Malcolm Hawker 

Hi, I’m Malcolm Hawker and this is the CDO Matters podcast. The show where I dig deep into the strategic insights, best practices and practical recommendations that modern data leaders need to help their organizations become truly data-driven. TuneIn for thought provoking discussions with data. IT and business leaders to learn about the CDO matters that are top of mind for today’s chief data officers.

Malcolm Hawker 

This is Malcolm Hawker. I am your host for the CDO Matters Podcast. It is my distinct pleasure today to introduce Mr. Saleem Khan, who will be joining us today from Discovery Data where C where Celine is the CD AO we’re going to talk about that difference between CD and CD AO. I mean, I’m I’m interested to to drill. Down a little bit on that. I I met Celine, I think it’s like maybe 2015, 2016 in. There, whenever you joined Dun and Bradstreet, I was pretty sure I was already there. So Celine and I overlapped at Dunn and Bradstreet, where at the time selling you were reporting to our Chief Data Officer, a woman named Monica Richter. And you were in charge of innovation at Dunn and Bradstreet for a number of years. And I can remember being a bit of a pest to Celine when I found that he was working on some blockchain related stuff and and I. Think that I was. I was. I was often pestering you. Related to concerns that I had about being blue addition of blockchain at Donna Bradstreet, at least from the perspective of of any sort of kind of disruptive force, we’re going to talk about blockchain today. We’re going to talk about a few different roles, including what it’s like to be a CDA O at a company that does data for living. So I think Selena’s got some interesting perspectives. There I would also love to talk about and plan to talk about data as a product, because I think for something for you it is it’s it’s very real because that’s what you guys are selling. So I’ll stop there. I’ll stop putting words in your mouth. So I would love if you could just take a minute and introduce yourself, describe your role and tell us a little bit about Discovery data. 

Saleem Khan 

Thank you, Malcolm, for having me on the podcast. Great to be here with prophecy a little bit about myself. So I am. You know, a recovering technologist. My early career was in the programming space, so I was, you know, out of college. You know, I had a. Computer science degree. I worked for a startup where I was employee #4 and I was a Java programmer. Did PHP development front end? You know JavaScript work, database design, you name it, I programmed. Most of it, but you know we built a product up built up to about 20 employees and we sold the company to a subsidiary of IBM. And so I actually, you know, made the entree into IBM, spent about four or five years there working in their global business Consulting Group, where I had the opportunity to work with some of the largest blue chip. Companies in the world, you know, everyone from, you know, the larger credit card processors through to insurance. And so I had a great opportunity to learn the full tech stack and and at. Right around that. Time we started to see larger data sets start to present themselves, especially in the banking world especially. You know when it came to people’s SMS alerting for their accounts and things like that and trying to determine. You know who might go below their their account thresholds and that sort of thing? So there’s a really neat introduction to some of the larger data sets that IBM shortly after that I I worked for standard and Poors. I was at S&P ratings for about 7 years there. I did everything from, you know, architecting solutions for the different types of bond rating systems. Everything from corporate bonds to municipal bonds to structured debt. So I did a lot of that and then I ran their innovation lab towards the end of that and one of the first forays I truly had into data science and. And the the data. The world was building out a prediction model where we built out a a sales prediction model where we would try to determine, based on a number of factors for publicly traded companies, which companies were most likely going to issue debt, right. So on one end of the spectrum, you’ve got Microsoft and Apple. Which are incredibly cash rich and they usually don’t borrow right, because they just have. A lot of. Cash. But then? On the other end of the spectrum, you’ve got companies that require a lot of cash because they’re very, you know, capital driven, capital intensive. And so they have to take out debt and whenever debts taken out, S&P has to give a rating on it. And so we actually had. A model that would predict which companies would need to issue debt and more often than not. Something like 80% of the time this this model was accurate, so that was my first foray into using larger data sets and you know, coming up with predictive models, Fast forward worked at Dun and Bradstreet, as you mentioned that I worked specifically in the alternative data space. I felt like a kid in a candy shop in that job. It was really neat because I got to work with mobile foot traffic data. I got to work with satellite data import export data, you name it. All the coolest new data sets that were coming out back in 2015. You know, we got to trial them, use them and actually put them to use. Should generate revenue, so that was a pretty neat experience. And now finally I find myself at Discovery Data. I’ve been with the company for about 2 and a. Half almost 3. Years now. Join with them right after the pandemic, but my job right now is chief data and analytics officer. And just to break that down, you know the job is really what I like to call 3 components, right? So I break it down into a cheeky little acronym, I call it rad. Research, analytics and data. Right. And so the function that I have. Number one is is research. So one of the things that we do is we provide a lot of research, a lot of content around trends within the space that we operate in, which happens to be the financial advisory space. So discovery data by way of background is a company that collects data. On financial advisors across the United States, so anyone who’s a registered investment advisor, anyone who’s a broker dealer, anyone who is an insurance agent. We collect information about those individuals collect. Profiles on those individuals and essentially sell that data as marketing data, right? And so part of my job is to look at trends, right? So the the research, the PR and the RAD. Is to look at what what is the financial advisors outlook for this year for instance. And one of the key themes that emerges this year? At least is look, equities are down, debt is not paying so well. So where do you go to find yield? Turns out alternative assets, things like private equity, hedge funds, real estate. Even things like artwork, right, and cryptocurrencies, those seem to be taking off as you know, viable investments at a time where you’re just not finding yield in the market in, in the you know the the normal capital markets, if you will. And so the the are in in RAD research is part of my job the A is analytics. So the analytics piece of it. Because really, how do we use this data set that we have? We have a very unique data set in that we not only have sort of a rolodex of individuals in the financial advisory space, but we also have some of their browsing behavior. So we have information about what these individuals are reading about. Are they reading about? ESG related topics or large cap growth funds or are they reading about how do I break away and start my own independent advisory firm, right? So we run a number of analytics to determine intent, right, which is something that you might remember from, you know, our days over the other place. But you know, intent data becomes a big part of the analytics piece. We also do analytics around diversity and inclusion in a couple other areas. And the final piece before I pause is is the D in RAD the the data operations is what that is and that is essentially making sure that you have. The apparatus this ability to take in multiple sources of information, make sure that you validate the data as it comes in, making sure phone numbers are correct. Emails are pintable IP addresses are correct. All that. Then you start to synthesize and normalize that data into something consumable and something that you can actually distribute to a customer, right? So there’s a whole framework we have, we call it frame. We could get into it in a minute, but there’s a whole data operations component here too. So I’ll pause there, but you know that’s that’s me in a nutshell. 

Malcolm Hawker 

That’s fantastic. Thank you that. So would it be fair to say in? Your role are are you responsible for? Kind of like what I would call more traditional internal data and analytics, right? Like like creating dashboards, reports, data quality, MDM for internal consumption. Or is that under more of an IT banner? 

Saleem Khan 

Yeah. So that’s more on the IT side. So part of my job as chief data and Analytics officer, I, I I say almost 50% of my job is sales and marketing and the other 50% is product development. 

Saleem Khan 

So I have a great team. I have the luxury of having a team of individuals that do research, analytics and data and they they operate the day-to-day. You know, you know, you know, sourcing of the data normalization and distribution of the data, but my job becomes making sure I communicate with customers, especially our largest customers. As often as I can to impart on them, what’s the latest and greatest thing that we? Have to offer. What are the trends that we’re seeing within the marketplace? They want to hear directly from the Chief Data officer. And so I feel the role has become. A little bit of a marketing role, right? It’s become a little bit of a sales role, especially at a data company because data is our product and so who better to hear from than the Chief Data officer about what trends there are and and where they’re taking your data set. You know what’s next in your data set? 

Malcolm Hawker 

Well, that’s interesting though, but that it does seem like a natural. Extension based off of what you guys do for a living. But I think if you’re talking about other large companies, those other large companies, even though they may not be in the data business, would certainly have a perspective on what’s happening in those businesses. An example would be, maybe I’m the CEO at a large shipping company. Would I have interesting insights that may be valuable or monetizable in regards to kind of? Global trade or global demand or or or even the high high level kind of macro factors. So it’s that that’s interesting to me. It makes it makes total sense that you’re kind of like veering over into more on the marketing side and even maybe a bit of a of a consultant sounds like a. 

Saleem Khan 

Yes, at times it is some consulting work, so helping customers think through scenarios. So for instance, one scenario I could provide. So discovery data, like I said, we are focused on financial advisor and insurance agent data, right. I think of us as sort of being the Yellow Pages for that, that information a customer. They come to us and say, hey, look, there’s a specific geography in the southeast where we want to sell specific type of assets. You could be a large asset manager that sells an ETF, for instance, or. People fund and you want to target specific advisors that have had a level of assets under management growth over the past 10 years. So will come to us and say, look, I only want to target advisors in the southeast that have had at least 10% year over year growth in their AUM and we want to be able to find that cohort. And and we want you to help us go after that cohort to be able to sell our products to be able to pitch our. And so that’s one scenario where we almost become consultants, right the the end user here has a very specific goal in mind. They have a business goal where they feel they’re not as penetrated in the southeast or southwest or whatever region there might be. And so they come to us with that specific goal. And we’ll have to walk them through how we would help them achieve that outcome that they’re look. To achieve and I’m happy to say that that we have nearly 1100 customers of which you know over 90% are are repeat customers. And it’s partly because of this consultative sort of approach. We take and. It’s not all on my team. I have to also kind of lump in our customer support team. We have an excellent client support team. That almost gives White Glove service to the customer to walk them through these scenarios to help them and handhold them. I think that becomes. Key you know. It’s not just on the chief data. Officer or the data team? To specifically walk a customer through and be that consultant, but to have that client support team that’s fully dedicated to making that happen, that is a key part of of success here. 

Malcolm Hawker 

I I I find this fascinating and and not, not necessarily, because this is what you do for a living day in and day out, and it’s obviously it’s working for you and. You you figured? It out, but everything that you just described. Could easily that process that you described, like the deep discovery, the consultative approach, understanding what clients. Need and and and. How? To solve for those needs. Could easily be. Applied in an internal focus in an internal use case, if I’m more of a traditional CEO who’s been given a mandate of help build out a data and analytics organization to fuel my business, and there there’s a lot of conversations going on these days about, you know, kind of more. I will just say traditional CDF’s at larger companies about them becoming more. Or product like and and to implement more product management type disciplines within the CD organizations. Because so many organizations are struggling to deliver value for their companies, right where so many CEO’s are struggling to get value out of governance, that’s another good question I could ask around governance. I have to. Assume that your your factory, so you’re you’re kind of managing the data. Factory kind of. And and that your factory has got very, very specific rules about what’s an acceptable use of data, what isn’t, that’s got very specific rules about accuracy of data, about all the traditional kind of metrics of data. Do you define that as a governance initiative with with within your organization, or do you? Call it something else. 

Saleem Khan 

No, absolutely, we do call it governance. We have a framework as part of our our overarching that RAD framework. As I mentioned as part of the data operation. We have another framework under there called frame, another cheeky name, but frame is an acronym. It stands for fuel, refine, analyze, magnify and execute. So what each of those components is is essentially different parts of the data operations life cycle. So the F and frame the fuel. Is essentially the different data sources we have, so we collect data from sources like the SEC, State Insurance Commissioners, different, you know, financial. Points as well as social media, right? So those are those are areas that we collect our fuel, if you will, from then it’s a matter of refining that fuel. So the refinement of it is partly the beginning of that governance that you talked about, right? So this is essentially ensuring that fields are validated properly we have. You know hundreds if not thousands of checks that we do on each individual field we take in, we take in something like almost 3000 different data fields across the financial advisory world, Couple 100 fields in the insurance agent world. And then I’ve got a number of other data sets that we tack on, which brings it to almost like 5. 1000 unique columns if you will, that we we take in and so every one of them needs to be validated and make sure that we have consistency around that. So that’s the the our in frame is to refine it. Then we analyze the data set. So for instance what we’ll do is what I call the dipstick test. You know, a more seasoned person would call it, you know, a stratified random sampling. But that’s essentially, you know, taking a a random sample of your data set a 5 percent, 10% sample, something that’s represented. And to actually manually review it. And so we actually have a team of folks that manually review that data to ensure its accuracy. You know, we’ll check sources like LinkedIn, SEC sites, you know, different state insurance Commissioner sites to kind of confirm that the data looks accurate. And so if it looks accurate and the representative sample, that means your larger sample. Is going to be correct. As well, or at least 9090. 5% correct. Then there’s certain areas where we we magnify. So this is the M in frame. Magnification is essentially creating derived data assets. So for instance you’ve got all this raw material now. I’ve got data on, let’s say Malcolm Hawker. He’s registered investment advisor who lives in Florida. He’s registered to be an advisor in four states adjoining Florida. He’s managed to build up a book of business that has 30 million in AUM, or 50 million in assets under management. He’s killing it. He’s doing a great job. And so now the question. 

Malcolm Hawker 

John, a financial analyst, not a financial advisor. 

Saleem Khan 

Is like right? 

Malcolm Hawker 

This is non financial advice. 

Saleem Khan 

It it is not absolutely, we’re just using Malcolm as. An example but. 

Malcolm Hawker 

Yeah, yeah, yeah. There’s a fictitional. There could be many of me. 

Saleem Khan 

Yes, that’s exactly right. So what we would do then in the magnify component is to say look, you know, if you look at Malcolm on a on A5 year time horizon, he’s been doing fantastic. And so how do you rate him on a score of 1 to 5, right? Five being the highest. And so we’ll actually create these derived datasets. We’ll create advisor success levels as we call them. So we’ll say look, Malcolm is an overachiever. This is someone that. Will continue to grow their AUM based on their their previous history and track record. And so if you’re an asset manager, a large asset manager, you know. That is distributing. ETF you most likely. Want to go after someone like this? Someone who represents someone who has this profile of growth, right? Because that’s most likely the person that will buy your product and then ideally sell your product to an end investor, right? And so that’s where the magnification piece comes in. And then finally, the last part of frame is execute. And execute is essentially how we deliver. Our products to our customers and so we’ve got multiple ways in which we do that and while. It’s not a. Governance theme. Necessarily, I think it’s an important topic to talk about because there’s multiple ways nowadays to distribute your content, your data to customers. So traditionally we’ve done portals, right, every every data company out there. Some sort of portal or user interface that you log into and you put in your search filters and you pull out a data set and you dump that into Excel and then you plug that into your Salesforce or whatever. It might be. We’ve also got. APIs as a as a mechanism to distribute data, but then more and more what we’re seeing is companies like Snowflake emerge right, and companies like that essentially are serving as data aggregators. So what customers more and more are doing is they don’t want to just see. Necessarily, data from one provider they want to see data from two or three providers that are. And take those two or three providers and create something new or unique that helps them meet their needs. So for instance, they might couple this financial advisor data that I provide along with consumer data. So data about you as an individual are you for instance a golfer or a wine? Enthusiast, you know, are you into fly fishing? What are your hobbies? Right. That’s not necessarily the data that we would collect. But you know it’s data that you could lump together, you know, kind of synthesize together and create a new derived output out of that to say that. Here’s a profile of an advisor that has a track record of growing their business. Also happens to be interested in golf also happens to have been a graduate of UCLA or Penn State or whatever school it might be as an asset manager, you want to know these things because as you market your capabilities to this individual. It’s great to go in there, not with a, you know, I’ve seen that you’ve grown your business by 10%. Year over year. I’d love to talk to you. You know, it’s better to go in there with. Hey, I happen to also be a alumni at Penn State, and I happen to like golf. Would you like to go out to a golf outing? That’s the sort of kind of synthesis of data that we’re seeing on on platforms like a snowflake, where you can start adding in these different data sets. So make a. Long story short, you know we do have this, this framework. We call it frame and you know it has all these components around governance as well as how you distribute data in the end. 

Malcolm Hawker 

So it sounds to me like governance is just baked in, it’s it’s part of the cake. You can’t you can’t bake the cake without having governance baked in. That’s is another thing that I’m hearing a lot. More of in my world. Which is. You know, a governance by design, right? It kind of kind of aligns to more of kind of like a dev dev OPS model where it’s operations by design. It’s just part of the cake. You bake it in. Sounds like that’s kind of what you’re doing. Where does it? Where does it start? What’s the kind of the the the the tip of the spear here? Right. Is it do you have? Product managers or analysts. Like who? Who’s out there saying. Hey, we, we. Need to build. A widget or we we need XYZ. 

Saleem Khan 

You know, believe it or not, nine times. To 10, it comes directly from customers, so one example I’ll give is in the middle of 2020, we had one of the world’s largest asset managers come to me and say, you know, we’re really thinking about our diversity and inclusion initiatives and we really need a data set to understand what the current makeup is. For financial advisors within the United States. How diverse is this industry? How many diverse advisors do we have? How many minority advisors do? We have and. So that particular idea came directly from a customer requesting it. And what happened was that large customer asked for it and then all of a sudden a dozen others asked for the same thing. And so interestingly enough. 

Malcolm Hawker 

By happenstance it just it just happened. Like just coincidentally. Or they they they they were all playing G1 day or like maybe. 

Saleem Khan 

No, I think I think I think that particular one was driven by certain, you know, cultural things that were going on in 2020. If you recall in the summer of. 

Malcolm Hawker 

There were there were, there were. 

Saleem Khan 

That that year. 

Malcolm Hawker 

A few things, yeah. 

Saleem Khan 

Yeah, a few things happened that. Year in addition to. But you know, I think a lot of that drove financial services leaders thinking about, OK, we need to have a strategy if we’re serious about diversity and inclusion, we need to have data to drive that strategy. We need to know what the current benchmark is, meaning, where are we today when it comes to diversity and how do we get to where we want to be? Right to be a diverse organization. And so it came out of that. I think that’s where a lot of. That came from. So customers definitely are one big, you know, source for us, another big source is just reading the tea leaves, right, looking at what is. Three to six months out, what are customers going to ask for? And so we have this product, actually a third party partner that we work with around advisor intent, right? So intent, you’re familiar with the marketing space marketing intent data. 

Speaker 

Yeah, yeah. 

Saleem Khan 

In the same way, asset managers are now starting to use this intent data to figure out which advisors should I be going after. And so we saw that in this world of asset management, intent was not even being discussed. Right. But we knew that six months from now or a year from now, it’s going to become an imperative because everyone’s working from home. You need to find a way to figure out this is the audience. I have to go after without having to fly all over the country to do that. Right. And so we developed this intent data set and it’s been something that we drove. Internally, we saw the need. For it and that then grew into something that was actually pretty big for us from a, from an ACV or annual cash value standpoint. So you know our products went from. Being you know you know mid priced products to being exorbitantly priced partly because of the richness of this intent data, right, they just brought a brand new way to surgically focus on and on audience. And then have that audience actually convert into paying individuals. So it was a very different sort of data set. So anyway to answer your question, there’s two sources. You know, customers definitely, I keep my ears open and listen to what they’re saying and what they’re not saying too. And then the other part is just having a sense of where the puck is headed. Right, that that is the two areas. From a product development standpoint? 

Malcolm Hawker 

I’m Canadian, so I get the Wayne Gretzky metaphor of where the puck is going to be, yeah. 

Saleem Khan 

Speaker 

But but do. 

Malcolm Hawker 

Do you have product managers working in your team or are they they called analysts? Or are they more customer? You know, customer success type people? What are the roles there? 

Saleem Khan 

So we do have product managers. So I do have a couple of product managers on. Team, they are more, I would say the traditional Agile product development sort of folks. So folks that will take requirements and convert those requirements into you know Portal feature enhancements or API enhancements or enhancements in Snowflake. But that spark of that. Idea could come from anywhere. It could come from me, could come from the product manager. It could come from a data subject matter expert. Right, we might have. You know, we’ve got a couple of subject matter experts on Canadian data. For instance, Canada is unique in that they do not. Canada has an interesting marketing stance in that you cannot market to individuals in Canada unless they actually opt in to that marketing. 

Malcolm Hawker 

Right. 

Saleem Khan 

And so the question becomes well. How do I get them to opt in? Without sending them an e-mail, and so conferences and other mechanisms become the way in which you reach these individuals. 

Malcolm Hawker 

Right. Exactly right. 

Saleem Khan 

And so that came directly from some of our subject matter experts on Canadian data. So that’s how we developed that particular product. But yeah, the the ideas come from all over and as a leader, I think part of the the job of a leader Chief Data officer is to synthesize everyone’s ideas into something that is tangible, something that you know, I always think as a salesperson, I try. To at least how do I put shrink wrap in a bow tie and a bow on this thing and say you? Know, here’s your here’s your. Final product go off and sell. Happy selling you. Know because because data sales can be very complex depending on the complexity of the data. So the more you can shrink, wrap it and and have a good narrative around why this is helpful for a customer, the better. 

Malcolm Hawker 

How much of your process is related to kind of training around the use of that data? So a a big theme right now in my world? Is is something otherwise known as data literacy, right? Like this notion of? Of the people out there who are consuming the data, don’t know enough about the data and that’s why they’re not getting value from it. I happen to be vehemently opposed to that phrase because to me it it turns it it turns the problem from to to a user problem instead of where I think it should be. On the creation side, right? If you’re creating a data product, if you’re creating something for consumption and people can’t consume it, they don’t know what it is. They don’t have to use it. They don’t have to drive value from it. I would argue that’s a product failure, not a user failure. What what are your reactions to what I just said and and and kind of? Large just in the theme of of of literacy and how you get your clients to the point of comfort of using your products. 

Saleem Khan 

Completely agree. I I think so. There’s different mechanisms. I’ve become pretty well versed in the sales space and using the Sandler method has become a big one. I don’t know if you’re familiar with that, but it’s it’s a lot of scripting. 

Malcolm Hawker 

I’m not. 

Saleem Khan 

It’s a lot of rehearsing. It’s a lot of, you know. So there’s a lot of player, coach sort of things that have to. Happen where I work with the sales team to to have this kind of talking point document so that they have these key points you want. To hit on the questions, become. The first thing a customer wants to know is what value do I get? Am I? Am I getting some return on investment? Am I increasing revenue? Am I decreasing headcount? Like what? What what is the value right you start with the value based conversation 1st and that for a salesperson is gold, right? Because they want to start with that value. 1st and then you can go into the bells and whistles and the features of how you derive that particular value. And so I find that. Your point when when you have a product that has a miss? There’s one of two reasons one could be that it was just a terrible product, right? Two could be maybe wrong place, wrong time. Maybe you’re a little too early with the product, but the products that do work out, they tend to work out partly because the chief data officer. For the head of product is working directly in tandem with sales and marketing to create that narrative, to tell that story of value to the cup. Customer and making sure that that customer has a very simple and crystal clear understanding of how this data set or how this product is going to help them achieve the outcome they want to achieve and so that that’s been my approach is. I I’ve I’ve. Had to kind of morph from being, you know, look, I was a data nerd for a very long time and a technology. That’s where I love to be, but I had to pull myself out of that and say, OK, I’ve got to kind of be a salesperson marketing person. Think like how? How would they go about selling this product, right. And so that kind of I think the CD overall, especially at a data company has evolved. You’d have to now think in that that manner if you’re going to survive. 

Malcolm Hawker 

Well, I would argue so when I was a Gardner, I was having these conversations and I was telling my CDO clients all the time. You need to think like a marketer. You need to think like somebody who may even be like in. A PR role of how do you get the message out there? How do you tell this and you use the word narrative? Love it. How do you tell the narrative of how to derive value from these data products? I think I think if companies within internal internally facing data and analytics organizations adopted. More of the practices like what you’re talking about, whether or not it’s a framed model there, there are other kind of their models. I love the framework by the way, or whether it’s going to be more kind of, you know, selling type methodologies. I think that that could do nothing but. Good for for kind of other kind. Of more traditional CDF? That’s that’s just fantastic stuff. Love it. Let’s let’s transition a little bit. You know, I mentioned kind of blockchain II, I fell down the blockchain rabbit hole a couple of years ago. I wrote an article for Forbes about a month ago now, and the title of it was how blockchain will save data governance. Yes, a little clickbaity little pithy I I get that. But but I. Actually, believe it. I I know that you were looking at this. Stuff a long time. Ago and long being 5 plus years ago. Where do you see things now of blockchain, kind of through the lens of B2B through the lens of maybe data management? Is it, where are we? Is it still? Still coming is it is it still imminent or have the wheels falling off or what? What do you think? 

Speaker 

Yeah. No, I. 

Saleem Khan 

I’m I’m a big champion. I’m a big. Fan of blockchain? Especially in the data management data governance space. I think look, 560 actually seven years ago now, 2015, it’s crazy how fast time flies. But but when we started, you know, the former company doing work around blockchain. 

Malcolm Hawker 

Yeah. Good kidding. 

Saleem Khan 

You know, it was in in its infancy, to be honest. You know it. We had a theory and we were using Ethereum as our our blockchain. It had a programming language called Solidity and what we were doing there was essentially writing smart contracts for businesses that we were tracking and the database behind the scenes was not Ethereum. It was actually I PFS. And so IPFS, the IP file system protocol it’s called essentially is like an open database, right? And So what we would do is we’d have like a, let’s call it, a rolodex card for IBM, right? And that would point to an IPFS file location to say. Here’s all the data on IBM they’re located in. Armonk, New York, and you know, here’s the CEO, and here’s all the information about them. But back then? You kind of. Had a separation between Ethereum as a blockchain and where the data was actually stored. Because in Ethereum there’s a limit to how much data you can actually store within a a contract, right? I think it was like a 10K limit or. Something like that. 

Malcolm Hawker 

Still is because the. The the the change would just get massive right? The the block. 

Saleem Khan 

Yeah, it it would. 

Malcolm Hawker 

Be massive, yeah. 

Saleem Khan 

And verification would take forever and you know we were. 

Malcolm Hawker 

Yes. Yeah. 

Saleem Khan 

We were just doing single node validation back then. So this is not. Like truly a distributed Ledger, and that, you know, you’ve got multiple nodes validating. That’s truly IBM and Armonk, New York, right, that that’s a true, you know, blockchain. But now you’ve got companies. You know, I recently noticed, like a big chain DB. I noticed that Mongo DB has gotten into this space and a lot of these. Companies, what they’re doing is. I I think where where the data housing and sourcing is headed is very similar to what I call. The Spotify model. And so when you think about that music player, the Spotify music player, what they did was they essentially created a digital rights management kind of, you know, platform for music. And what they said was they said, look, the music industry is fragmented. You’ve got all these record labels. They’re putting out all these records, you’ve got to go to this store to get a record there. You’ve got to download something from here and there. Spotify just made it easy, they said. Look, just pay 10 bucks a month. And you get. All of your content in this one place, and so I think when you think about blockchain. What blockchain is going to do for databases is similar to what Spotify did for music in that you’re going to be able to take in datasets from multiple sources, create contracts for the usage of that data right to say that every time you use this particular data asset, you know X amount has to go back to the originator. Of that content. So you’re going to have this sort of like capability where you can price and digitally store and do the rights management and governance around data. And so, you know, I look at some of these newer technologies now and I’m I’m kind of wowed by how far it’s come in the past set. Years, I think now more than ever, it’s become viable to start storing data within a blockchain native system like a like a big chain DB or MongoDB and not necessarily endorsing any of them, but just kind of using them as as an example as databases that are using blockchain as a mechanism. To do the digital rights. Management, but also pay back the content creator, right? That’s a big part of this because there’s there’s two sides to the coin here. There’s the person that wants to consume the data. The person that actually went through the the trouble of normalizing, validating, scrubbing all the work that goes into getting data just right so that you can consume it and get the right result, right? So there’s there’s. Two parts here, but that’s my two cents on where I see. 

Speaker 

It’s so so. 

Malcolm Hawker 

I love it. It is. It’s amazing to. You brought up DRM and that you brought up Spotify. In in my largely unsuccessful attempts to explain my thinking here, particularly to friends of mine who have the well, you know, OK, who is going to pay $1,000,000 for a JPEG of a board app? And beyond that, why would I do that when I can just right click on it and download it anyway, right? That’s kind of that’s kind of what I hear. When I, when I when I go on my rants about. And what I always say in response to that is, well, a long time ago, when, trust me, this is real Gray hair. When I worked at AOL in 1996, we were saying the exact same people were saying the exact same thing. Why would you distribute me? Why would you even bring your music into this platform? If people can literally just right click it and download it a zillion times and send it to a zillion people. That was a. Valid question at the time and then along came DRM and you can say whatever you want about DRM. I’ve I’ve got gripes about DRM from a fair use perspective and other things related to copyright. Law. But putting that aside. DRM said, hey, it’s this person has the ability to do this to this piece of copyrighted music and it it. It missed the, I would say it kind of missed the compensation and incentives piece. It just did the rights. Piece again, maybe a separate issue, but blockchain does that quite well. It doesn’t send us and compensations, really. Actually quite quite well. And what I see we’re going on right now is DRM being built for blockchain, whether that’s. A smart contract. Whether that’s something else or whether that’s a layer two or layer three that’s sitting on top of these smart contracts, I see the exact same thing happening where? In another few years, you won’t be able to right click on that Board 8 and download it, because whatever infrastructure you’re using to look. Edit will have integrated that DRM software and we’ll be able to and apply those rights, because if they don’t, they’ll be in violation. Of copyright law. So I I love the metaphor of Spotify, and I didn’t know about MongoDB, but like Amazon has a Ledger database, Microsoft has a Ledger database. That that are being used for these things. So I I couldn’t agree more. You you what you laid out is is, is is bang on to what I believe will eventually be. A world of kind of just hyper decentralized data ecosystems. I’ll. I’ll just. Call it a data ecosystem, for lack of a better word. We’re we’re companies potentially, are willfully sharing certain data assets. What do you think about that? When I, when I, when I ask people, when I say OK, well, I’ve got this idea, but it’s going to require some form of sharing of data and maybe even a little bit of transparency that kind of goes along with blockchain from the perspective of sharing data. What what do you what do you? Good to hear that. 

Saleem Khan 

And look, I have to preface this by saying these are my own opinions on blockchain, not my companies. 

Malcolm Hawker 

Ditto Ditto, yes. 

Saleem Khan 

But yeah, I think I think you’re right. I think what’s going to end up happening is in the same way I’ll use the Spotify analogy again. You had you had fragmentation in the recording industry, right? And the fragmentation caused inefficiencies for the. End user the end user. They didn’t want an entire album. From, let’s say I don’t know Nirvana, right? 

Malcolm Hawker 

Spice Girls. 

Saleem Khan 

The Spice Girls? There we go. That’s a good one. You wanna just the one song they liked? But then they also liked a song from another band and another band. What became important? Was the playlist right? That became the important thing. That was the killer app. I think for something like Spotify and so I see a similar thing playing. With data right now, the data data industry in general is incredibly fragmented, right? You’ve got like hundreds of data providers, but the end user wants a little bit of this data. A little bit of that data they want. Their playlist of. Data so that they could synthesize it into some sort of analytic or some sort of outcome some. Sort of report. Right. And it’s it’s never one place that you can go to get data. And so I think blockchain and and DRM. Play a very. Interesting role, I think over the next five years. Of getting the inefficiencies out of sourcing the data, getting it into one place, and then being able to use it to synthesize some outcome either in analytic output or some sort of, you know, derived output that helps you to reach some outcome. So yeah, I mean just to sum up, I think, you know, five years from now. We might be talking about something similar to Spotify. We’ll call it data FI or. Something like that, right? 

Malcolm Hawker 

Yeah. Well, what? Royalties. 

Saleem Khan 

That’s right. That’s right. If the data file comes. Up we want robot. 

Malcolm Hawker 

That’s right, because we just copyright it. We talked about it. It’s in a fixed medium and. We’re good to go, right? I I I could literally talk about this for hours, but Celine, thank you so much for joining us today. I really, really appreciated the perspectives and the insights from from from somebody who is the CEO, where data is the product. Because as I said. About four or five. Times in our conversation today I I think that all CEO’s could learn from this if they want to implement more product centric disciplines within their organizations, if they. Want to build data. Products, if they want to be more customer driven internally, I think following a model like what you share that frame model is a fantastic first step. So thank you so much for coming. Thank you for spending a portion of your day with you with us today. It’s good to see you. 

Saleem Khan 

You too. 

Malcolm Hawker 

All right, let’s keep the conversation going. Great to see you. Thank you again.