PAW 2019 Podcast Series

PAW 2019 Conference Series – Gerhard Pilcher – Elder Research

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 Gerhard Pilcher, CEO of Elder Research.

Find Gerhard Online:

Email: gerhard.pilcher@datamininglab.com

LinkedIn

Elder Research


[00:02]

I’ve been working with data in the technology field, but in a weird…  I have a really weird… I have a really crazy career path. 

[00:11]  

All right.  Let’s hear it.

[00:12]  

Before I went to technology, I was CTO for a telecom firm, inventing the new DSL subscriber line (high-speed over copper).

[00:20]  

What year is this?

[00:21]  

This was in the years from about ’90 to about ’99.

[00:26]

So early 90s, all the 90s.

[00:28]

All the 90s.

[00:28]   

That was a crazy time, right?

[00:29]

It was a crazy time, and things were booming.  It was insane. Well, then had a tragedy sort of happen in my family and realized I wasn’t spending time with my young sons and wife; so, I bought a road construction company.     

[00:42]  

Was that better?  I don’t know.

[00:45]

I don’t know.  It was crazy. It was like being a farmer.  You know you’re dependent upon weather and uh…  But the interesting thing is technology was just starting to come into the road construction business.  So it was fun. I got to do some things like do oil sampling in predictive maintenance on our large equipment, which allowed us to avoid downtime and more costly repairs.  I got to bring in GPS grading. And, instead of having people out there pulling strings with wooden stakes and doing traditional surveying, the operators had computers in their equipment, and they could see where the road grade needed to be.  And people don’t realize how much goes into actually building a road and compacting it and getting it ready for pavement. And so, that was a lot of fun. And I ended up we were cutting roads at a higher quality than the state was able to measure that quality; and ended up getting invited to Washington, D.C., to talk about grading and that kind of thing and how they could bring their quality processes along because they were actually taking quality out of the roadways.  So a little bit crazy career there.  

And then I decided at a late age that I’m going to go back to grad school and study more about this statistics and analytics stuff.  And so, my son was just going to be a freshman at NC State and was looking for a place to live. And I made this decision with my wife, but he wasn’t quite aware of what was going on.  And so, I called him one day; I said, “You know you’re having trouble finding a place to live. I’ve got the perfect place for you. It’s an older roommate, probably can cook and clean a little bit for you, but it seems like a decent guy you know.  What do you think?” He said, “Well, give me his number. I’ll call him and check him out.” So I started rattling off my phone number, and he goes, “Wait a minute. That’s your phone number.” “Oh, yeah, that’s my phone number.”  

[02:27]

That’s a crazy story.

[02:29]

“What do you think about spending a freshman year of school with your dad?”      

[02:34]

I don’t know if I’d like that very much, but it’s funny, as a dad, I’d love that.  So, fast forward where we are now, there’s been so much evolution inside of our space.  What do you see as one of the megatrends that’s going to like the next kind of wave that takes us the next three to four years?  

[02:51]

Gosh, I think the next wave really is about calming down from some of the hype of data science and realizing how valuable it is but realizing that I have to have a foundation to really accelerate data science through my organization.  So I think people are starting to think more about their data strategy and how do I treat data more like I treat cash flow into my business. Cash comes into the business; I want to increase the value of that cash as it moves through the business so that I’m worth more at end of line.  We call that internal rate of return. Well, I think people are realizing that data is in that same space. We can actually use data as a currency to improve our business. And some signs of that and some signals of that, if you think about it… If you go back to the mid-80s and you look at what the difference was between our book value of companies, on average across all the publicly traded companies and what the market capitalization or what the market was willing to pay for that company, there was about a 15%-gap on average (some more, some less, of course).

Now today fast forward, that gap is about 87%.  What is the difference in that gap between the market valuation and our book value?  And many people think we’re not accounting for the value of data. Interesting we’re in the information age and accountants haven’t figured out how to put data as an asset on our balance sheet so we can account for that.  So, there’s a lot of trading of data for other services and things that are all off-balance-sheet and non-taxable and been used as a currency but not being recognized officially as a currency or as value in the way that we account for things publicly, right?  And so, many companies are starting to value data internally, and it helps them understand how to make better investments of data and how much to invest in data     

[04:44]

I have never heard – like I’ve talked to a lot of people – the analogy that you just did of connecting the dollar that goes into a business to the dollar that goes out of a business, right?  What’s the x-return on that dollar? Data has not ever been framed like that. Maybe, I’ve been a bad listener but in that exact way. But I think that’s exactly the way that we need to be framing it, and it’s about how do we maximize the return or the overall value, which ROI on insights is something I’ve heard a lot, but not framed like from an accounting lens.  And I think that’s exactly how we have to be thinking about it. That makes perfect sense to me.   

[05:24]

Wow, great because I’m trying to take that message to CFOs and everything and get them to think that way ‘cause they’re saying, “Why are we doing this data analytics?  Why are we spending this money?” I said, “Look at it the other way. Why do you spend money in your business?”  

[05:36]  

See it as an asset on your balance sheet.

[05:38]

Right.  How do you manage that?  How do you manage that asset?  You’ve got ways to understand that.

[05:42]

And it’s funny.  And it’s more transparent than something like goodwill.  

[05:47]

Right, it’s much more transparent.  You can measure it. Back in ’86, a guy named Appleton talked about data, the data we create like in normal operations or the data we might buy.  Once we start combining that, putting context around it, asking questions, he said it becomes information. Once we start building models around that and making decisions around that, then it becomes knowledge or intelligence for our business.  And we work in the intelligence space as well at Elder Research, and we know the value of that data and increasing value of that data. So it’s one way to think about it. It was actually a Gartner person wrote a book last year called Infonomics.  And I’m sorry I can’t pull the author’s name out of my head.  But it’s a great book, and it gives you some ways to measure data as an asset.  Measures that are used by the M&A community and things like that. And so, I think they are good ways, and I’m encouraging the companies that we’re working with to begin thinking about that and measuring data in that way.   

[06:46] 

Info what?

[06:46]

Infonomics.

[06:48]

Infonomics.  So, tell me about your business; tell me about Elder Research.

[06:52]

Elder Research, we’ve been around a little more than 25 years.  So, we were doing this thing we called data mining that’s now got this new cool term called data science for a long, long time across a lot of different verticals and have gotten a lot of experience.  We actually started out building hedge fund trading models. That’s a hard place to make money and to survive, right?  

[07:16]   

You go to be right.

[07:17]

Got to be right.  And you got to know that you’re right.  And it’s one of the things that we evangelize.  So much of data science today you see done in not a disciplined manner.  So, coffee’s good for you; coffee’s bad for you. Wine’s good for you; wine’s bad for you.  So you just drink both, and you hope they balance out.  

[07:32]      

I do.  If I’m right once and wrong once, maybe to your point, but average I’m perfect. 

[07:39]

Average, I’m good.  I’m going to the mean.

[07:42]

That’s right.   

[07:44] 

So, anyway, so I think that that part of it is really important.  So then, what happened in 2001 after 9/11, our founder, Dr. John Elder, was really well known in the space, and the President asked him to come sit on a committee to look at how our intelligence community could share data better.  So you now have the Office of the Director of National Intelligence, who is supposed to bring data together from our intelligence agencies to make sure we don’t miss another 9/11-style attack. Well, that got us into the intelligence world and a large piece of our business now is in that space, trying to help with counter-intelligence and insider threat and things like that.  Then, we expanded into the federal civil government, working for places like the IRS, FCC, Health and Human Services…

[08:35]  

Highly regulated spaces.

[08:37]  

Highly regulated, Department of Labor, trying to help them with some of their problems and some of the things they need to tackle so that we can use our tax dollars in a very efficient way.  So, that’s a little history.

[08:48]  

Give me your favorite customer story.

[08:51]

Wow!  We have this thing called recency bias.  So I’ll give you a recent one. I can’t really reveal the client, but what was fun about it and ties into what, I think. the questions you’re getting at and digging around here is they asked me to come in and help them with a long-term analytic strategy plan.  Now this is a company that had sort of put together an analytics team, hired 40 people, and then said, “What do we do with them?” Which is exactly opposite the way I would approach the problem or encourage them.   

[09:25]

Ironically, big companies that do it this way.  

[09:28]   

Yeah, yeah, they try to figure it out as they go.  

[09:30]

“We need AI.  Let’s hire a bunch of people, and then we’ll figure out why.”

[09:33]  

Yeah, and then we don’t even have the business questions, and the people are doing this, and they start having attrition and everything, you know?  And so, I said, “OK, well.” They shared with me what their two-year plan was. And they said, “We got the two-year plan. we think, down pretty well.  We want you to help us with three years. Tell us what technology we should use and all this. And, by the way, you have three days to figure it out.”

[09:54]

No problem.

[09:55]  

I said, “OK.”  Well, I’s spent some time with the company and spent some time in their operations to understand it a little bit better.  And so, I really thought about this, and I really thought about it from the technology point of view. And here’s what happened.  If you think about technology, and we’re in it all the time in data science: GPUs, MS processing, natural language processing, and those are some of things that we’re interested in.  What we thought was state-of-the-art four months ago has been surpassed four times in that four-month period. So, what I actually went back to them and said was, “I think you’re focused on the wrong thing.  Don’t focus on technology. Focus on what analytics can do for the business in the three- to five-year timeframes.” So, as an example, one of the things in their two-year plan was self-serve: how do people come in and get some of our analytics and data in a self-serve model?  So, I said, “That’s good. That’s a good two-year plan. Three-year plan: instead of deciding what technology ‘cause it’s going to be different by the time you get there, let’s look at how do you turn that into what I would call instant awareness. So, instead of self-serve, you’re pushing now information that, based on your analytics, you know they should be thinking about.  Think about how that would work in a health care space. It could be amazing, right? And so, now they’re self-serving on things that you’ve already given them information about or a direction you’ve already pointed them in. So, rather than thinking of it through a technology lens, look at it through a business lens and say, ‘What should we be doing in the business in three years?’  And then, you’ll be better able to select a technology as it comes along during that period, right?” So. That’s it.

[11:37]  

That’s super interesting.  Thinking about the space in general, it feels to me like we are, it feels to me like we are entering into like a big uplift in businesses generally becoming more analytics in their decision-making process.  Is that what you’re feeling? Where do you think we are on that J-curve? Are we like at the beginnings of it? Middle of it? Cresting? 

[12:01]  

I think we’re just at that knee of the curve where it’s really, really taking off, and they’re figuring out.  And part of that is because there’s a cultural change that has to happen, all right? And people have to think that way.  And a lot of it is too, we’re overhyped. And I love technology. Don’t get me wrong. And I’m not trying to slam technology or product providers.  But what I’m seeing is, look, they’re out there saying, “Hey, I’m going to give you analytic platforms and things are going to be wonderful.” But, if we hadn’t defined that business problem well, if we’ve just bought a platform and hired a bunch of people and we don’t even know what business problems we’re going to solve or even alignment with our mission and strategy, then we’re going to fail fast, you know, and hard.  And it’s not going to be pretty.  

And then we start getting the data organized and together and visible to everybody, transparent.  Then the data science. We really have great tools in data science to be able to build models once we have data and a problem to solve.  And then, the really hard part that we need to think about is how do we put it into somebody’s hands so it fundamentally impacts their decision process on a day-to-day basis, right?  And we don’t think about that enough. We think, “Oh, we just put it in a BI tool or whatever.” But now we have dashboards of dashboards of dashboards, you know. And nobody looks at them.  So, we hadn’t thought about that delivery process that says. And if you think about analytics, the whole purpose is to impact, to give better information into a decision process. If we haven’t done that, then we’ve failed.  But part of that is delivering it to you so that it makes absolute sense to you; it’s intuitive to you; and I’ve anticipated the next three questions you’re going to ask, right? So then, you can ask those questions and get some confidence in the data that you’re seeing.  But we have to do those things well, and we don’t think about them enough. And then, beyond that is how do I begin to get people to adopt and change over to that.  

[13:45]  

This is an exciting time.  

[13:47]

Yeah, it’s very exciting.

[13:48]

We are probably going to get run over.

[13:50]   

Yes, we are.  There are people all around us, making lots of noise.  Oh, my gosh.

[13:54]

It’s terrifying.  If somebody wants to get in contact with you or someone else at Elder Research, how would they do that? 

[14:00]  

They can go to the web and look for DataMiningLab.com or ElderResearch.com.  And either way, they’ll find us and jump on our website. I’m listed on that website.  Everybody else is. We love to get contacts in, and happy to talk to people. 

[14:15]

Gerhard, thanks for being on the Happy Market Research Podcast.

[14:18]

Super, thanks for having me. 

[14:19]

Everybody else, I hope that you will take the time to screen capture this, share it on social media.  Really appreciate your time and attention. As always, have a wonderful day.