Transcript
Dave Mariani: Hi Everyone, and welcome to AtScale’s Data-Driven Podcast. And today’s special guest is Tom Redman. And Tom Redman, otherwise known as the Data Doc is the President of Data Quality Solutions. So first of all, Tom, welcome to the podcast.
Tom Redman: Yeah. Thank you so much for having me, Dave.
Dave Mariani: Yeah. It’s great to have you, I know you just wrote a piece for the Harvard business review. That’s called a, your data initiatives. Can’t just be for data scientist. And so really want to talk a lot about that and, and, and, and the content they’re in, but, you know, before we get, get to all that, Tom, just, I always love to sort of open these podcasts up with a little bit about, about you. and, so tell us Tom, about yourself, about your company data quality solutions and, and your path into data and analytics.
Tom Redman: Yep. Well, I’m happy to do that. And first of all, I mean, nobody’s had it better than me. You know, I, I met the girl I wanted when I was 20 and we got married and off to graduate school a little bit. After that, I was trained as a statistician. I went off to Bell Labs, with an assignment to work with teams who were trying to get more bits through the network, as the network became less analog and more digital and, and, more voice, but way, way more data. And, and it was a time at Bell Labs where, you know, you really expected to look out for problems that may be facing telepany and pull on threads and see what you got. So I’m this hardcore, you know, statistician studying network performance and, and, and, but, one thing led to another and I, and I pulled the thread that led to data quality.
Tom Redman: And it is really interesting that, you know, at and T could tell you within three, how many of the, you know, tens of millions of phone calls got dropped, right But, but of the data, wasn’t very good and it impacted company performance and it impacted in some cases, the running of the network. And, and so I was lucky enough to set up the data quality lab. And for me, it really was the perfect job. I had one foot in these enormous 80 and T problems and the other in a lab setting where we’re trying to figure out why, what worked worked and why, what didn’t work, didn’t work and develop the underpinnings of, of data quality. And, and frankly, I never would’ve left that job, but, but at and T encountered hard times and, and, at some point, you know, one of the divestitures or try investitures, I left and set up Navesink and, data quality solutions.
Tom Redman: And, and, and basically I’m an advisor. companies are loaded with data quality problems and, and, and by and large, they know this, everybody knows garbage in, garbage out, but they’re not attacking those problems correctly. And, and there’s some organizational issues that get in the way. And, and of course, so that’s where I spend most of my time. I spend some time on data science issues too, because the, the, the organizational issues are strikingly similar. And, and boy, do we, have we meaning, you know, companies, and then we society have some enormous opportunities in attacking these problems correctly.
Dave Mariani: Yeah. You know, that’s, that’s an amazing background being you spent what 15 years at, at, and T bell labs, and data quality. My gosh, it’s like, that was even, you know, that was even at the birth of the data warehouse probably is, is, you know, we’re where you had started. So, so, you’ve been working in data quality for your whole career. So let’s just start with a definition of data quality. So what is data quality Tom, what’s that mean to you
Tom Redman: So let’s try this by, by way of example. And, and so, you know, just suppose you’re, you know, you’re, you’re on your commute home one day and you get a call from, from you have a teenager or your teenager’s principal calls and, and says, Dave, I’m, I’m sorry, but, but, your teenager has been in a fight, you know, we’re not sure if, if they caused it or not, but mandatory suspension for a week, kind of thing it’s about, okay, you know, you’re trying to wrap your head around this and, and you get home and, and you call, you call the kid down and you say, you know, well, what happened in school today Right. And the kid enthusiastically says, dad, I got a B plus on my Spanish test. Okay. And, and, and, and the reason I tell this story, it, it captures, a bunch of little things around data quality. First of all, you don’t, you’re not questioning that the kid got the B plus, you’re pretty sure that’s correct. Right. But the kid did not answer the question that you wanted answer. Right. Th they did something that was relevant to what was going on in your mind. And by the way, you knew that, and the kid knows that.
Tom Redman: Okay. And, and so like, like quality is not just, is the stuff correct Right. Quality is, is it relevant to the task at hand Right. Do we have enough of the stuff Is it, you know, in some cases it’s, it’s free for bias. Everything in the data, quality space is context dependent. You know, what, you know, what are you, what are you trying to do with it And, and the simplest way I know to wrap this up in a, in a definition is, is that, you know, data are of high quality. If they’re fit for purpose and explicitly fit for purpose includes, is it the right stuff Right. And is the stuff right And, and, and, and, and at least those two aspects bare and every data quality problem. And then sometimes there’s, you know, there’s other ones, a lot of AI problems now are demanding lots and lots of data and, and, and bias free data. I mean, again, depending on, on what you’re trying to do. And, and, and that’s the idea that this fitness for use is, is trying to capture,
Dave Mariani: I love the fit for purpose. That’s an awesome, I love that phrase. And then in your example, obviously, there was a lack of completeness there wasn’t there. So the kid actually didn’t mention the fact that, you got suspended just about the fact that got a B plus on a test. so you’re right. The, you know, data quality is a multifaceted thing, and it’s not just whether the data is correct. It’s it’s, I like your, I like the umbrella of fit for purpose, because that really covers it all. So, you know, just along that lines then Tom, I mean, it’s, it seems like, whenever I hear data quality and talk to people about data quality, it’s, it seems like it’s like, a mountain to, to high decline. like how do people, how do people grow up that, how did he get started How you develop a strategy on implementing, on delivering data quality when it comes to, did the enterprise, like, where do people start
Tom Redman: well, so, so look, there’s two different questions. There’s where, where do they start and where should they start And unfortunately, most start in the wrong spot.
Dave Mariani: Right
Tom Redman: Right. I, I, he, he, here is the way I advise people and companies to think about this. Most of us don’t ever think about it at all, but we serve two roles every day. One is we’re a data creator. Right. And by that, I mean, we are creating stuff that is used by other people, right Similarly, we are a data customer, meaning we are using stuff created by others. All right. And, you know, you may not think about yourself in, in those roles, but what now, you know, it’s just sort of, you know, you’re in a sales capacity, you take in data, leads, data that marketing has, has provided you. And, and you look at that data. And, and there’s a lot of flaws in it. And a lot of salespeople report, you know, they spend two or three hours of their day cleaning up the data.
Tom Redman: So they’re not really being good customers. So what would, you know, if it were a car they were buying, right. They would not tolerate the bad defects in the car. Right. They would reach back to the manufacturer of the car and they would say, Hey, we gotta, you know, you can’t, you can’t show me this stuff. Right. And, and so the rule of the data customer first and foremost is to recognize that you’re, that you have some responsibilities. If the stuff you’re getting doesn’t meet your needs, then you got to speak up and you got to find the place where this stuff came from the creator, if you can, and articulate what you need. And, and then if you’re a creator, what’s G you know, usually what I find is creators say, I never knew you needed that right. Where the problems are. And, and when they do that, and, you know, you, you find that, that, you know, there’s lots and lots of very simple root causes that the creators, once the customer has come to them, they can eliminate.
Tom Redman: Okay. And, and so this, and this is these two moments, these moments of when data’s created, and a moment when data is used in when the creator and the customer are doing, these are the moments that matter. And in data quality, they’re happening, you know, millions of times a day. And so, you know, thousands, and even the smallest company, and, and this, this chic grit is to get people to become reasonably good customers. First of all, you have that rule and to become reasonably good at it, and to get creators, to acknowledge they have that role and, and become real, reasonably good at that. And, and that when you make that happen, it’s, it’s like a miracle now, you know, it doesn’t solve all data quality problems, but by, you know, sheer numbers, it solves so many of them. and, and so the question too, you know, when you start, if you want to do this at scale, You know, somebody may have a company with that name. And Then, then the secret is, is how do you turn on large numbers of people to fulfill these roles Okay. And frankly, it just isn’t that difficult.
Dave Mariani: Well, you know, it’s like, I mean, it’s when people think data quality, I think they immediately, at least me, I think of tools and technologies and data science. Right. But when you’re talking about it’s P it’s all about people. Right. So, so, so talk to me more about that. It’s like, what’s the organization, what’s the, you talked about creators and consumers. Like what other sort of, I guess people architectures or organizational kind of, considerations do, do, do enterprises need to think about, when they thinking about data quality
Tom Redman: Well, I mean, look, first of all, I mean, I, th th the organizational thing is, is first, you have to put people, regular people, right. Front and center in this role. And this was what, you know, the HBR article was
Tom Redman: About. Right. You know, almost all of the interesting things that are going on with, with, data right. Involve regular people. And, and so look, the, the, just with respect to data quality, we’re where we’ve had the most success is, is we establish a small core team, right. You know, meeting the data quality efforts, securing, right. The resources providing the training and, and then between them and regular people is a network of, of people we’re calling embedded data managers and an embedded data manager sits in, in the line. Right. So, you know, any team has, an embedded data manager, if it’s a part-time role. Right. But they have the responsibility for, for data quality within that, within that team. and so the core team will, you know, figure out how we’re going to do it and provide the training and, and, and all that kind of thing. And then, and then work through the embedded data manager to get everybody buying in, into these roles. And, and, and so for most people, by the way, this embedded data manager, it’s really cool job.
Dave Mariani: Yeah. Because
Tom Redman: You’re moving means something to your organization. You’re not a data person. I mean, you’re, you know, you’re, you’re, you know, you’re, you’re, you’re either a petroleum engineer or a know your customer person or a, you know, to working in a branch and a bank, whatever it is, but you’ve got these additional responsibilities and, and they put you on the leading edge of data in your company, which, which is really, really exciting for, for a lot of people.
Dave Mariani: So, it sounds like a hub and spoke model, Tom, is that, is that, is that the kind of the way where the core team is the hub and, and your embedded data managers are sort of the spokes to that strategy Is that the right way to thinking about it
Tom Redman: Well, so, so I, I, I guess I sent you a slide, pull that up a little bit.
Dave Mariani: Okay.
Tom Redman: Okay. So, so, so look, I mean, I I’m really emphasizing that, you know, we put regular people in the middle, you know, it’s a regular, people are always an afterthought, right. You know, it’s the digital transformation project failed because of the wetware.
Tom Redman: I mean, and I I’m really reacting very strongly against that and urging companies to put regular people right. In the middle. Okay. And, and then you see that, you know, embedded data managers, right. They’re in the same team or the same department as, as regular people. And here, we’re talking about data quality. So you see the core data, you know, the core data team. I mean, that’s a data quality team. And in this, in this particular case, right. I mean, it’s, it can’t, you know, discount the importance of, of leadership, and, and leadership playing some really key roles in getting these embedded data managers named and, and, and, and taking on the mantra that taught you how important this is to the company and, and, you know, dealing and, you know, taking the savings that they get from improved data quality. And then this is really key. I mean, as you mentioned, technology, look, I w you know, once you get the organization, right, then you can really take advantage of the tools.
Dave Mariani: Right.
Tom Redman: The mistake too many organizations make is they start with the tools and they don’t have the organization structure in place and, and the tools just founder as a result. so, so, so I urge people to, you know, start with this model and, and quite generally for everything in the data space, you know, this other component of, of fat organizational pipes. So I’ll just say a word about this. I mean, you know, we, we have to get over silos. And, and so in tech, we, we think about pipes. We think about data pipes and, and, and the analogy here is these are people, pipes, you know, get the right people talk and make it easier for them to do so, take on issues of disparate language and, and, and so forth. So, so, so, so look, I mean, this, this slide there’s a lot going on in this slide. The overall structure is really, really bearing fruit and, and, you know, so we’ve got lots of details to work out, though.
Dave Mariani: I love the, I love you to regular people like, you know, we talk about, you know, at AtScale we talk about the semantic layer, we’re trying to make data available for everyone, not just data engineers, not just people who are skilled in working with data, because we want everybody to be making data driven decisions. And that means that it’s, you know, that even having a, a title of, of data analyst is, is almost wrong, you know, it should, everybody should be a data analyst and it should be part of everybody’s job. So I love that. I love your regular people concepts. So talk to me a little bit about the yellow box and the leadership. So who does, who does, lead the charge here, like, is that, talk to me about that, Tom, about that, which what’s the role or roles, and what do you mean by leadership to sort of orchestrate this whole process
Tom Redman: Let’s, let’s start with, let’s start with data quality. The observation I’ve made is the data quality programs go as far and as fast as the business person perceive to be leading the effort demands. Okay. And, and, and, and there’s, there’s a lot going on here. So any team, any, any, you know, team of any size, if it’s leader, right, that’s a manager with three people reporting to, to them, right. Can, can take up data quality. An individual can take up data quality. If an individual of a leader of three of three person team takes it up, the quality program will extend to those three and maybe to the three that, that those people demand a at a higher level. If a department takes it up in some, some finance departments and say, you know, look, we’re sick of correcting all these mistakes. We’re sick of spending three quarters of our time doing data quality, not finance kind of thing.
Tom Redman: Right. Then, you know, you can take things up, up across all things, all things finance, and, and, and at the end at the industry, and if you want to do the entire, enterprise, then the organization’s most senior leaders must be demanding. And, and in terms of what they have to do to this idea that you’re a data CREC creator and a data customer, it isn’t once obvious. And it is so transformational. And, and so all the things around, you know, transformation and leadership, right. Sort of, sort of being in front of that are, are the kinds of things that, that, we really need leaders to do. You mentioned, you know, data-driven culture, the culture, again, your, your culture is, is top down, right at the end of the day. I mean, there’s can be a lot of bottom up forces and not everybody has to agree, but across an enterprise, it’s a top-down thing.
Dave Mariani: And I hear that a lot about also about data literacy, Tom. So, you know, how does data delivers literacy sort of interplay, with this sort of model that you’ve put forth
Tom Redman: So, so that’s a really good question. the, the thing I observed is that practically everyone today can do something right, and practically everyone can do something. It, we, we, it doesn’t take a lot of training for people to become better data customers, right. It takes a little more to become a good data creator. in terms of, you know, I, I like this, we, we push this notion that everybody should be a small data scientist. Everybody should have the ability to, to figure out how to improve their team’s work. Right. So to form a problem, to gather data relevant to, to that, to that problem and, and make improvements and, and put in place controls and, and, and, and so, so literacy, I mean, it, it, it just starts, I mean, very, very fundamentally, right. Me it’s, it’s like, you know, do I understand these roles, right Can I do the work associated, associated with these roles And, and the best part of it has been how almost everybody can do something now and wants to do something. And, so, so, you know, that’s that th that, that that’s the connection to literacy. I think too many things in literacy start to technical
Tom Redman: Training people, how to use a tool, right. When they ought to start with, okay, here’s this role, here’s what this role means, right Take tools out of it. Here’s what it means to, form a problem. Remember the scientific methods, right. You know, you know, the science to be a data scientist, even a small one, you must understand the scientific method. and, and so I’m really excited about, I think this is, the data literacy problem, and the data literacy opportunity facing us. And in 2022 is the same as the literacy problem facing These in 1900. and, and it will require that sort of scale of effort to get people to be literate.
Dave Mariani: Yeah. You think about the state of literacy data quality. I mean, it’s all building trust and data. Right. I see. So many of our customers are so many customers out there that are struggling with people just not trusting the results because of inconsistencies, or just lack of knowledge or, or confidence in the quality of the data, which
Tom Redman: I want to build on that. I don’t think, I think most data is not trustable. People show us,
Dave Mariani: So Tom should use it. Should people take the, should they take the attitude of, I don’t first, I don’t try, you know, prove it to me or I don’t trust it first is that should be your out of the gate sort of response to data. Yes.
Tom Redman: And, and from my experience, all data is guilty until proven innocent when it, when it comes to quality. And I’d say, look, we’ve we’ve. sometimes I do these things where I train in and, you know, in public settings and I ask people to make their first data quality measurement, you know, and there’s a simple process. They follow it only takes a couple hours and, and, but then returns a score on a zero to a 100 scale and, and, and you know, well, how good does your data need to be for you to trust it Right. You know, and everybody says, oh, high nineties. Right So I work in healthcare. If our data’s not good people die, or I
Dave Mariani: Work,
Tom Redman: It was all these things about how the data needs to be high nineties. And well, while only a couple of percent come in in the high nineties, you know, the average over, over this is, you know, over the years we’ve been doing it since the fifties.
Dave Mariani: Wow.
Tom Redman: It’s kind of scary. And, and so, so the, you know, obviously there’s some data that does meet the standards, but most data should not be trusted. just the same way your teenager should not be trusted
Dave Mariani: Back to our original story.
Tom Redman: Yeah. I mean, obviously, you know, that obviously you want to trust your teenager, but what you really have to develop the evidence that the data is good and, and the, and the, the, the most important thing is most of it, the most root causes can be made to go away and the data can be made trustable pretty darn quick.
Dave Mariani: So, so Tom you’ve, so, measurements, so how do you know, how do you know So, so you you’ve talked a little bit about the course and sort of coming up on a score of zero to 100, but, so how do enterprises know where they stand when it comes to data quality What, how did they measure it Like, what is, what are some of the tips and tricks
Tom Redman: The first thing that, I recommend they do is, is, is make something that, we call the Friday afternoon measurement here. And in parts of the world, we call it the Thursday morning measurement because Friday is part of the weekend. But, but what I asked them to do is just, take something that you do. And, and, and so some people say, okay, well, I’m, I drill for oil kind of thing. Okay. Well, what’s the important data there and said, well, we’ll call it a well header kind of thing. You know, it’s a, what are the 10 or 15 most important attributes data attributes of, of drilling a well or data, or know your customer, or I sequence genes, right. You know, what are the 10 or 15 most important things you need to do to do that work and, and just look at the assemble a spreadsheet a hundred by 15, or so, the last hundred Wells you drill the last hundred, know your customer, exercises you did and, and so forth and assemble this a hundred by 15, and then get two or three people who understand the data and in the room.
Tom Redman: Okay. And give everybody a red pen and work line by line and just make an X where the data is obviously wrong. It’s a work line by line, and then add a column and goes, you know, red, or w was there a red or no red, and then count the no rents. And when I mentioned, you know, those numbers in the fifties, that’s what people are getting. They’re getting these, these numbers and in the fifties. And, and, and the reason I asked companies to do do it this way, this, you literally can do it an hour and a half, right. You literally can do it an hour and a half to people, and the people who are using the data and who kind of thought IAM, well, maybe the data is not so good, but I think it’s okay. And they find out what we’re at 63, right.
Tom Redman: It is an eye opening experience for them. and, and so, and, and look, I mean, I do believe in automating things, which you shouldn’t be doing this every Friday, right. Or every, every Thursday, but, but what you should be doing is you should be looking at that and going, okay, well, that’s fine. Now, where are my most important problems You know, there’s usually two or three of those. Let’s go knock those off. Let’s measure again, once we get it into the, you know, the nineties, let’s put some automated controls in place, right. Is a, maybe we can, we can, you know, figure out a way to count errors when there’s only a few of them going forward and, and move on to the next problem. and, and so companies that will do this, you know, something, you know, so maybe they’ll start, I, I just worked with a company that, that, you know, in their first few months they made 40 such measurements. and now they’ve set up the improvement to, to go get after the most important sources of error.
Dave Mariani: So, so that’s awesome. We talked about, you know, that’s people, we talked about process, let’s talk a little bit about technology, so, so how does technology fit in here, to really sort of, sort of, complete the picture here, and complete, you know, complete, your picture here. We got information technology. So, so talk to, talk to me a little bit about that, Tom, w what’s that involved
Tom Redman: So, so look, there’s a, we have some real problems, Dave. the, the, one of the problems that, that took sort of the hierarchy of data quality problems are, can I find it
Dave Mariani: Right
Tom Redman: Do I understand what it means Is it relevant Is it accurate kind of thing Well, you know, this business of can I find it, you know, organizations have, have become loaded with technical debt
Tom Redman: And, and it makes it difficult to find things. And, and if you’re a data scientist, you know, chances are, you’re, you’re trying to pull data from, you know, from, from a bunch of sources and, and, and what you find out is that, you know, customer over in marketing, that really means prospect and over and in sales, what that really means is the person was sign off authority over in finance. It, you know, it’s, it’s, it’s the person ultimately responsible for the bill. And, and somehow you have to make sense of, of this mass. Okay. And, and, you know, and it’s layered on because everybody has their own systems and, and, and so forth. And so look, the biggest technological problem that we have to solve. And it’s going to take a while is, is that we have to simplify, we have to simplify the architectures, such that people can, can, can get what they’re entitled to, or, you know, stuff they’re not entitled to, it’s a bunch of GDPR requirements. No, you can’t do that for right. You know, this reason you can’t have it. but, but we, we need to, to, to sort of get that sorted out. that’s number one. And number two, we talked about before, but I am a, I am a big believer that that technology is once you have a working process, increase scale and decrease unit costs,
Tom Redman: I think too many of our companies have done their, it departments, a disservice by asking them to automate things that fundamentally are broken. Right. And, and I think it departments have to push back and go, I’d love to automate that for you, but what the heck is it Right. You know, it’s got to make sense before we can, we can do that. And, and so, you know, to me, these are the, the, the two areas where right now for most companies, the focus should be. and I think, you know, by the way, there’s a, there’s a whole bunch of great technologies that are sort of sitting there in the stable ready to be used, but, you know, there’s just too much, too much technical debt, too much overhead to get over. And until we clean out the clutter on, on those two particular issues.
Dave Mariani: Yeah, that’s a, that’s a good that’s, that’s obviously what, with our, you know, with that scale and the semantic layer, one of the reasons why I started the company was that, there is a lot of technical debt out there, and it’s not going away anytime soon. and that just makes it a really complex environment, like you said. And so, you know, the semantic layer seeks to really simplify and also, get some consistency when you, along those terms and those metrics, like you said, I love it. I love your, I love your customer example. it could be a prospect. It could be the person who pays the bill. You’re absolutely right. It means different things to different people. It’s really important for people to be speaking the same language. so Tom was like,
Tom Redman: Well, I mean, there’s a trap there, right I mean, because a prospect is really a different concept. Correct. Right. And, and you have to preserve those. You have to preserve those things. You may even have to build on those, on those things yet. There’s times when we need to work together, here’s the common language,
Dave Mariani: Right. Totally. Here. You need, you need all that nuance for, for all, you know, for people to be able to get their jobs done. I mean, you can’t, you can’t, aggregate it away, papered away. No doubt. so
Tom Redman: Too many architects have missed that point, right It, yes. Common language is important, but people still have to do their jobs.
Dave Mariani: Yeah. I mean, details matter. and, and, and so, and taxonomies matter. So, I love that. So, Tom it’s like, we covered a lot of ground here. We’ve got people process technology. What else, what have we not covered that you think that the audience needs to be aware of or something they need to take away
Tom Redman: well, I look, I think I th I think we’ve hit the highlights. One thing I, I really, I urge people to do, and that is, is, you know, that the technology, the technological capabilities, or way ahead of the data and people, capabilities and, and if all you care about is the technology, right. Well then, you know, you really ought to be spending more of your time on, on data and people, because it’s, it’s, what’s holding, it’s, what’s holding you back. And, and of course, by the way, most people aren’t just technology people. So most people are business people, right. And, and for them, the messages is slightly different. You know, you probably, if you’re, you know, if you’re not in your twenties or thirties, the technology’s probably out there to see you to the end of your career, but right. What’s holding you back. Are these, are these other dimensions
Dave Mariani: Yeah. And I love that. So, just to close it out here, first of all, just remember audience that, Tom, just, just, take up his, his, Harvard business review article called a piece called the, your data initiatives. Can’t be just for data scientists and w so it covers a lot of what we talked about here, but in much more detail. So, what’s it, it’s a great piece. So Tom, before I let you go, I always love to, I love, always love to ask, my fellow podcasters to predict the future, or what, what should we be keeping our eyes on with regards to data and analytics and data quality trends, Tom, for, for, for the future
Tom Redman: well, so I want to answer a related question and it’s a little bit of a historical perspective. So if you were to ask most people, what’s the most, you know, important information technology invention, right. In all of history, a lot of people would say the Gutenberg press, okay. You know, moveable type, how, how, however they would say that kind of thing. And, and what’s really interesting about that is at the time there were a, a large group of professionals, it’s a bright, basically monks and friars who were, who were there before the press and what they were doing was copying the Bible and, and other ancient works from Greece and, and so forth. Right. Right. And so, so if all the press did was replaced those people, right. Nothing would have changed. There would have been any more Bibles.
Dave Mariani: Okay. Right, right.
Tom Redman: Kind of thing. But what had to devote what had to change was a bunch of management things around the press, right. So a publishing industry had to develop, right. People had to write books about things other than, you know, the Bible that, that people would be interested in, people had to learn to read. Right. So that, you know, so, so, so that there was greater and greater demand for, for, for the, for the books. And, and, and, and to me, there is just something that is so apt in this analogy,
Dave Mariani: Right
Tom Redman: Our next generation of progress, our big boost, right. Akin to what we got from the Gutenberg press is about management. And it’s about people. And, and, and, you know, and to me, I just think there is something so human and so exciting and, and beautiful about that. Now I’m not going to predict when that’s going to happen, but I will predict I will confidently predict that it is necessary for the explosive growth in the benefits that we all really want to get, out, out of it. And frankly our, our data and people as well.
Dave Mariani: Yeah. I love that. So it’s not just about automation, right It’s about people and management. That’s a really good way to think about it, and to think about data and analytics and where we’re going to go. it’s been very technology focused as you, as you point out. And so I guess if anybody, w w what’s my take away here is to think more about, about process and people and not just technology. It’s really a three-legged stool, isn’t it Tom, this has been fantastic. I want to thank you for joining the podcast and, and, giving me an education. And I hope it’s, it’s been useful for everybody listening. So for everybody listening and for Tom, thank you for joining, joining us today and be data-driven. Thank you.
Tom Redman: Thank you, David.
