine, and can the machine help the humans with their decision-making process, can it augment. When the human is confused, can the machine help him or “This is a better choice for you today,” “This is how you should react in your business,” and “This is the possible outcome.” So, that is the whole motion around diwo. That’s exactly why we call it “data in, wisdom out.” We want to help the business community with the decision-making process where we are able to reduce their cognitive load of their human brain, reduce cognitive confusion and, most importantly, augment the decision-making process.
[01:10]
That is really interesting. I like the way that you’re framing that out: basically, making decisions a little bit easier while still using insights as the premise. So, when did you guys start the business?
[01:24]
So, it’s about four-and-a-half years ago.
[01:26]
Fairly recent. You have a favorite customer story?
[01:29]
Yes, I do. We actually started our first engagement with a fast fashion retailer. When you look at the process that they had on how they were managing quite a few of their stores across the country, one of their biggest challenges was which product needs to be on the shelf at what location and at what time. When you look at these simple questions, you may think the retailers have figured it out, they have so much experience in business. But the real problem that this customer was facing is they can’t put the right product at the right time at the right location because the customers are not reacting to it or either the product is not on the shelf at the right time, right? So, when you look at the whole business model, it was directly impacting on the revenues because, if a product leaves a warehouse, it’s never coming back. Either you sell it or you promote it or you throw it. So the challenge that we were asked to solve for them is how do we allocate the right product at right location at the right time for the upcoming season. So, diwo, as a platform, it was able to really consolidate the customer behavior, the product behavior, the location behavior and, most importantly, it was able to embed the complete business process around it, which means now all the merchandisers, they know, “OK, 20 weeks ahead, this is where I have demand” “This is how I need to create a buy plan, which means now OK, I need 1,000 units for these 10 locations.” Now, they have enough time to procure the material; they have enough time to manufacture the product; and then, when they manufacture the product, diwo, as a platform, is able to tell them the merchandisers that “OK, this is how you need to deploy,” which means, “Now, OK, you put 100 units in store A or put 100 units in store B or what are the cases.” So, we are really closing that whole loop. This what we mean by taking insights to a whole new level, right? How can you help that business community start consuming analytics instead of producing analytics? It’s not about creating data science models or producing insights. That’s where most people stop, but we want to address the last mile in business. Is diwo able to contribute towards one of your milestones in your business process. So, that’s where we are really focused. So, that story that I just described is now creating a significant impact on their business, which is really saving them millions of dollars.
[04:05]
That’s fantastic. I love the framework of just-in-time retail inventory management and then building predictive models that illustrate consumer behavior, which just empowers the retailer in a tremendous way. It’s a massive consumer advantage. Give me another example of where you guys are rolling this out. Is it predominantly in retail management or are there other sectors?
[04:31]
Diwo is very agnostic of any retail. So, our second use case started with an insurance company in Australia.
[04:38]
Oh, wow! Totally different.
[04:39]
So, one of their biggest struggles was they wanted to identify what could be your pricing philosophy for the upcoming renewal season. If they react to a certain customer segment where the policy is increased by 5%, they don’t understand the impact of that, which means, “Is the customer going to stay with me or am I going to churn?” It’s very challenging situation for them in Australia. The reason is because of all the data privacy policies they have in place. If you as a driver, if you have five accidents in a given year, and if you want to just go switch to another insurance risk provider, you could just say, “I have no accidents.” So you, basically, have no access to his driver’s history or nothing. So, in such difficult situations, now you’re always about retaining your existing customer-base rather than going after new customer-base. So, diwo, as a platform, was able to tell them, “OK, for the upcoming season, these are the segments that I see that are highly risky, but, if you react to a policy change, there is a 90%-likely chance that you will lose these segments and these are the reasons why.” So helping them really stay ahead of the pricing optimization for the upcoming renewal season was very critical because even 1%, 2% churn in their business is a significant impact for them in billions of dollars.
[06:08]
Wow!
[06:08]
So this is a great use case that we solved for them. And, as we speak, we’re also working with another major financial firm in New York, helping them with how they balance the aspect of delinquent customer segments. So, if you look at most financial organizations, they pretty much know that if a customer segment is going to be delinquent. It’s inevitable; they can’t control that. They, at least, have to figure out, “OK, if my delinquency rate is increasing, how do I add new revenue streams into the business?” Diwo, as a platform, was able to address that key problem on how do you manage your risk around these portfolios and these segments are likely to be delinquent. So you’ll have to approve more accounts in this segment. Even though they are highly risky in nature, there is a good chance that you will make more revenue for a shorter time for this high-risk customer too. So very interesting use case that we are solving for them too.
[07:08]
Yeah, that is very interesting. You’ve been in the industry a long time. What do you see as a major trend and what do you think is going to change in the next three to five years? How are we going to different?
[07:20]
I have a positive impression about this and a negative impression about the changes that we are looking at in the industry. At least from what we see, most organizations are adopting technology just for the sake of it, not knowing what the outcome of that technology should look like, which means technology is never aligned to the business application. That’s one major challenge that we see in the space. And the second aspect, especially when you talk about AI in general, again, most communities are so passionate about the experience part of the customer: “Did I detect your face?” “Did I read your emotion?” That’s all great, you know. It’s very technologically savvy kind of a process that you can bring in your business. But people are forgetting the fundamental aspect around the operational aspect: “Did I first increase the efficiency in business before I put experience part?” and “When I combine experience and efficiency, can I make that more effective?” That’s what people are missing out. So I think probably we’ll see some adoption from the business community, but, if you are just stuck with experience, I think that is a problem that we all have to solve or put our heads together to help people educate. You have to climb this ladder one step at a time. You don’t come from top-down; you go bottom-up. That’s where we think the industry is headed.
[08:58]
I really like that framing. I think you’re right that it is a 1+1=3 as you incorporate what you’re talking about, which is the basic business structural insight function and then also layering in, whether it’s the emotional connection or what have you, experience-based part of it as, again, one of the legs of the stool that help the business make the right decisions.
[09:22]
Absolutely.
[09:23]
I think we’ve seen a democratization of access to insights over the last five years. It’s been kind of unprecedented. So, it used to be the case it was really hard to get insights, you know, data, and then ultimately… You know, we went from a… We just been in such a data gathering stage inside of the corporations. And yet, at the same time, we feel like – or I feel like – we’re not necessarily smarter just because we have all the data. Is part of the thesis then figuring out… like taking the customer’s data and then structuring it in a way that can be analyzed and useful for insights?
[10:04]
At least our perspective about insights are experience that we have gone through working with several large enterprises. Is this whole notion of, “All right, let’s build a data lake; let’s hire a data scientist; let’s build insights.” That’s almost you’re trying to dig for nuggets in a desert. You don’t start your journey with data. You start your journey with value, right? What is the business problem we’re trying to solve? What is the value that you want to demonstrate coming out of that exercise? Is there a business reason to do that? Is there an optimization aspect that we’re trying to address or is there efficiency or whatever the case is, right? Unless you start your journey and reverse engineer the process back to the data, this whole aspect of building insights will just remain another deep forest with no direction for anybody. That’s at least what we think is happening today as for insights because building insights is a very highly technical data scientist-kind of a role. Now, when people don’t understand the true business problem, what insights are you building and what value are they creating? That’s the biggest question.
[11:22]
Yeah, it’s not enough just to have insights for insights sake. You really got to have your objective in mind. What your KPI’s are and how that’s going to be a lever and move the business in the right direction.
[11:36]
Absolutely. It’s all about business levers. Unless you cannot move some of those levers with these insights, they’re no good to you.
[11:42]
Right. Yeah, that’s fantastic. If somebody wants to get in contact with you, how would they do that, or the business diwo?
[11:48]
We have a sales team; we have a business development team. They could visit us at getdiwo.com. They could simply reach out by just filling a simple form. They can always shoot an email to hello@getdiwo.com. And one of us will reach back out to the audience to help them understand what we deal with here.
[12:11]
Krishna, thank you so much for being on the Happy Market Research Podcast.
[12:14]
Thank you, thank you, Jamin. It was a pleasure speaking to you this afternoon.
[12:18]
And all of you who found value in this episode, if you please take the time to screen capture this episode, share it on LinkedIn, share it on Twitter, social media platform of your choice. Love to get the feedback from you. Have a great rest of your day.
Recent Comments