Jonas Christensen 2:45
Okay, we have a very special edition of Leaders of Analytics on today, because today we have not one but two guests on the show. My guests today are Carla Gentry and Whitney Myers, both from the company called Zuar. And I am so excited to have both of you on the show today. So Carla and Whitney, welcome to the show.
Whitney Myers 3:07
Thanks for having us. Yeah, we're already getting to do the jinx it. We're gonna kick it off right there out of the gate.
Carla Gentry 3:16
Well, and that's the thing. I think that's why Whitney and I get along so well. It's that we're on the same page and that's what we're here to talk about. It's getting everybody, small, medium business, not just the large guys, but everybody on the same page.
Jonas Christensen 3:30
Brilliant. So you can hear listeners, we're straight out of the gate here, because today we're going to explore what it takes for businesses to succeed with data in 2022 and beyond. And we are going to cut through the fluff and get right to the detail. But before we get to that, we want to learn a little bit about the two of you. So perhaps, Carla, you could kick us off by telling us a little bit about yourself, your career background and what you do.
Carla Gentry 3:56
Well, it's quite lengthy. 23 years and I've been with a lot of different organisations. So I'm just going to kind of cut to the chase here. I mean, I graduated from the University of Tennessee, Go Vols. Actually, it's at Chattanooga, so it's "Go Mocks". But when I graduated in 1998, I couldn't get a job. I mean, it was like, I'm like sleeping on my sister's couch with my two little kids. And I just happened to send my resume to a company called Ronald J. Krumm & Associates in Oak Park, Illinois. And the vice president of the company walked by right when the fax was coming out. And at the very bottom of it, it said: You know, I'm a single mom. I've dragged two little kids along with me through my entire six years of college because I doubled major. I have a degree in mathematics and economics. And he saw that and said "I'm going to be the one that gives her her break". So thank you, Ronald J. Krumm for giving me that break. Now I started off on a Mainframe and SAS. I stayed with them for quite a while, almost five years. And then you get to that point, you're climbing the corporate ladder, so you want to see what else is out there. And I've did a lot of corporate climbing in the ladders. I've worked with a lot of different organisations and we went from a Mainframe and SAS to ETL and Microsoft chop shops, as they call them. And then what you find is everybody has different versions. This company's got an Excel. This has got an in commerce separated Value kind of thing and it's kind of getting a little scattered. So I thought "I'm going to open my own business" and opened Analytical-Solution in March of 2011, to be able to bring that type of insight to all companies, not just big companies. It doesn't just have to be all Deloitte and the Firestone's and the Procter & Gamble's. What about the little guy? What about the Samtec and the Shoe Sensation and the Talent Analytics? They need help as well. And that's just some of the clients that have been under my wheelhouse over the last 11 years, almost 12 in March. You know, it's been a wonderful progression to be able to go through that and, you know, having your own business and stuff, and then you can partner with wonderful companies like Zuar. And that's what I do for them. I spread the word. You know, I'm letting them know what their capabilities of data migration, data transformation. What is realistic and what we can actually do for you. So that's why this partnership with Whitney and and Joel and the company and all the other great employees there has just been a godsend, because it kind of lets me talk about what I love, which is data. And also my other love, which is Zuar and they can do so much. And I'm so happy to be a part of that. And now I really haven't heard this story before, so I really want to hear how Whitney got started.
Whitney Myers 6:39
On my side of the house, yeah, hello. Whitney Meyers. So I'm now the CEO of Zuar as of the last few months. I have about 13 years experience, which is surprising because I honestly still think of myself as being fresh out of college, even though that's far in the rearview mirror. I began in sales operations. And so a lot of it was taking data and having to get multiple different departments working together. And I cut my teeth in quite a few roles in which I had responsibility but not authority. And for anyone starting out in anything that is analytics related, I encourage you to hone the ability to get results where you have responsibility, but not authority, because later in life, it'll serve you well. Because you'll get people to get on the same page as you, where you have to say something better than ''I'm your boss''. So I started at sales operations at a few different tech companies in Austin. Eventually found my way to analytics. I worked at Tableau for about five years in their sales leadership team, and then joined Lauren 2019. Wow. And it's really exciting. Because Zuar is kind of a culmination of everything I've seen in different roles I've had working with data, because it's this combination of really smart passionate people that realise sometimes we just make the problem a little bit more difficult than they need to be. So we have both developed solutions and technologies like Mitto, where it's kind of an out of the box, all-in-one system. You know, it can connect to data, bring it into a database, transform it. Includes the storage and the staging areas, but then also everything we build has this extensibility to it. So for those who have already made existing investments or have a team of data scientists or have a team of data engineers, they can actually extend these platforms to do some really deep analysis and leverage investments they've already made. It's a nice space to come together. And after kind of 13 years of learning everything that can go wrong when trying to make data become information, it's really nice to be at an organisation where people first have empathy for the problem and then we really try to empower our customers to be able to solve that.
Jonas Christensen 8:51
Yeah. Beautiful, beautiful. So Whitney, if you were to give us your 30-sec elevator pitch, I think they call it, could you tell us what Zuar actually does and why you exist as a company?
Whitney Myers 9:02
Yeah, so we are a data technology company in Austin, Texas. We kind of make two investments in both products and practice. So we've developed solutions, like Mitto in our portals that connects to data, helps you turn that into actionable insights and then present it to stakeholders. Why we exist as a company, because we're not the first ones that come up with the concept of ETL or ETL+ or data staging or analytics, is that we really focus on, and what I'm excited to talk about today, is we focus on how you can quickly deploy these solutions, and then be able to iterate whenever the landscape changes. So many of our customers are fast growing companies where their strategy is changing, sometimes every quarter, sometimes every month. And so coming up with an annual plan or an 18 month roadmap just really isn't feasible. So we help a lot of people deploy analytics for the first time or change their analytic strategy very quickly, using a combination of products and practice. I don't know if that was 30 seconds. But that's what we do. Yeah.
Carla Gentry 10:10
Yeah and one of the things that Whitney didn't mention is its continuous support. Now I'll go to one of our examples. A million years ago when we brought in Cognos - love Cognos, blah, blah, blah - after they left, they left no support. If you wanted support, you had to pay ungodly amounts of money to get support. It wound up costing twice as much as what the actual estimated package price was. And then we still wound up not using it hardly at all. So going back to your comment of you seeing companies spend $100 million on unutilized platforms. I mean, Zuar is not going to leave you stranded. So they're not going to sell you some stuff and then say ''Okay, you're on your own. See ya!''. They actually offer the whole package. So say you have a problem, they're coming up, ''Hey, you know, I'm looking and I've got, like, all of these different connectors. You know, I've got Dropbox. I've got SFTP, which is coming in from our call centre on Five9. I've got HubSpot data. I've got data from Salesforce. What do I do with this?''. And then actually, you'll have a team of advisors that will work with you and set up everything that you need. They'll help you with the connectors and they won't leave you until you have a workable dashboard. I mean, you go back to our website: Zuar.com. And you'll actually look down at the bottom in our case studies and you'll see that there are clients there that will actually tell you, you know, kind of how we help them. So it gets into more than just ''We're gonna sell you some software''. You've been to all of these technology conferences. We're all just patting each other on the back and selling software. What we really need to do is teach you how to use that software and what's going to happen when you run into a problem with that software. Because it's easy to make a purchase. We can all go out and buy stuff, ''Hey, let's go by and let's get in. We're gonna go with.. Let's get in Salesforce. Yeah, let's do this call centre stuff. That's all great. Yay, we spent all this money''. Now you've got data silos throughout your entire company. Your HR departments using PeopleSoft. Your accounting department using QuickBooks. Your marketing departments got 100 different API's connected to everyone under the sun. But now when the president of the company wants to know ''Hey, what's the bottom line? How are we going to do our projections for 2025?'' They don't have a clue. We know that the accounting people have to put their data because if they don't, the government's knocking on the door to do an audit. But what happens when you do all of this data manipulation and transformation and migration and your data doesn't fit or doesn't match what you have over here? So it's great to say ''Oh, we're data driven''. But if you've got people within the company that say ''No, you can't have access to my data. That's our data''. How often have you ever tried to get data from HR? Half the time, those tables are locked. You can't get any information from HR. You go into finance: Those tables are locked. You can't get any information from there. So we have to be able to know that, for us to have company wide usable insights. Data may be king, but insight has got to trump the king. It's got to be even higher than a king. I mean, it's like the Pope or something, I don't even know. I mean, it's gotta be. Insight is what we're doing all this for.
Jonas Christensen 10:24
Absolutely. So business problem first. And Carla, you've already taken us into the topic for today, the main topic, which is hard to businesses actually succeed with data in 2022. One of the things that you mentioned in a recent article on KT nuggets, which was an article where you and nine and eight other AI and data science experts are giving predictions for 2022 was exactly on this topic of data silos. So here's what you said. And I'll just read it out, so listeners are on the same page. So you say ''As COVID continues to be effective in all that we do, companies have pivoted on staffing remote versus in-house data in the cloud or on site. One thing that has become clear is the fact that we are at a crossroads with data and its ability to make a difference. Data sprawl has become a real and costly problem inside organisations and it's hurting innovation''. So, the two things I picked up from that is one that we're at a crossroad here. And two, that we have this data sprawl problem that's really hurting innovation and then your prediction for 2022: ''We will continue this data siloed path since each department will always have its own agenda and needs'', which you kind of alluded to just then, ''Companies that can bring multiple data platforms together will be vital for businesses who don't have the talent, time or ability to tie all this together themselves and make executive decisions. Throwing good money at bad ideas is no longer acceptable. ROI is must be attained. Let us embrace innovative technology. But let us also keep in mind that data itself is useless unless you do something with it''. Now, that's your scene setting for 2022. What are the challenges that we are seeing out there today? And what is this crossroads that you're talking about? Could you elaborate on that for us?
Carla Gentry 15:32
Well, the crossroads is to say that you're data driven, but not be able to get results, as you alluded to before $100 million has been spent on unusable technology. And what we need to do now is, when I started out in 98, we had one platform. We were SAS and we use SAS, and we had a Mainframe. That was it. There were no other connectors in data. But now what we have is we have this these siloed departments, where you know, you have all the various departments of HR, accounting and marketing and that's just the small amount of the actual integrations of the company breakouts. I mean, when you see a schema of who works, where and what department, it's mind blowing. Now, when I walked into career education, they had their schema on the wall. You could actually look and they were arrows and connectors and unique quantifiers or variables and how they were connected on the tables right there, in front of God and everybody. It was very obvious, there was no hidden secrets or anything like that. And we've gotten away from the schemas. Matter of fact, I talked to a company like a couple of months ago, when I mentioned schemas, and they looked at me like I had three eyes. I mean, like, am I talking a foreign language here? Now understand if I start homoscedasticity versus heteroskedasticity. But that's really: Is your data similar or is it unalike? Like, I mean, if you've got a million variables, that variable reduction is to get it down to where it's usable. So all of these different things that are going on in the data sphere, at the moment, this moment of reckoning is how much money are you willing to throw out the window before we finally realise you're not being successful at this. Do you need to call, and in my personal favourite it is Zuar, but do you need to call a company similar to the Zuar? Do you finally just need to give up and say ''We can't hire the data scientist we need''. We already know from statistics of looking at that Hakoda article, the state data and the high cost of data sprawl. And I mentioned that and actually wrote a little bit more elaborately at it in an article for Zuar, on our blog, at Zuar.com/blog, that talks about developments in analytics, data science and data transformation in 2021 and key trends for 2022. So it goes into a little bit more. And what I will talk about is that moment of reckoning or that point where we've spent all this money, we're not getting our ROI. And I know that you say that it's very difficult to look at your ROI. But it's a simple calculation. You have to look at the price of the software, the subscription prices, which a lot of these times you have to pay yearly for them. Look at the cost of all of the data API feeds and all of the connectors that you have and make a decision. We're spending $2.5 million on all of these and we're not getting our money's worth. Look at the data that you're actually getting the most usable information from and then it may be time to cut ties from some of these other guys. I mean, we don't have to collect data from every source out there. Because like I said, when I started 98, there was one source. Now I can just rattle off 100 right now, waste our time in this call doing that, but I mean, we went from ''Hey...'' - what was difficult was ''Hey, we have an FTP site''. ''Oh, an FTP site. What is that?''. ''Oh, it's a File Transfer Protocol''. Now we have SFTP, which is just the same thing, but it's secure. But it's still just a way of moving data back and forth. So when I talk about sprawl, you're taking an existing database of which I wouldn't have had to have like given three children and my right arm for, back in 1998, that IT which has lost control of their data, but IT departments now will let you connect to anything. Marketing departments are connecting to all these API's. It used to just be like, you know, we were doing pay for clicks and we add a third party company that we did business with. Now we get Salesforce and it's got sprawls all over the place. So we have to get to that point, that reckoning: Do we give up or do we go forward? Nobody wants to give up on being data driven. But if you can't get the staff you need and you don't have the talent, then you need to say "Okay, enough is enough. We've spent enough money trying to do it ourselves. Let's go out of house. Let's go third party. Let's look for someone who can actually help us". I mean, there are companies like Zuar out there that that what they do. That's all they do. When people talk to me and they asked me "What do you do?". " I'm a data scientist". First off, I have to explain what a data scientist does and then I tell them "I just play with data for a living. I'm a data nerd". You know, but really what it is, is you have to be the liaison between the C-suite and the IT staff. The liaison between sales department who's making promises over here and marketing who's making promises over here. When you go back to operations and then they go "What the hell are you doing? We can't do that". So stop overpromising and under-delivering. Do what you say. We can't get our politicians to do it, so maybe we can get the corporations to do it. You know, let's go where they won't go. Let's go into the avenue of being transparent. We're not going to tell you how we did it because that's why we pay engineers a lot of money and we don't want you to steal our crap. But we'll give you all the help you need. Now you go and read some of our blogs. We'll tell you step by step how we did some of the things we did. Now, we're not going to give you the secret to our sauce. But you're very informative. I mean, a lot of these sites out there will tell you how to programme in Python. But if that's all you care about is Python, you're a programmer. So know your domain, know your niche. If you build things, you're an engineer. If you analyse data, then you're an analyst. But if you come back to a company and you tell them which direction they should go and what they're doing right and what they're doing wrong using data, then you're a data scientist. So don't think just because you can use Excel and Python that you're ready to join our space. I've been in it for 23 years. I learned something new every day. So never stop learning. And the only way that we get this done and we get this corrected and we move into a positive path, of what you've mentioned, of 2025 is that we get on the same page. We get our sales staff. We got our marketing people. We get our operations. We get them on the same page, because you know, sales staff, they're not techie. They may send you a napkin with some numbers on it. I've gotten that before. And that's not a joke. I've really got numbers on a napkin from a sales guy. So be prepared for them to give it to you in any way, shape, or form, they can. They're doing the best they can, but the salesmen are your bread and butter. So just use that information and try to use it company wide. We have to get on the same page in order for us to be data driven. It's more about really walking the walk and not just talking the talk.
Jonas Christensen 22:30
Brilliant. So what I'm picking up here is the technical foundation is key, of course. The people who work in it, they need to do that. But we need to have a strong leadership built into our data science processes. And everyone is a leader, whether they like it or not in this space, because we are setting the scene and we're moving the company forward using data. Everyone out there, this is why this show is called Leaders of Analytics because we need leadership as much as we need the technical data experts. Whitney, did you have anything you want to add to Carla's comments there?
Whitney Myers 23:03
I mean, Carla, she, like always just hit it right on the head. But if you think about kind of what helps maybe drive progress, as a lot of people look to like the sales and marketing teams, because they're generating the revenue and the growth that allows companies to make investments elsewhere, they also follow the path of least resistance. And so whatever analytic strategy you have, has to take into account export to Excel and the Google Sheets random sheets that are out there that are on a personal Gmail account and a Dropbox that someone set up a trial and then forgot existed. So you have to kind of have this: How do we account for flat files, the rogue data, the dark data, the information that might be on a personal machine and is there a way to automate that? So then the investments in technology that you wish everyone would adopt and these bigger cloud platforms and applications, they're set up with API's. They already are intended to be queried and have the data pulled out. But the majority of customers that we see the problem isn't ''How can I make my Salesforce and HubSpot data work together?''. That's actually a much simpler challenge. It's ''We've invested all this money in the CRM and like an SMS platform and we have Google Sheets running quotas, and being used in forecast calls''. So there's kind of this like, rogue element that if you can find a way, which we have with our solution, but any company if you can find a way to say ''How can we capture even kind of those people that maybe go off the guided path, whenever it comes to working with data and information?'', that starts the process of bringing everyone onto the same team. And in the past that maybe that felt optional, but we're all seeing people write about the great resignation. We know that everyone who left in 2021, in 2022 they're most likely going to be joining new organisations. And every time someone new joins a team, they come with them new ideas, new technologies, new ways and systems of wanting to capture. So you also now have the challenge of kind of flat file management and Rogue data, plus ''Old system meets new'' and how you can get those to give you a single picture that's consistent. So you can actually do year over year analysis versus having every report have an asterisk of ''Oh, well, that's as of March of 2020. Prior to that, we used a different system''. So those are the challenges that whenever I think about, kind of we find ourselves at a crossroads. You know, as Carla mentioned, there's always been a desire to be data driven, but the obstacles, a lot of it really, it's not a lack of willingness. It's this ability to - you almost get frustrated at the onset of the journey. The mountain seems so big. And I have to say, just from personal experience, if the plan to climb that mountain is to say ''Well, we'll just force everyone to follow the process we've outlined and use the technologies we've purchased as instructed'', you're kind of set up for failure. Because you're already ignoring the very human element, which is a lot of the team, especially if they're working remote, especially if you know they're global, located in multiple locations, you need a strategy that it can account for our core systems to give us 70% of the picture. How do we capture the rogue data? How do we marry it with the new insights coming in? And it's a big challenge, but it's like, and I think Carla talks about this more in kind of a more in depth article: Any kind of strategy, you have to allow for this element of iteration and change. Because last two years have taught us, you really just can't expect what you're going to have to do. We didn't just eat our own dog food for fun. And as a marketing pitch, we did it to survive. You know, because we had to make a lot of decisions in 2020 and 2021. So whenever I met Carla, she and I connected because we were both very passionate about the topic.
Carla Gentry 26:57
Can you tell?
Whitney Myers 26:58
Yeah, because I mean, first firsthand, I've seen: I have personally been in situations in which you look in the future and you say "I do not know". And there's so much uncertainty. And the big conversation you want to avoid is anything around layoffs or downturn or shutting down operations. And there's a lot of fear in those kinds of conversations, too. And how you combat that fear is you start with data and you say "If the priority is retention, if the priority is growth, if the priority is being able to pivot, what information do we need? And how can we make the most informed decision and then how can we track our success? So we can calibrate if we were wrong". So I had to do that a couple of years ago, and now do it, you know, every quarter. But it's a much more optimistic place, I think entering 2022.
Jonas Christensen 27:50
Absolutely, it is. But still, as you highlight the world is uncertain, and it feels like we're trying to look many years ahead, as you should. But we're also moving into week increments, because we don't really know what's going to happen next in a global pandemic.
Carla Gentry 28:03
Which is what's interesting is five years ago, I had people that would not hire me because I wanted to work from home. I have an 85 year old mother. Father passed away. It's been almost five years now. She's like my third child now. I mean, I take her to the store. I buy her stuff, just like I would do the kids, but I can't be gone for long periods of time. And in the contracting industry, they want you to travel. They want you to come there, you know. I mean, it's like zoom, and all these Skype and all these options wasn't even an option five years ago. Pandemic hit. We got forced into the situation. So I mean, why is it that we have to be forced into something before we even give it a consideration? But why is it when say we're data driven that we have to wait until we have no usable information and our company is about to die, before we finally reach out to someone like Zuar and go "Help me". You know, I've seen companies in the past where the employees, I want to ask them for like, "Well, do you have the master copy of blah, blah?" and they go, "Well, yeah. It's on my C-drive". "It's on your Cdrive. Really? It's on your Cdrive. Your personal C drive? Okay, great, thank you". But for them, it was job security. They didn't want to share the information that they had. And to her point here about the great resignation, we've got so much people moving around and you had talent here that didn't document anything. When I started programming in SAS in 1998, you would write a line of code and then you would document why you did that. Write some more, then you would document why you did that. Why? Because I was a junior analyst in 1998. As I moved up, I would pass that off to someone new who just started with the company. I'm handing them the keys to the car, but I also gave them the manual showing them how to drive the car. Now new employees come in and I've heard "Well, the guy that used to do this left, took all the technology with him and he didn't document a thing. Can you rebuild everything?". "Yeah, I really don't want to". But yeah, I mean, you get put in these situations. Like we're faced with a pandemic. We have to change everything thing that we think about. So that's what we talked about being predictive and not reactive, meaning like proactive. You see these problems come in, make sure that you guys are documenting the stuff. Required that. I mean, if you're using Python, there's ways to put little marks. You can document why you did that line of code. And that's for all coding. Make them give you documentation of what they're doing, I mean, not bit by bit by bit by bit, because they'll quit because they don't want to do that. It's like, now we have so much upheaval within the company, we really are thinking about self survival. So when we think about self survival, we think about what's really important to us. But now that as hopefully, we're coming out of the pandemic and moving into the new era, we can look to the future. I mean, what happened during the Spanish virus that they had in 1918, Spanish flu 1918? 2022: It was the roaring 20s. My grandmother used to tell me about she was a flapper and she was doing the 23 SkiDoo and they were making bathtub gin and all this stuff. They really made the best of it. They came out of it and they thrived. They just did so many more things that they never thought about doing before. And that's what we need to do. As we come out of this, we need to think about and then Whitney's gonna allude to this later on, but I'll sneak a peek and tell you that it's in AI and machine learning. And please don't overgeneralize. Like my sinus medication says, it works for 24 hours. At 22 hours, my nose starts running. So don't assume that everybody falls under the same characteristics when we generalise. So I'm off my soapbox now.
Jonas Christensen 30:41
That 2022 version of bathtub gin is AI and machine learning. We look forward to hearing more about that in a minute. Now, there's a couple of things here that I picked up on, you both said, which is we have a situation where it's almost like a classic 80/20 problem where you've got 80% of the data in a nice format that's sitting where you can access it in a system. Then you've got this 20%, really important stuff that's living in Dropbox's and Cdrives and what have you. And you also have lack of documentation. You have lack of process, generalised process. Lack of consistency across teams in the organisation, and so on. And part of that is, you could call it pseudo politics. You alluded to Carla that some people do those things for job security to an extent. Nowadays, they sort of hoard the knowledge for themselves. Another element to it, I think that's just as important, is that some of the stuff I've just described here is what you'd call the "boring stuff". The stuff that is not the final solution, but the stuff that makes the solution work for long-term. How do we get people inspired to do the boring stuff and how do we elevate the importance and the visibility of that in organisations? Whitney, what's your opinion on that?
Whitney Myers 33:09
Yeah, I'm laughing because I need to send you a T-shirt, because the last conference we did at the very end of 2019, we made T-shirts that said "Life is short. Automate the boring stuff". Yeah. Because it's one of our core beliefs. Because a lot of what - another thing we joke is that nobody really wants to talk about plumbing, but everybody wants a hot shower. And so much when we talk about analytics is the data strategy side of it, which is the plumbing of how do you connect, architect, transform and ready for analysis. But I think to your point, how people will get excited and be willing to invest in plumbing is you have to find what is that hot shower that gets everyone really pumped and excited and I think that's where you have to think about, "What action am I expecting people to take with this data?". So I think so many of us have this view of reporting, as a "check-the-box" or something we're supposed to do for leadership or our customers or our stakeholders. It begins with the design and the format of what you want it to look like and then you fill it with the data. And I think if you can turn that on its head and instead say, "What am I hoping people do differently because of what they see in this dashboard or what they see in this analysis or what they see in this platform?". Then what you start with is a place of curiosity of "What questions do I need to ask? What I hope they do with the answers? What is the next question these answers will spark?". And you enter this kind of cycle of analysis. That if you come from that mindset, not only does it inform a better data strategy because now you even know what you should be querying in the first place, but it's a lot more impactful with what you're doing. It also can be a little bit of a carrot to get that 20% of data, that is rogue and living elsewhere, into a more easily accessible place because it matters for the first time. So a really key example, Redbox talked about this at a conference,I want to say in 2015. They literally had disrupted Blockbuster, just in time for Netflix to hit the scene, right. And so they had to completely rethink their entire good market strategy and they use data and analytics in order to do that. And what they realised was they needed to hone in on the natural competitiveness of their sales teams. And so instead of just saying, "Here's a pipeline report. Here's a quota dashboard. Here's a..", instead they made leaderboards. They employed this kind of gamification of data. And so suddenly, they had this nine month plan, in which they were hoping to implement a change in how they both rolled out machines, decommissioned some, stocked them differently, and the minute they put it in front of their sales team, they were able to achieve that in about three months. So an entire two quarters faster, just because they put so much thought into "What are we hoping people do with this information?". And they realised, if we want salespeople to change behaviours, and get excited and get motivated and compete, we have to provide analysis that shows them where they stack rank, who are the winners, who are the leaders. Even tying that to like spiffs and bonuses and incentives, what you stand to make by being a star performer, and having it be that, even if it means not having normal KPIs. Because then you actually see that shift in behaviour, and then it bubbles back up into, again, kind of the underlying data: Where does that information pull from? How to communicate it? I would say, that's probably the big thing. The boring stuff is boring, because it takes time, and a little bit of expertise and kind of technical knowledge sometimes to implement. But it stops being boring whenever you see what it can do. If you eat the chocolate cake, you're willing to read the recipe. That's kind of what we do. It's you have to start with: What is that cake? What is that hot shower? What is the exciting thing that's going to make someone be willing to change and see the value in changing? And then you either invest in a product or call a team. Whichever one is, kind of, the preference and then handle the boring stuff. So also invest in automation. Please don't do mind numbing tasks more than once. We aim to try to help someone do something once and then automate it forever. So make that a part of the plan, too.
Jonas Christensen 37:17
So one comment I'll make to the listeners here is when you build, if you're a bi professional, for instance, and you build a dashboard, typical that you think of it as a series of charts that you putting together to look nice, and someone will have access to data. But for the end user, this is actually potentially almost like a software design where you're giving them an interface into a world. And I think your example there of getting this competitiveness into a sales team by using the same data but structuring it slightly differently and visualising it differently, so that you can create that is actually a real sign that dashboards or reports are not just houses of information. They can be real drivers of business outcomes and incumbent upon us as data professionals to think about the data in terms of the technical aspect, but also the psychology of what we produce and how it's going to be used and so on. So that story for me was really important. So, thank you for bringing that up. Carla, any stories from you that you could share that are in a similar vein?
Carla Gentry 38:19
Well, I mean, I totally wholeheartedly agree. It's like everybody wants that hot shower, but nobody wants to know how we get the water hot. I mean, you know, they don't care anything about the pipe. Nobody wants to. When I'm programming, nobody cares if I'm using Python or SQL or Hadoop or whatever. It's "get the answer". Create a dashboard. Do you want to know if your data is right? Publish it to the sales department. There'll be the first one to go, "Hey, that number is wrong". And I love the gamification aspect of it because we all are competitive. We all want that tied to some sort of, you know, I mean, money of course, but self satisfaction too. "I'm the number one salesman for the month", that may not seem like it means a whole lot but with everything that's going on in the world today, that may be the only control we actually have in our life. We can't control our teenage children or our adult children or even our grandchildren or our spouses or our neighbours or our parents. You notice when you're driving, you have road rage because people are like "I'm behind the wheel of a 2,000 pound vehicle and I have power. Get out of my way". But we all want to feel like - Right now, it's like we've all been, like, trapped in our houses for two years. That in itself to be the number one salesperson or the be the number one anything - I mean, the competitiveness of it to feel like you're getting some type of satisfaction from the job you did. I mean, everybody talks about how great data scientists are but we're glorified data janitors. We make sure that the data is clean and usable. That doesn't mean if there's missing data that I can go in and assume "I can input that data". If it's missing, it's missing for a reason. The customer didn't give you any information. So I cringe every time I hear a data scientist say, "Well, we'll just take the mean average of the blah, blah". No. Now you've just screwed up everything and skewed the data. And that's where we'll get into the AI and the ML too. Overgeneralization, which Whitney's gonna talk about in a little bit. But it's like, don't create these titles that are so glorified. I love DJ Patil and think about it: I applied for the chief data scientist for President Obama. I applied as well. He got it. I was like ''Way to go, DJ!''. You know, it's like, we all want to be successful, but in this industry, we need to have everybody succeed. The partners that you're working with: You want them to succeed. Your data scientists and your engineers: You want them to succeed. We have to bring some type of positivity out of this pandemic and get away from this "Everybody's our enemy. I can't be successful unless I take this company down!". Do what you do. Do it well. You'll be successful. Don't worry about the other company. You know, if you want to look in the rearview mirror, "Oh, they're catching up with me. I need to do more". You know, I mean, just like the sales guys. I was number two this month. That makes me be number one next month. That gives me an incentive. And in order for us to all be data driven and survive and move forward. We have to try to think, you know, "I can't do this by myself. It's a group effort. I have to work with my team. I have to work with the entire company. I have to work for the third party. Whatever I need to do to make me successful". So don't down another company. Make yourself more successful, so that you can say, "Hey, I'm number one". You don't need to knock somebody out of first place and think, "Oh, I knocked him out. I just became number one". So we have to be more positive about what we do and the way we work as human beings together. Here we are having a conversation. It's almost five o'clock here. And it's what almost 9am there. But we're having a conversation. I have a 7am call every morning in Hyderabad for a scrum. You know, this isn't just the United States. This isn't just Australia. This isn't just the UK. This isn't just, you know, Russia and China. We're the world. We're one planet that just got our butt kicked by a virus above that kicked the entire planet. But let's put our big girl panties on and let's head forward and thus be successful. Let's move forward. Don't be negative. Let's work together. Let's realise that there are other departments. They're not out to get us. We can share data with other departments but it has to be company-driven. It has to be like you said: It starts with culture.
Jonas Christensen 42:45
Way to go, Carla. Thank you for that rally call.
Carla Gentry 42:49
I get on a tangent. Can you tell?
Jonas Christensen 42:50
Yeah, there's so much in that. One thing I picked up very early on in that comment was the fact that we're data janitors and I think that's really important for people to understand. Because one of the things that happens when a job becomes sexy, and data scientists have been called the 60s job of the 21st century, is we tend to believe that that it's so glamor and fun, and you're going to do cool stuff all the time. But like any other job, there's plenty of stuff that's menial and not that enjoyable. You can automate some of it, but you've got to eat a bit of dirt along the way.
Carla Gentry 43:23
Yeah.
Jonas Christensen 43:24
It's the same as entrepreneurship being glorified by the 20-something billionaire on the cover of Forbes and so on. But actually, it's very hard to succeed when you're in your own business. Data science has gone a similar path. Now, we've talked a little bit about data-driven organisations. Whitney, in your opinion, when can we actually say that an organisation is data-driven? What does this look like inside the organisation and what are the typical outcomes that those organisations get?
Whitney Myers 43:53
Those milestones are going to shift as an organisation grows. But I would say probably one of the easiest indicators is: Do people know what is happening in the organisation? What's changing and if it's good or bad? So for an executive, there's quite a few things, whether it comes to revenue, customer churn, pipeline, different revenue streams, key marketing indicators, there are things like that, where even just that first question of what is happening at my company, if that is not immediately something that can be answered, then you know, you're not really data-driven yet. I think kind of a soft measurement would be a willingness, even within a meeting of stakeholders to kind of show analysis and debate, maybe where the data is coming from or what the calculations are being used, or what the point of the information is versus what so often happens, a question that someone wasn't prepared for. They go back into a room and they say, "I'll come back to you in two weeks or a month with information". So, kind of a softer indicator that you're becoming more data-driven is whenever people feel have the comfidence in real time to actually talk about what they think state of play is and what's changing. From a system perspective, I don't know if there's a right answer. I mean, in an ideal state, you have all data in a database. Everything is fully automated and optimised for analysis, and you've invested in kind of the gold standard of BI. But that feels more like a pitch. I think the reality is: It's kind of like saying, - you know, a little bit of worrying about if you're going to be a good parent is, sometimes just that concern is enough to show you'll be okay. And then what actions have you taken since the last time you thought about that to be better let you know if you're on the right path. So I would say leaders, if you're truly wanting to say, "Are we a data driven organisation?", is you ask yourself, "What type of questions do I need answered? Am I able to get those answers quickly and easily?". And just kind of have a mental account of how many times you ask for something that feels simple, that not a convoluted complex "This has never been asked in the history of business", but a normal thing like "How many deals did we close yesterday?", "Who's our biggest customer?", "How often do people re-up their subscription?", "How are we cross selling across different revenue streams?". Like small things that should be a very quick answer, if you're in that place, you're doing it right. But this is all kind of through commercial business. With not for-profits, it's going to be more: How are we meeting our mission objectives, tracking donors, tracking fraudulent activity to help keep fees down from accepting donations, if they you know, were given using stolen credit card information. So things like that. But yeah, I would say in a nutshell: if you can see a reduction in the time between answering a question, in which you should know the answer to about your business and getting the answer, that is how much you know if you're data driven,
Jonas Christensen 46:54
Nice. So if you're a senior executive listening to this, I think the task upon you is to define what does this look like for you. What are these questions? And if you're not a senior executive, the task is upon you to go and work with the senior executives in your organisation to define this stuff, so that you have a yardstick for where you want to get to. I love that.
Whitney Myers 47:16
I think that's probably a really important point that I just glossed over, which is: I don't think anyone should expect the data scientists to come up with what are the KPIs or the OPRs for an organisation. That needs to be top down. So if that is what is the struggle in becoming data-driven, is you don't even know what your leaders need to see, start with that. Please don't architect a schema and a data strategy and automation process and analytics launch until you actually know, "What are the three primary vital few that my manager, director, VP and C-suite need to see?". Otherwise, you're going to be making your best guess and I don't know if anyone's ever been right in that scenario.
Carla Gentry 47:16
Yeah, but to her point: Your data scientists should not be in a position where they have to make executive decisions.
Whitney Myers 48:08
Yeah, it comes top down.
Carla Gentry 48:10
So to your point about how do you know that you're a data driven company: If everybody that can go in there without being scared to death. Because here's what happens when you have a corporate meeting: You got the head guy up there and everybody's like, "Oh, please don't call on me. Please don't call me". If you can go into a meeting and you can hold your head up and you can be ready for those questions. And to Whitney's point: How long does it take between question and answer. They ask you, "How many blah, blah, blah" and you're like, "Tsch, tsch, tsch". "How many blah, blah", "Tsch, tsch, tsch". Be prepped. Be ready. Then you're data-driven. So if you go into a meeting and not be scared crapless, then you may consider yourself data-driven. If you can answer company-wide questions, without it taking a year and a half. And I say that because I've worked with education data. By the time you get it as two years old. It's not doesn't have to be real time. Because you work with the client. They're like, "Yo, we want real time data". And then I tell them how much it's going to cost. They're like, "Oh. No, it could be a couple of weeks old. We're good". You know, but you have to be on the same page. And that to Whitney's point is you have to have that company support. You shouldn't be making your data scientists make executive decisions about what the KPIs are. That should be something that should come top down.
Jonas Christensen 49:28
Hi there, dear listener. I just want to quickly let you know that I have recently published a book with six other authors, called "Demystifying AI For The Enterprise: A Playbook For Digital Transformation". If you'd like to learn more about the book, then head over to www.leadersofanalytics.com/ai. Now back to the show.
So we've talked here about a data-driven organisation, but we also need a strategy to actually do it as well, and to make sure that we've covered our bases. So what does a great data strategy contain and what does the great execution of that look like? Perhaps we can start with you this time, Carla.
Carla Gentry 50:12
Well, I mean, you have to know what kind of questions you're going to be asking. I mean, and I'm not just talking about "How are we doing as far as our monthly sales?" or something like that. I mean, really, you need to know what's going on within the company. But you know, as far as being for the future, I mean, we have to think about questions that are easily answered. If you want to know what your sales are, I mean, break it down and then break it down by departments, it's easy if you're looking at total sales. But when you start breaking that sales down by categories, those categories have to be defined. So, I mean, think of it like the dictionary. There's a lot of words out there that are used every day that I'm like, "Okay, what does that mean? Let me go look that up". Now we Google it. We used to, in the old days, look it up in the Encyclopedia. But we need to have that type of metadata and the schemas, that data documentation that lets us know: What is the definition of this? When you say successful, what do you mean by successful? How do we define that success? Is that we made a lot of money, or this was what we expected and then we went above? You really have to have those questions before you start collecting data. Because if you want to ask questions, and then you come to me and I can't find anything, our sales are not broken out by category. We work for this electronics company that has 18,000 different components. But we don't break it down to specifics like, you know, "This is for cars", you know, "This is for Ford", or, you know, "This is for your computer". So think about, it's not too much information, it's the right information. It's breaking that down and talking about your leaderboards in sales. How we can look? If we wanted to look at our sales by - we're coming up in February here for the white sales - so, if we wanted to look our linen sales and we wanted to look at sheets versus pillowcases versus ruffles versus, you know, bathroom stuff, it has to be broken down into that way and it has to be categorised. When you talk about machine learning and AI, we overgeneralize because we put things into this big bucket and then we expect it to drip down. These have to be defined. We have to say, "Put these in these brackets and put these together and have them kind of not sideload, but group for classifications". Think about it. 10 years ago, when we were looking at - Machine learning didn't know the difference between a chocolate chip cookie and a Chihuahua. So we had to look at every minut detail of that Chihuahua. And I'm getting into XYZ trajectories and how close the eyes are and the ears, but every little thing, all of those specific things, you need every single bit of that. You have to drill down and then go even deeper and then go even deeper. Because if you want that information to come out like that, then you have to put it in like that. Because your boss later is not going to just want to know, "What's our company wide sales?", "blah, blah, blah, trillion dollars". "Well, how much did we do in this division?", "I don't know". So think about the questions that you're going to be answering and prep for them. You know, let someone talk about the data scientist being a data janitor and having to cleanse but we also have to recategorize data. Like, I'm looking over here, "oh, this pillar case is in the wrong section. It should be over here". So it's about ensuring that your company has confidence in your data. Because I hear a lot of people say, "Oh, we got this database, but we don't use it because it's kind of crap". Okay, well, yeah. "That's a legacy system. We don't really touch that anymore, because it's not very accurate". But you're still paying for all of these subscriptions and for all of these things. So when we're talking about, you know, being data driven, and what we want from all of this, we have to define it, if that makes any sense.
Jonas Christensen 54:08
It does. And I've alluded to it before the show, and you've alluded to it a little bit, Carla throughout this conversation. But in my career as an analytics professional, I've probably seen, I estimate about $100 million being wasted on on data platforms and other things that just didn't have the ROI or they were promising before they got delivered. So one thing is strategy. Another thing is execution. And Whitney, what does good execution look like?
Whitney Myers 54:37
Yeah.
Jonas Christensen 54:38
What does it do to the organisation who's involved? What does it take?
Whitney Myers 54:42
So that's the thing is, and it's kind of a little bit the back half of what Carla said, that you need to identify - I think of things as kind of like people, process, products and then kind of all under this layer of governance and security. So you know, Carla talked about the step of identification. Who are the stakeholders? What is the process today? We think about products: Where's the source systems? Where we made investments? What is our governance policy? So identification is kind of where you make the plan. The execution, I would say you start with a series of: What is that first key or quick win? What is something that in the first 30 days, we can actually make progress on? Please don't bite off. If anyone is new to this, please don't take the biggest behemoth project that your predecessor moaned about, as your first thing to cut your teeth on. A good execution of a data strategy is to identify what are some key quick wins that if we were able to achieve this, it will have been worth it. A good example would be if your number one priority is customer retention, how are we tracking renewals? Do we have a process for customer satisfaction and support? Do we need to actually say how are we tracking cases, CSAT scores and then the actual renewal of any kind of services or revenue? So you identify what are those key quick wins, the stakeholders involved, the process to achieve it and then what products would need to be involved, whether those are resource systems or products you've invested in to help you with that. That's execution itself. And then you also need to think about - definitely I mentioned governance, this thought of who actually is able to access this and who needs the access. And that will vary by organisation, partly because of company culture, and sometimes also by strict requirements. And we've done a lot of work with healthcare. We've done work with financial firms and trading firms. It is not that they want to have like, - there's some data that will always be intentionally siloed, because there are mandates that require it. There's regulatory compliance that they have to meet. And so a good execution of a good strategy is you identify all of this upfront. You say: What are these first kind of quick wins that if we were to - that it's an early indicator of life? I encourage customers to have something called a canary group, just like canaries in a coal mine to see if there's air, as set of key business stakeholders that you roll something out to see, "Is there oxygen in this movement?". And then from there, you almost then say, "We're going to treat these first set of wins as a prototype and then we're going to build upon that to kind of set that next set of milestones to build a really solid foundation to then scale to the rest of the organisation". But along the way, the reason that even as you scale, you have to continue to remember and focus on the people, which was one of the first things you mentioned at the very beginning, Jonas. Because there's study after study that shows the number one reason that analytics deployments fail is the human element. And it's not because people are dumb. People are smart. They're creative. They're passionate. They care. It's that maybe no one's taken the time to help raise their level of data literacy. They know how to do their job, but not how to interpret a report. So this as you grow, part of the planning after that quick win, as you go beyond that canary group of trusted individuals that know how to do something with the information you've given them to the rest of the organisation where this may be something new for them. Where now you're encountering the politics of fear of job security, or that they're going to be replaced, or that you're about to automate something that accounted for 10 hours of work a week. And now they're worried they may not be around instead of the upside, which is now you get 10 hours that to do something you enjoy. As you encounter that, you have to think, "How do I educate my users, in addition to communicate with them?", so that as we continue to expand analytics throughout an organisation, we're bringing everyone along with us. And that's where just as important, you know, we've talked a lot about tech and systems and plumbing, this third element of creating a community. It's where if you see anyone talking about centres of excellence, or internal user groups, it's because they've hit on: The data is one thing. The analytics dashboard is another. The third is you have to have a plan for pulling people together. So the stakeholders and the informed and the data scientists and the people and anyone that's actually a data worker, whether it's via zoom or in person, once a year, every week, you have some mechanism to bring them together for a dialogue. So you can have this internal group discussion around, "You asked for this. You got this. Is this actually what you needed". And that is, whenever you're talking about successful execution, if you get that right, everything else will follow. Because as long as you get people talking to each other, the data and the analytics, that's gonna fall out of it.
Jonas Christensen 59:32
So we have here a bunch of technical people who actually are responsible inadvertently for a master communications exercise across the organisation. I think that is really key for anyone working in data. It's as much about winning hearts and minds than hurting people along this journey, which is on tread path often, It's intrapreneurship and it's innovation and it's thinking out of the box with the stakeholders. It takes more than building the perfect neural network to actually get success with this stuff. And it's the thing that sometimes people will call boring, but it's incredibly important. So I love those comments from both of you. And I want to connect that now to the next topic, which is politics. So let's talk about corporate politics, not the stuff that happens at government level. We don't want to bring that in here. That'd be terrible. But in Carla's prediction for 2022 that we discussed earlier, she stated that the silos will continue to exist, because departments and necessarily the people inside them will always have their own agendas. So if that's true, what does it actually take to be successful in an environment like that? So one approach is to break down the silos, create one team, one agenda across the organisation. Another one is maybe to take those politics as a given and try and work with them in other ways. How do you succeed in this environment, Carla?
Carla Gentry 1:00:54
Well, I think we have to just agree to disagree. There's going to be the sales: They gotta meet their monthly quotas. For the marketing, they're trying to get their web traffic up. HR is trying to reduce turnover. I mean, accounting is just trying to get the numbers right and make sure that everybody's system is working right. Now see, in Scrum and Agile, specifically Scrum, there's a CSPO, which is a Certified Scrum Product Owner. That product owner, which was me when I built the data warehouse for Tera9, you have a scrum master that removes roadblocks and then you have product owners responsible and I'm responsible, no matter what. If Joe Blow over here screwed up, they don't want to hear about that. They just want to know why is my product not ready. So if we think about that company wise, I mean, we don't have a scrum master but we could kind of have if we have a natural leader. If we had some sense of leadership within the company, we would have that scrum master that would remove those roadblocks for us. So if we see that there's a problem, an issue between two departments not wanting to share data, we understand that some of this information is PII and we can't share it. We understand that some of this information is regulatory and we can't share it. Some of this is "it's ours and it's part of the company and we can't transpose down and let everybody know what's going on". But just like we have Scrum in the mornings and we talk about "This is why I didn't get this project done" and then the Scrum master goes, "Well, I can remove that roadblock for you. We'll have a conversation with them". We need that top down leadership that says, "Okay, I realise I have 14 different departments. Not all of you play nice". But we have to make some concessions. How do we become successful is we compromise. We know that everybody has different agendas but we have to have that one overall major agenda, which is company success. There is nothing that one department is doing that's going to be the be-all, get- all. Our salespeople are our bread and butter but there's a Contact Centre and there's all these marketing people that led up to that sale. So we're patting the sales guys on the back, but they might not have gotten that sale if it wasn't for an article that got written by one of your bloggers, or a campaign, an email campaign that one of your marketing people sent in. We have to understand that there are no, "I'm number one". You have to go back to us being "we" again, where we do things collectively for the company. I've had people say, "I can't do that. That's not my department". I don't give a crap if you need me to sweep the floors. I've been in a meeting where somebody put something hot and smoke was coming up and they're like, - Everybody's sitting there looking and going, "You know, there's smoke coming out of that garbage can". Okay, well, I got up and went over and took the garbage can and poured some water in it. And what had happened is, yeah, it was just smoke from the coffee thing but it could have been a fire. We're all sitting around talking about it. So we need to do less talking and more doing. And then back to your point about kind of the canary, we used to say, "Let's run it up the flagpole and see who salutes". If we have an idea, we bring it up and it gets shot down, that's going to make us not want to bring up another idea. So I'll try to be a little more kind and realise that there are more departments out there than just ours. That ultimately what the name of the game is to make this company successful, not to make us personally successful. So we have to get back into the "we" and we have to compromise and we have to be so unlike politicians. We have to be like human beings. So how do we succeed and make it to 2025? Let's be human beings. Realise that our employees have a life. That they have problems. That maybe the reason that their project failed was because they were having a rough time. You know, so a little compassion goes a long way and we can get things done if we work together. When we're working against each other, we're not attaining our goal which is to be successful as a company.
Jonas Christensen 1:05:00
Thank you, Carla for bringing up those points and also for bringing in the conversation around Agile, or these sort of new ways of working, which I think have been hijacked a bit by the number of certifications and clever phrases that you need to have to be a certified scrum master or what have you. But at the core of it, it's really designed to make people work together and have clear accountability and responsibilities. Such decisions can get made without having Chief XO actually having to make a recall on stuff they don't know about, but bringing decisions down to the factory floor where the knowledge is. And the other thing is this element that you also alluded to Whitney, with your canary club example there of actually trying to bring these departments together and work together and create a shared responsibility. So what that really points to is leadership. That is so key to actually executing some of this stuff. We can have all the right data and so on. But if we don't have the right leadership in organisations, there's a big chance that we'll fail. And as data leaders, we are actually the CEOs of data in a sense. So I'm interested, Whitney, what kind of leadership do you see and organisations that do this stuff really well? So I'm talking both here about the data leaders, but also the leaders across the business because it is a team sport, as we've just established.
Whitney Myers 1:06:23
Absolutely. I think probably the thing that I see across regardless of role is there's a natural curiosity. So there's a desire to know something more. And that is what's fueling even the search for answers in the first place. Because we know anything, especially data is a tool. And so the same hammer that can be used to build a house can be used to strike down your neighbour. So the person who's wielding the tool is as important as the tool itself. And so those who really succeed, there's a natural curiosity and there tends to be a desire to build. That doesn't mean everyone who's good is positive and optimistic. But I often think about that George Carlin quote, that "Inside every cynical person is a disappointed idealist". So even the cynics, eventually come around the curiosity. The desire to build or improve or change, people who have that are really successful. Because implementing some of this, it requires change. If the reason an organisation is changing is not necessarily because of this kind of towards thinking of what they could unlock or improve or build or grow, if it's instead much more risk adverse of seeing what they're losing, because they're not, you may or may not find that same kind of natural curiosity. Instead, it's more being incredibly uncomfortable, because they're in a painful position and they would like that to stop. You can still make progress with that. It's a harder road because it means that that improvement is going to be stunted. That that quick win is going to be the last one. Until the next kind of wave of pain comes and then the next milestone comes about. The natural curiosity, a desire to kind of build grow and change and I think we're seeing especially both with kind of new generation coming into leadership, and then just world events and kind of humanity being right on the cusp of our own kind of Renaissance coming after the plague, I think empathy is really important too. So we've talked about, especially when it comes to politics, so much of the reason that things break down is because there's this opportunity to understand someone else that just isn't taken. So those that succeed have some element of empathy, where if they see a dashboard and the numbers are skewed, the first thing they ask is, "Where did these numbers come from? Are we sure they're right?", before having the emergency meeting saying, "This is the big change we need to make because of this data". Yeah, so that's what I see succeed the most. And then I don't disagree with you, Carla, about kind of getting back to the wheat. But I do also think that there is an opportunity, if really done well, for there to also be kind of a personal achievement of either your professional growth or kind of showing where you can shine. I mentioned, you know, kind of a roadblock to growth is if somebody has a very away mentality. If there's also a lot of fear about what the data is going to show, that can be a blocker as well. It's odd. I'm using a lot of emotional language to talk about business and data, but it's because it's run by human beings and we are buckets of meat powered by consciousness. Like, we're very emotional creatures. So with that, just be prepared. If the reason you're changing in the first place is not because of a desire to become data-driven, If it's a fear because you're not and you're already feeling the pain of not changing and kind of being stagnant, it's going to be a different journey to get to where you need to be because the first thing that will have to happen is a little bit of a cultural reset of why is data important to you now. Because again, if it's "Well, without it, we're getting Eclipse", or "We're going out of business or our customers are leaving". And you're trying to say, "What is the MVP version of a data strategy, so that we can get back to businesses normal?", you're just not going to go very far. So there needs to be kind of this adoption of, we're going to go through the process of identifying the stakeholders, coming up with the processes, identifying the products, making investments in change and then sticking with it, because we actually believe in the future that's possible by going down that path.
Carla Gentry 1:10:30
Yeah. And I didn't mean that we can't do the "we, we" thing. What I kinda mean was: It can't be about just one department or can't be about one person.
Whitney Myers 1:10:39
Yeah.
Carla Gentry 1:10:39
I agree. Domain expertise. When we hire people, we hired them because they're their expertise in their field. But I've been told a million times when I gave someone a report, "Oh, well, okay. Yeah, I see the data but my gut tells me this". Okay, when I can start measuring your gut, I'll listen to you. But other than that, if you hired me, because I'm a domain expertise, let me do my job. You want to know why the great resignation is going on? Because we're all sick of doing our job and then having someone else tell us that, "No, you should do it this way" or "I don't really like this. I'm gonna go with my gut". So we have to look at that data. We're talking about being data-driven. You can have all the data in front of you and still ignore it. Somebody at Blockbusters told them that they should probably start thinking differently. They were ignored. Someone at Zillow said, "Hey, I don't know about that AI model that you built over there. I think we may be buying too many houses" and they were ignored.
Whitney Myers 1:11:45
Yeah.
Carla Gentry 1:11:46
So it's the people that get ignored and you don't listen to him, that's why we quit. Because we're frustrated. I'm good at what I do. Let me do what it is that I do and stop telling me how to do my job. So that's what we need to do. We need to let the departments within themselves shine. But then we need to be collective and like that, you know, Scrum session, where we do have that Scrum master. Someone who has that overall last decision that says, "Hey, this is a roadblock and I just can't remove it. We're going to have to do something else".
Jonas Christensen 1:12:18
Thank you, Carla, for that insight. I have now three questions left for you. We're towards the end here. And of course, I've saved the big question for the end. So I'm interested in both for your views on this here. And we can perhaps start with you, Carla. What is your vision for the evolution of data between now and 2025?
Carla Gentry 1:12:41
Well, I do know that artificial intelligence and machine learning are going to become a big part of our future. Like you said, if we're doing something that's mindless, automate that crap because nobody wants to do it. That's why we can't get anybody to do it. If it's something like a chatbot, if you get 1,000 calls a day and 999 of them asked you the same question, why not automate that? What you're doing is you're freeing up your salespeople in your call centres to be able to handle problems. We all hate the fact that we have to push one for this and push three for that, that automation gets us to being able to actually get problems solved. So we have to think about: We automate the boring stuff to Whitney's part. But we need to really use that time that we saved by doing the automation to make a difference, to make a dent to make an impact to change things. And when you do something, the sample size is what kills me. I hear people, "Oh, we did this study and we had three hundred people in this study". Three hundred people, really? What are there, like 4.5 billion people on this planet or gotten close to 5 billion yet? Pick a representative sample of what you're looking at. If you're looking at diabetes, you need thousands of people who have diabetes and then thousands of people who do not have diabetes. Now we can do an accurate study. So I know, Whitney is gonna go more into that automation thing but for me, it's about be careful how much you depend on these things. Hey, I do a lot and I've seen a lot. I've lost count. I mean, I'm sure we're into zettabytes by now the data of that I've looked at in the last 23 years. I can take a mouse with a scroll feature, scroll through millions and millions of records. You know what happened? It's a pattern. Everything that's in tabular form, if I look at it and this is all character and this is all numeric, as I'm scrolling, I can see that there are numeric in the character and their character in the num. So look at your data. I've went back to my engineer and said, "Well, I looked at that data you sent me". He said, "Oh you mean you really looked at it?". They're so surprised that we actually opened it up and looked at it. We just do this and people do this and then we pass it off. Open it up, look at it, see if it makes sense. You know, give it a little twirl there. Look at what's going on. So let's be realistic about what we're asking because if we're asking our company to be data-driven, that each and every person within that company must think data-driven as well, so don't expect us to be data driven when half the company is flying by the seat of their shorts. So my view for the future is those companies that will be successful are those that have learned how to work together and put their differences besides and can actually have productive meetings and not just meeting for the sake of meeting.
Jonas Christensen 1:13:47
Brilliant. So in summary, automate the boring stuff and start working together, everyone.
Carla Gentry 1:15:46
I can't steal Whitney's tagline. That's hers.
Whitney Myers 1:15:49
It's okay. Once it's on a T-shirt, it's the world's.
Carla Gentry 1:15:51
Yeah, it's a T-shirt. Yeah.
Whitney Myers 1:15:52
And I'm pretty sure we literally sent the T- shirts to print and then somebody said, "The kind of definitive Python textbook talks about that". And I was like, "Well..."
Jonas Christensen 1:15:54
We came up with it, too. So, Whitney, your prediction for 2025.
Whitney Myers 1:16:10
I do think that probably over the next four, - Gosh, I guess now it's three years, - we'll see where data is coming from change. So as we invest more in AI, as we invest in machine learning, as it's not just as a bunch of nested "if then" statements, as it actually begins to work, there's going to be a lot of data generated by these machines and processes that never touches humans. And so how we calibrate, it means we're fast approaching a place in which robots are talking to robots about what robots are doing. So kind of finding a way to understand and calibrate and augment data that was not generated by any system that's really being controlled, I think will be a new challenge in the future. But more importantly, I think, also, what's going to be different about data is kind of what we think about when it comes to security and retention. You see this in the UK with GDPR laws. A lot of people in Europe are talking about the right to be forgotten on the internet. And so even as we have new and diverse systems generating data at a rate and clip faster than ever before, you see all these large companies springing up to savour the answer to help you store it. And there's now this kind of third wave and movement of you actually have to segment and silo it, and then forget it upon a customer's request. My hope, my positive hope for data in 2025 is what I have noticed in my career, is that any kind of data analysis is only as good as the capture. And I think our ways of capturing data are becoming much easier. And I think by 2025, we'll get to a place where it really just is as simple as speaking and it's captured, translated, catalogued. And the moment we do that will be the moment we have an actual perfect Salesforce forecast. Whenever we don't have to have sellers typing in a pipeline update. When they can just - you can record a phone call, and then just have that translate into a CRM. I see data involving both who's producing it, how it's being captured, and then also what we are allowed to own and store? And that doesn't even get into kind of that bigger conversation too, about who even owns the data that's being generated on an individual level. I think that will also change over the next few years, as I think so many times corporations are being held accountable by government bodies. And I think lawmakers are finally getting to a place where they realise they need to be educated in what is actually even being captured and shared and sold. And that will just trickle down. Yeah, that's very big. A lot of things will happen in the next three years. I see
Jonas Christensen 1:18:50
Lots of things on the horizon. And I think this question could be a whole episode in itself.
Whitney Myers 1:18:54
Yeah.
Jonas Christensen 1:18:55
But we are at the end of today. But I have two questions for both of you, which are short and sweet questions. So the first question is one that I always ask the guests on here and that is to pay it forward, in the spirit of all working together. So who would you like to see as the next guest of Leaders of Analytics and why?
Whitney Myers 1:19:14
I would absolutely love for you to interview Kelly Wright. She's the CEO at Gong. Previously, she was the global sales leader at Tableau. And then she served on a few different boards. You want to talk about how analytics shapes and rapidly grows businesses and the role that it plays in going to market and also AI and machine learning is now doing that in the selling process, I think she'd be fascinating. She's been a mentor for a long time and she'd be a great guest.
Jonas Christensen 1:19:40
Brilliant. I will definitely reach out to her, Whitney. Thank you for that recommendation. Anyone from you, Carla?
Carla Gentry 1:19:46
Well, I actually and you may or may not have heard of him. He is the Seattle data guy. He is Benjamin Rogojan and that's R-o-g-o-j-a-n. He is just a remarkable guy. He actually just quit his job at Facebook. And he's like, we actually tried. I was like, "come on join us at Zuar". I mean, he's gotten phone calls from Snowflake and Hadoop and everybody under the sun. I think he's decided to just do contracting but he is just a really cool guy and he's on Twitter. And it's SeattleDataGuy. And I actually introduced Whitney to him as well. And another great gentleman, Brian Wallace. He owns his own company and the name of his company - He's the founder of NowSourcing. We've actually talked with him quite a bit. We have a common love, which you'll see on one of our blog sites. We actually talk about cannabinoids and cannabis and the Medical beneficiaries of getting us away from this very addictive problem that we have with pills and opioids and things like that. So two wonderful, of course, my favourite, of course, is Joe Sellner. But she says, "Well, he works for us and we probably can't talk about him". He's actually one of the co-founders of the Zuar, so we don't want to be greedy and keep it within Zuar. But Brian and - Well, both of them, Brian and the other, you know, Seattle Data Guy, they're both wonderful gentlemen.
Jonas Christensen 1:21:16
Excellent recommendations. Carla, thank you. And I will be competing for space in the Seattle Data Guy's inbox, I can hear so maybe forward to that. Now, last question of the day: Where can people find out more about you and get a hold of your content? Perhaps you can start, Carla.
Carla Gentry 1:21:32
You can go to Zuar.com. We have blog. I'll have all of our services there. We have case studies. We have actually learning and tips. I think Whitney may go more into the community and tell you about how if you're actually working with us as an active client, what we can do with you, but Zuar.com is a great place to start.
Whitney Myers 1:21:53
Yeah, I would say the same thing. Zuar.com. Zuar.com/blog for our blog. Community, that's Zuar.com. It's where we have our customers get together to talk about topics like this. So we even have on our site that you can contact us to see a demo or just make a new friend, because at the end of the day, we're all data nerds. So Carla fits in perfectly around the team.
Carla Gentry 1:22:19
Now, you know, if you're working with Tableau and you run into a problem with Tableau, you can go on to their Tableau community site and voice your question. But what if your question has never been answered before? But what if your question has a million different responses to it? You're gonna get lost in the community. What happens here at Zuar is if you have problems in the site, you could reach out to one of our experts and they'll tell you exactly where you can find that information. So like again, it's not the "Set it and forget it and here you bought your software. Adios, we're out of here". We're also going to support you throughout the project as well. And that's why on our site, you can also see that there are community places where if you have questions, if you're having a difficult time trying to solve a problem, get on there and someone in the community has faced that problem before and if not, we'll figure it out.
Jonas Christensen 1:23:15
And I have been on Zuar's website and there is some really interesting blog articles on there. So I do recommend that listeners go and look at sort of their disparate type of information on there but all relevant for the analytics professional of 2022. Carla and Whitney, thank you so much for being on the show today. I really appreciate it. It's been a wonderful time for me. I've learned so much. I look forward to learning even more about you guys on social media because you are both very active on social media I've noticed. So for listeners, go and check out Carla and Whitney's as well on LinkedIn and Twitter. And other than that all the best for the future. I look forward to seeing the data world expand in 2022 and beyond.
Carla Gentry 1:23:57
Thanks for having us.
Whitney Myers 1:23:59
Fantastic. Yeah, thank you. It's been a pleasure.
