Marketers consider themselves as creatives rather than statisticians. But if marketers want to be successful in our world today, they must know how to effectively use data.
If you’re working in the marketing industry, there is always constant talk about data. Data provides valuable information for business decisions and has become one of a company’s greatest asset. Specially in our world, data is in constant abundance. However, despite its value and importance, most companies still fail to make a data-driven marketing strategy. Why is that?
If I were to ask you right now how much ROI you’re getting from all you’re digital marketing activities, chances are you don’t have a clue. This is one example of data that is either not tracked or simply ignored. You need to track the return you’re getting from all your digital marketing activities. And we have the perfect assessment for that. Click here!
Data-driven marketing is based on facts rather than biases. It’s more accurate and leads to better outcomes. Yes, data can sometimes be risky, specially if it’s bad data. It’s also complicated to work with without technical help. But you don’t have to be driven solely on data.
Companies have a wrong notion that they have to make every single decision with data. It all starts with data, don’t move without data, your opinions need to be backed by data, etc. Yes, data is important, but it’s also important to understand that data supports you. It’s not at the heart of everything. Don’t rearrange your company around that data.
Like both sides of a spectrum, companies need to properly balance their insights, gut feelings, or biases and the data they have. If you’re serious about generating more revenue for your company, you have to know how to properly and effectively leverage data.
In this episode, Ruben Ugarte shares his expertise on data and marketing and helps you get the most out of it.
Ruben Ugarte is an expert in data, decision-making and the author of the Data Mirage: Why Companies Fail to Actually Use Their Data. He has helped over 75 medium-sized and large enterprises, including the Fortune 1000, across 5 continents, use data to make higher quality decisions.
These decisions helped companies significantly boost performance, increase profitability, dramatically lower costs, and make their teams world-class. He also maintains a popular blog that over 100,000 readers have read. In his free time, you can find him dancing or trying to learn something new.
01:00 – Introduction
01:49 – What do B2B Digital Marketers need to be thinking right now?
03:24 – Ruben Ugarte’s background in data and B2B Digital Marketing
04:54 – Impacts of the pandemic on data
06:28 – When to use data on decisions
08:26 – Accounts-based marketing
09:44 – Fears in using data
11:26 – What being data-driven really means
13:37 – What Ruben recommends to entrepreneurs when using data
14:25 – Data-driven vs. numbers-driven
17:22 – Where data is being used now
19:43 – How much AI has affected data?
24:01 – Baseball coach analogy
28:53 – Dealing with bias
30:52 – Doing experimentation
38:25 – Connect with Ruben Ugarte
“You need to be able to make decisions without data. Because one day you won’t have them. If you’re stuck, if you’re paralyzed without it, then you’re really going to have issues with uncertainty.”
“Uncertainty is the name of the game now in our world.”
“Data in the right amounts can give you insights into the answers you’re looking for.”
“At the end of the day, it comes down to what data did you learn? Something that you can take and change a campaign or change a messaging. And if you’re doing that, then you’re being data supporting.”
Episode Links and Resources
Ruben’s website: https://rubenugarte.com/
Ruben’s LinkedIn: https://www.linkedin.com/in/rubenugarte/
Digital activity ROI assessment: https://b2bdm.com/digital-activity-roi/
More episodes on data: https://b2bdm.com/how-to-leverage-data-in-digital-marketing/
Episode TranscriptClick to access unedited transcript
Jim Rembach (00:00):
Okay. B2B DM gang. Uh, I’m excited to have the guests that we have on today. Ruben Ugarte because we’re going to talk about something that I’ve been exposed to for about 20 years, and that is working with customer experience data. But that’s not the part that we’re going to talk about. We’re going to talk about taking data, driving insights and making better decisions. And for even organizations that think they’re doing a good job of using data in order to do that, they may not, not be in. Ruben’s going to share that with us. Also, if you’re trying to sell to other companies, you need to understand a little bit more about what they are doing or not with their data. Ruben. Thanks for joining us today.
Ruben Ugarte (00:38):
I’ve got to be here. Thanks Jim.
Jim Rembach (00:40):
Before we get into learning a little bit more about you and your background, if you could share with us, what do B2B digital marketers need to be thinking right now?
Ruben Ugarte (00:50):
Wow. You know, uh, the pandemic has, of course, they’ve talked about ad nauseum for a year now, but nonetheless, as we get into this recovery, there’s a few things that marketers should be thinking about. One is, uh, personalization, right? The, this idea that be able to communicate with, with customers directly, uh, to do in the way that that’s, that makes sense to them. I think it will become even bigger and even a bigger issue. And we can dive into examples of companies that do this well versus not. Uh, I do think data in general will become more important. We’re seeing, for example, retailers, as they mix a switch to a more, uh, to direct-to-consumer model and now what big role data might play there. So if you’re working with those other companies, you’re in that space, uh, now is really the time to really get a handle on your data.
Ruben Ugarte (01:40):
And then lastly, I think it’s also about understanding the limitations of, of data. Um, you know, we saw a lot of companies in the pandemic that were unable to make decisions. There was so much uncertainty, so much lack of data, so much lack of understanding, and they got stuck and frozen and that became a huge deal, right? Because they, they had to move, they had to make choices. Um, and I think that same thing will apply now as we go into recovery, there’s still the uncertainty. So those are things I think marketers need to be thinking about, uh, as to how they approach the work, their companies, or even their clients.
Jim Rembach (02:14):
Thanks for sharing that. And we’re going to get into many of those key points here in a second, but before we do that, tell us a little bit about you, your background and how you’re going to add value to this conversation we’re going to have today.
Ruben Ugarte (02:24):
Yeah. So my background is, as I mentioned near, and, uh, that’s really helps with data. You know, I find beta as any marketer can attest. It’s very technical. It’s a lot of details to it. There’s bugs, there’s errors, and you have to typically deal with some kind of engineer persons of technical person to get it going. So my background has allowed me to work primarily with marketers. I’ll say, you know, it’s mostly marketing teams that bring me on other companies and then help bridge some of the gaps that may be having with our engineering team. Some of the pushback they may be gathering and I help companies really get the most of that data. You know, I work with very small companies, five, 10% companies, even smaller startups all the way down to public companies and the problems are surprisingly quite similar. They differ a little bit of course, based on stage, but there’s lot of similarities and it’s really all about how to get more insights out of whatever data you have. Doesn’t matter if you have a lot or little and highlight change behavior and how to make better decisions.
Jim Rembach (03:26):
Well, you had also talked about, uh, the shift of, you know, what had occurred in a lot of organizations found themselves with the, COVID not having either certain pieces of data or not clean, um, you know, data in order to be able to make a decision and move forward. But I would dare to say, is that vulnerability, did that vulnerability take place? Because for the most part, when you look at year over year, things were somewhat consistent. So therefore I didn’t have a worry.
Ruben Ugarte (03:55):
Yeah, I would say so. I, you know, I think a lot of companies found themselves in no man’s land. They had never seen numbers like this. You know, I work with tourism agencies here in Canada and they all kept telling me we’d never seen drops in visitor numbers like this ever. Um, so they’re faced with effectively crazy numbers. And how do you deal with, how do you market, uh, where do you market, who do you market to, what do you say? Um, those were all questions and all the, you know, many companies and teams, uh, even companies I think were doing really well industries that did well. I think there were still faced with the same question of what do we actually do. And I think that’s some of the things that I cover in the book that, you know, it needs to be able to make decisions without data. Uh, because one day you won’t have them. So if you’re, if you, if you, if you’re stuck, if you’re paralyze without it, then you’re really gonna have issues with uncertainty. And uncertainty, I think is the name of the game now and in our world.
Jim Rembach (04:52):
So when you say, um, you know, decisions using data for me, I think for a lot of organizations, they end up using it more as a backup, um, than they do as a driver in their decision making process. And what I mean by that is if I was to say, well, when am I going to look at the information in order to be able to decide it comes later in that process versus the beginning. Now I, for me, you’re more skilled at this. Where should the data come when I’m actually formulating my decisions?
Ruben Ugarte (05:28):
Yeah. So I think that she actually come at the very beginning when you’re trying to decide what to do or where to go. And let me give you a marketing example. Uh, I’ve worked with a company that they were really good at driving leads, but the quality of those leads wasn’t as high. So the sales team was struggling to convert them. So a very classic problem, but the marketing team didn’t actually quite know how it changed us for years. Their focus has been, let’s drive more and more leads and that’s not just, they had to recall the they fine and the sales team, the best, their problem, you know, they’re not able to cost them. So to kind of figure out where to go from there as a sort of major shift in strategy for the marketing team, they were able to go into the beta and then want to make it clear that this is not, you know, this massive data science machine learning project, where they spent six months in a, you know, kind of like a chamber going through all the numbers.
Ruben Ugarte (06:21):
They were, they just have to say they just started to segment their audience and say, okay, here’s, you know, we have a thousand leads. One of the different segments we have here, what segments tend to convert, uh, higher than other segments. And in this case that we’re dealing with, uh, students. So they had different types of students, um, whether I know community college students or Ivy league students, and then from there start to figure out what are the different demographics? What do they care about? Who should we focus? What, what should the market and say towards those people, where do we find them? So that’s, to me the role that data plays, it’s not very complex, but it starts to give you hints as to here’s what we should go. Here’s what we need to learn next. And here’s where we can start shifting our marketing campaigns, our emails and so on.
Jim Rembach (07:07):
So for me, as you’re talking about that, I start thinking about if I’m a B2B organization, there’s been a lot of talk around this account-based marketing type of strategy. And, uh, so for me, um, that’s whole segmentation, uh, account based marketing, all, all of that is data driven. Isn’t it?
Ruben Ugarte (07:27):
Exactly. Yeah. That’s the same idea. Whether we’re talking about students, which are consumers, of course the same idea applies even more to companies, right? Because who do you go out? Do you go after every single company, what’s the characteristics what’s going to make them likely to buy. What’s gonna make them likely to use the product and be retained. I think those are all questions about data and data in the right amounts can start to give you insights into what those answers actually look like.
Jim Rembach (07:53):
Okay. So this starts getting into a whole strategy issue. And when you, when you look at a lot of the research on organizations of all kinds of different sizes, one of the areas where they struggle the most is on the entire strategy piece. So in marketing, we’re talking about creating strategies that do not get into that level of, uh, just say precision, uh, to be able to really, you know, target, uh, because ideally if we were to say, I have 10 different customer types or customer segment groups, you know, it’s a lot, I’ll work in effort to make sure that I have the relevancy and the personalization to be able to make sure that I’m building trust within them. And they’re saying, Oh, I’m a good provider. They’re talking to me. And then therefore I take that next step. I mean, it’s a lot of work, so could I potentially not be using the data because I know it’s going to mean more work on the back end.
Ruben Ugarte (08:43):
I think so. Yeah. You know, it’s that stuff that you hear about people who, you know, don’t want to go to the doctor because they’re afraid what the doctor, my fine. Right. It sounds crazy when you think like you’re sick, you should definitely go to a doctor, but Hey, you might be even sicker. You, you might not even know. Right. So I think the same idea can happen with companies when they think, you know, like we’re doing well. But if we start diving into the numbers, we may really discover that we’re not doing that well, that there’s serious holes, but there’s a space that when you, when you talk about it, right? So, so you have to do it. And I actually think, uh, companies are spending a lot of time, small, large doesn’t matter, simply doing things, right? So if they’re already busy building campaigns and running AB tests and writing copy and all those things, but they’re not quite sure if this is where they should be spending their efforts, right.
Ruben Ugarte (09:28):
They’re missing the strategy because they may be targeting the completely wrong customers. And I’m sure we come across examples, your audience in your software, this was the wrong segment altogether. They’re never going to buy it’s, it’s almost, you know, but it’s a significant portion of your time. So I think that that’s the mindset that companies should be approach, uh, poaching data, not as in who knows a well fine, but we can really start to shift our time and spend as much time as possible working on the right things and reduce the, you know, the amount of time that is wasted. Yeah.
Jim Rembach (10:02):
It’s counterintuitive. I mean, we essentially, you know, shut that door, you know, but sooner or later that thing’s going to bust through and it’s going to be quite nasty. Right. Okay. So the first chapter in a book, you kind of set, you know, a very compelling question that I think I would like to dive into here. And the chapter title is the reality of being data-driven and why your company isn’t. You got to tell us what you’re talking about.
Ruben Ugarte (10:27):
Well, you know, data-driven has been the thriving for us, a mantra for maybe 20 years now. And I think it’s a little misguided. I think it’s forced companies to think that, you know, you’re going to make every single decision with data. It’s all going to start with data. If you don’t have data, you don’t move. If you have an opinion, that’s not backed by data. Didn’t even bring that into, you know, into the head to the boardroom. And I think it’s, it’s the wrong approach, you know, companies and I talk about later on in the book should really think of, of stablish in a data supportive culture, where data supports you and your decisions along the way it gives you insights gives you guidance, helps you go back in the right track, but it’s not the annual appeal, right? And it’s not, you don’t have to completely rearrange a company around data and put data at the heart of it. Some companies do, we can talk about that, but really it’s just about, are you learning things that change your behavior, right? Are you doing new things? That’s really what it comes down to at the end of the day, what data do you learn? Something that you can then take and change a campaign or change the messaging. And if you’re doing that, then you’re, you’re, you’re being data supporting that and you’re doing well, but you don’t have to be driven by data. You don’t have to let data sort of take the lead and everything.
Jim Rembach (11:46):
I think you bring it. That’s a very, very important distinction. And I even find it with, you know, um, individual entrepreneurs may be just running their own business. You can call them solopreneurs, if you want is they may be doing a lot of activities that don’t have visibility into the impact and performance on it. And you know, it’s not supporting them. Um, and oftentimes they just don’t admit it and they just go off of habit in regards to, Hey, well, this is what everybody else is doing. I need to do this as well. And they kind of fall into, you know, what you can call as far as the sea of sameness. Right. You know, I’m, I’m doing everything just like everybody else. So I would dare to say, if somebody needed to really change the way that they approached the things that they’re doing from a technical perspective and make sure that they do have the data in there, what would you recommend to them?
Ruben Ugarte (12:37):
Yeah, so, you know, I w I would start, I think this applies to, uh, an individual person, a small team, or perhaps even the larger team within the larger company is to start with a handful of KPIs that matter too, right now, because of the way data is collected, you have this ability to track an unlimited number of KPIs. Uh, you know, anyone that open go with the Lennox ever knows what, what I’m talking about. So there’s an unlimited number of KPIs, but only a handful of really makes sense. So for an individual person, a coach and entrepreneur, it might be simple things like sales, uh, maybe some of the number of calls you’re making to clients or prospects, your conversion rate, things like that. And that’s where you start your tracking. It can be simple, right? It can be just a spreadsheet, it could be a dashboard.
Ruben Ugarte (13:24):
And then you build from there. And if you’re, if you’re missing data to calculate some of those KPIs, that’s where you then start to determine, okay, here’s what I’m missing. Maybe I need a tool here, maybe an upgrade to whatever technology I have, but you go from there. Uh, instead of started on the other side, which is what could I collect? What could I track, which is endless, and it’s just going to drown you in data. Um, and I think that’s what companies, a lot of companies do. They, they start with it where they can track, collect, and then they’ll say, and then we’ll figure out after we’ll figure out how to use it after. But the, uh, the usage is quite tricky.
Jim Rembach (14:02):
Oftentimes when you start talking about data, um, I hear different things that for me, I started wondering, are they the same? Or is there, you know, different mindsets or philosophies behind it? And then what I mean by that is people will say, well, we’re data-driven. And then other people’s will say, well, that we’re numbers driven. I, to me, I don’t think they’re necessarily the same thing. What have you found?
Ruben Ugarte (14:25):
Oh, that’s a good point. You know, I seen that distinction and I would say, I hear a lot about numbers driven from, uh, the financial side of the company. Uh, whoever manages the finances that they’ll say, you know, here are a number of series of profit, a P and L uh, here’s how we manage the company. Uh, I seen that. And then of course, maybe marketing teams might talk more about the data driven. What’s the data saying that distinction to me, uh, would come from, I think if we look at financial numbers, they are straight up numbers. Here’s your revenue, your margin, uh, your costs. And you’re sort of driving the company and making decisions based on that. They can, that, that works. Of course, it’s, it’s one approach. And I think we’re companies or teams, or people say we’re data driven. They are talking much more than just numbers. Yeah. You have numbers like landing page conversions and clicks on ads and things like that. Those might be a quantitative, but then you also have qualitative, um, uh, NPS or what customers are saying on surveys or some other kind of qualitative measure. So you’re being driven or you’re being guided or supported by this combination of quantitative and qualitative versus just a pure numbers.
Jim Rembach (15:41):
Okay. So as you’re, as you’re saying that I’m starting to visualize certain pictures in my head of, you know, certain numbers and data of that have traditionally been used in certain parts of the organization that don’t necessarily go to others. So I’ve spent a lot of my years working in contact centers and customer service. And, you know, I’ve been using that from a, both an inbound sales perspective, as well as customer retention and servicing perspective. But I think with the shifts that have occurred over the past couple of years, have you seen certain parts of the organization using data that they traditionally haven’t used and now they’re finding value from it?
Ruben Ugarte (16:22):
Yeah. Yeah. You know, not nice, much more enrichment, for example, uh, being able to bring external data in. So for a long time, it was all about your data, your customer, data, emails, names, phone numbers. Now you see much more enrichment of that data to break demographics, uh, gender, if you’ll have a lifestyle numbers. And again, you start to really flesh out a much more holistic, more human picture of actual who your customers are. It’s not just a data point, a bunch of people who paid your money, but here’s what they care about. Here’s how they use our product in their everyday lives. Um, I also think you you’re seeing much more qualitative. Quantitative has been sort of the, the, the major thing companies want, right. Financial, as we mentioned before, they look at the numbers, right. And if it’s not profitable, it’s not profitable, but it also misses a few things.
Ruben Ugarte (17:19):
Uh, some of the customer support some of the customer happiness or customer satisfaction, it’s not always fully captured by a number of metrics. So now you hear much, uh, companies talk more about what are customers thinking? What do they like, what do they don’t like, how can we improve? How can we, how can we use that in a market, in a salon? So those are two areas where I see some new data coming in. And I think it’s just easier to track now to run some of those surveys automatically, uh, to talk customers at scale without having this massive effort.
Jim Rembach (17:54):
Well, I think what you said right there as a key thing, I mean, one of the reasons that I’ve found that qualitative, uh, analysis was, you know, never really considered is because of the whole effort associated with doing qualitative data. I mean, it, it’s, it’s a whole, I mean, when you start talking about coding, when you start talking about cleansing, when you start talking about all of the things that you need to do in order to make the qualitative data, you know, an effective data set, um, it takes a lot of effort, but like you said, it is also true with, uh, speech analytics technologies with a lot of the ability to convert, you know, voice to text, and that’s getting better. Uh, some of those, um, you know, models in regards to costs versus value chain quite a bit. So how much has AI and artificial intelligence started playing in all of this?
Ruben Ugarte (18:43):
Yeah, I mean, AI is already here, right? It’s it’s, we have all likely experienced that in some shape or form. I would say it’s primarily used by larger organizations. Um, I do think an individual and a small company should be generally aware of what’s possible and what’s hype and what’s not, uh, it doesn’t mean it might be applicable either to yourself or to your clients, but at least, you know, what the landscape is. So we can say, you know, there’s a handful of use cases where AI is, would be useful, um, for example, anything to do with pattern recognition. So we see it when Facebook recognizes faces and photos or Google photos does the same thing, a fraud. I love payment processors, uh, companies like Stripe, PayPal, and some of the biggest ones. They now use AI to detect fraud before it happens. Uh, so they minimize some of the costs of those chargebacks.
Ruben Ugarte (19:37):
You might see. And in some cases we see AI in a sort of very high innovation fields. Uh, self-driving cars might be the best example right off, Hey, I have this beam that’s trying to be used here. However, it’s very experimental. You know, self-driving cars have been around for 70 years now, right. The research on it, and we still don’t have it. And it might be here in the next 10 years. It might not be, it’s unclear. It’s, it’s a really hard problem to tackle, but those are the use cases, uh, that, that started to become relevant and helpful for companies if they have the scale and they have the data to take advantage of them.
Jim Rembach (20:13):
And on the marketing side, I mean, we’ve covered this with quite a few guests on our show in regards to how artificial intelligence is being used in content creation. Uh, we’ve talked about how artificial intelligence is being used and strategy development, and those actually can be done at an individual level. Um, so we do have some of these SAS based solutions and tools that are out there, uh, that will allow, you know, um, organizations to create better content. That’s already going to be SEO optimized. And we talked about that on a previous episode, you know, we talked about how to create a strategy based off of, you know, who my targets are and things like that. So it’s, it’s definitely becoming, you know, a scenario where organizations of all size can start leveraging it more and more. So I’d be interested in to see from a data perspective, how we can see some of that also taking into effect when we start looking at, for example, modeling, right. Um, we start doing some sophisticated data modeling and we start throwing in some AI insights into it and pick a cause us to potentially take a direction that we’ve never even considered before. Is that true?
Ruben Ugarte (21:17):
Yeah, exactly. Yeah. So, you know, I think a lot of those cases fall into sort of like basic pattern recognition. And I’ll give an example that I think a lot of marketing teams have started to see on a regular basis, uh, which is, uh, attribution models, whether it’s last click or first click and so on. But the sign of that traditional model is based on data and Google analytics. For example, maybe the most popular attribution tool out there for a lot of companies, they have now released a, I believe they call it a algorithm attribution it’s effectively AI attribution. So they, they provide an AI model and it, they let you attribute your conversions based on, on machine learning or AI. And it’s meant to have, or to give you a much more balanced way of looking at conversions and the multiple touches that might be required to get to a conversion.
Ruben Ugarte (22:07):
So I would see examples like that, uh, and that helps companies manage their spending, right. Be able to decrease their spending by being more efficient, of course, any company that runs Facebook ads, Google ads, you’re being held by AI. Those, you know, the attribution models that Facebook uses to optimize your ads are now fundamentally run by AI. And there are, and you know, the reason now is you basically just tell Facebook, here’s my conversion. Here’s my ad. You do the heavy lifting it’s because those AI models have much better than what we might be able to do with, uh, with targeted. So those are examples. Like I said, I think in those, in the largest examples, the AI is as hidden. You don’t see it, you’re not writing the code yourself. You take it advantage of a software tool or something that was already built. Um, and it’s great because you don’t have to go hire, you know, a very expensive data scientist. Uh, you can take advantage of someone else who already did this
Jim Rembach (23:01):
Talking about that, um, data scientists. I mean, so there’s, there’s there, there’s a lot of people, um, who, well, let me, let me tell this story. Uh, I think this is going to be important, um, to kind of set the stage for where I’m going with this and trying to get your insights on it is I was, I’m actually a certified pitching coach. Some who have listened to the show on a regular basis may have heard me talk about that. And for me, when I don’t have to work anymore and I can retire, you know, I want to be a college pitching coach. That’s, that’s what I want to do. A lot of those programs.
Ruben Ugarte (23:33):
So I started as a baseball pitcher, coach baseball that’s right.
Jim Rembach (23:38):
Yeah. Because a lot of the, the smaller schools, you know, division three and two as well, they only have two paid coaching positions. They rely on, uh, on unpaid coaches in order to help them to run the program. So for me, I’m like, Hey, I have, I have learned all these things and, you know, I’d love to come help you and Oh, by the way, you don’t have to pay me. Win-win right. So, anyway, so anyway, I, I was, uh, attending this particular, um, clinic, um, with a bunch of different college coaches and performance coaches. It was just on pitching and that’s what I want to specialize on. And, uh, I was talking to this pitching coach who is a coach at the university of Missouri, Fred corral, uh, just a dynamic guy, um, very, very knowledgeable. And we were talking about analytics, baseball, analytics, you know, and using analytics in order to be able to make better decisions, right.
Jim Rembach (24:31):
It’s, it’s it’s data, right? It’s the same thing as a business, same thing in sports are doing it all the time now. And, and, and we were talking about interpreting of that data and he was saying how a lot of college students come out of these, these sports analytics programs. And they start doing all this data analysis, and then they say, look what I found. And he said, but the problem is they don’t have the practical understanding and application of what it takes in order to be able to play the game in order to be able to coach the game in order to be able to develop players. And he said something along the lines of, you know, you just can’t teach new dogs, old tricks.
Ruben Ugarte (25:12):
Jim Rembach (25:13):
And so I think that’s really important when we start thinking about interpreting data, right. We can do all this analysis and I mean, we can make it, you know, throw AI in and do all this other stuff. But unless we have that understanding of the business and the business acumen, I mean, aren’t, we greatly and potentially hampered when we start looking at interpretation.
Ruben Ugarte (25:36):
Yeah. And, you know, I think you brought up a couple of examples where that there has been perhaps too much focus, uh, brought to the alter of analytics, of course, Moneyball swept through the baseball world maybe a while, about 10 years ago. And I imagine every baseball team out there now has some kind of data science team that helps them choose recruits and strategy Island bout that. Um, I actually follow NBA, uh, alpha baseball. And then the, in the basketball, the Houston rockets are known as the B analytics team. You know, they, they were driven completely by the licks, their data determined that they should should more three pointers. So I started showing more three pointers that anyone else in the league, and they’re a good team, right. They make the playoffs, they’re consistently a team in contention, but it never, they actually haven’t won a championship, I don’t know, 20 years.
Ruben Ugarte (26:23):
So Dana can get you there and can help you choose those recruits and maybe tweak your strategy. But there’s definitely a human element. Uh, this, we might call it instinct or gut or whatever else, maybe just experience that is really the final. What really gets, gets you over the hump. And sometimes companies, I think any metaphors teams make it stuck with what’s the data tell you. And if the data doesn’t tell us that we’re not going to do it, but then you complete this regard, you know, people like the D the example you mentioned, right? People like Fred who have all this experience. And he may say, you know, I see this if the data supports me, but I see this from all my years of experience. So finding the balance between those two worlds, I think is what the best companies do. Well, they don’t get they’ll treat data like a hammer, you know, in search of nails.
Jim Rembach (27:14):
Well, then even, okay. So, okay. Okay. So I think there’s a, there’s a piece in there that I think is an important, even more important message there. And you really started talking about the whole discounting, right. Um, then the other piece and other side of that could be bias associated with it. So in other words, if it doesn’t support my hypothesis, I will discount that easily. Yeah. I mean, so how do you find that balance of not, you know, inducing or allowing bias, you know, to cause you to really take advantage of what’s sitting there right in front of you?
Ruben Ugarte (27:53):
Yeah. I mean, what we’ve seen companies do really well to deal with that biases on your own. There’s only so much you can do. There’s a recent, this sort of bias can, can get sticky. So what I see companies do is get them to defend different roles. So you might have a team, a marketing team that has the data in front of them, and they say, you know what? The data is telling us to lean towards more paid advertising. Um, does anyone, can anyone take a position to defend why we should spend more money on organic search or what should spend more money on referrals? Let’s, let’s kinda go through this debate, um, and see where we end up with. And I find that’s a great way to deal with bias. Companies just have to find a time which can be hard and be able to go through this kind of process.
Ruben Ugarte (28:42):
And sometimes it’s just about, Hey, just play devil’s advocate, take the other position, tell me why this won’t work. And other reasons we may fail and what this also does. And it can help you surface things that you might, uh, that may actually derail us. So you may discover that, Hey, here’s a three or four issues that if we don’t tackle them, if we don’t prevent them, that could really affect our strategy, let’s deal with them now. So that’s one way to start deal with bias. You know, you become aware of it, but then you need to, I think, take advantage of group dynamics to deal with that. And I, you know, I think, uh, going back to Moneyball, right? I, I, I remember the baseball teams, the way they recruit that they will get the recruiters around the table and they would debate the players and say, this player is good at this better. It’s not good. I don’t like his head. He walks by me or whatever it may be. So this kind of process can, can be effective with teams. I, they did a state debate numbers and they will treat them as, as a Holy, like, here are the numbers, no one can debate them.
Jim Rembach (29:45):
So you also talk a lot about, um, you know, experimenting, making better decisions. Talk a little bit about what you mean by that.
Ruben Ugarte (29:52):
Well, you know, experimentation culture, I think it’s at the heart of a lot of what really great companies do well. Uh, they have big debts. They have big home runs, but they’re also making a lot of our first base heads. Second base heads about walks in a lot of times when I’ve been with teams and we’ve been stuck in a debate, should we do this? Should we not do this? What I tell them is, Hey, you know what, in a few weeks we could find out, right? We could actually run the test, do it in a way that’s not intrusive to customers. You’ll have to show it to all your customers, uh, your entire website traffic, and you can start to get a little more insights. So what we need now is just to figure out what the hypothesis is, how, how did the side the status, but we didn’t necessarily have to answer this major question.
Ruben Ugarte (30:37):
We can test our way to it. And I imagine there’s been a lot of talk about AB testing in your podcast, right? It’s, it’s a, it’s a great way to test ideas. And, uh, we can talk about maybe how to do that, but I think it’s beneficial in general can take the pressure off teams to say, this is the right answer. You know, this is the one idea that we’re going to commit for the next three years in our strategy. And simply say, this is what we believe right now. We’ll test it along the way. And if something changes, we can change our mind, but we don’t have to make this, you know, never ending commitment, uh, for the next few years.
Jim Rembach (31:14):
But I also think what you’re jaded on there, um, you know, add on to that is to think about what a realistic plan would be like that. So for example, uh, you know, I, I go through and I make the decision, I’m going to do some split testing. And how long does the process take? You know, if it actually meant the transformation. So like, for example, you know, um, I did some testing, it’s on a smaller scale. I have a certain amount of sample. I have a certain amount of confidence, you know, and then, then where do I go from there?
Ruben Ugarte (31:47):
Yeah. Uh, so, so once you’ve got that small test, then I would scale that up. So I would either do a bigger test or I would try to roll that out and again, have some beta to track the result, not just at the short term, but the longterm. Right. So sometimes you see,
Jim Rembach (32:04):
No, that’s what I mean. It’s like, you know, so thinking about this is that, um, do I go through two rounds of bigger sampling? So, you know, I take it from here, which was a small sample. I’ve got, you know, I got, let let’s, let’s just get the straight numbers. Okay. So when I dealt with data sets, dealing and measuring the customer experience, um, you know, we deal with competence levels all the time. Cause people would want to shoot down results, you know, if it wasn’t of a certain level of confidence, right. Um, and especially the people who are executives who went through MBA schools, Oh, well, that’s not a, you know, a 3% confidence level and they just shoot it all and shoot the whole thing out. And I’m like, well, wait a minute. No, let’s, let’s understand about how variability and competence levels actually works.
Jim Rembach (32:46):
Right. So, I mean, if this number is off, it’s only off by this much. Right. So put it into perspective. But if our steel, you know, for dealing at smaller scale, what we used to tell clients is that, look, when you start getting above 30, 30 samples, you’ll start seeing consistency in your data. And if I’m starting to see consistency in my data, and maybe I don’t, you know, I can’t realistically do a sample of, you know, 500 cause I’m a smaller organization. Right. Um, w when can I potentially take that sample of 30 in a, doing a split test and maybe take it to a larger sample? Um, maybe take it to a larger sample before I do a full rollout. Cause that’s like, it’s almost like putting yourself in the pool. I do a toe, I do a foot, I do an ankle and you know what I go knee up. And then, you know, how long does it realistically take for somebody to go through that process?
Ruben Ugarte (33:38):
I would say it’s a bit tricky. You don’t have to guess it would depend on, on the traffic. Um, I would say generally speaking, uh, an experiment will probably be at least in the, in the number of weeks. So at least the number of couple of weeks to be able to see a full cycle that has a full week, that you might have some seasonality between, you know, weekdays versus weekends. Um, so to go through the process, you just mentioned where you might have a couple of tests that get bigger and bigger before our rollout. I think realistically, you’re probably looking at at least a few weeks, maybe a month or so to, to fully do that. And you know, it may not sound like a lot, but of course, some companies are thinking, why can we do this this weekend, right? Why can we run one test?
Ruben Ugarte (34:21):
And then next Monday, we’ll roll it out to everyone. But that’s, I think where you run into issues, you know, there was a story about Uber and they, they discovered that a big portion of their paid advertising was not actually driving conversions and they just turn it off altogether. And they’ve, you know, commercial is dropped a force, but not as much as they expected. So even a company like Uber that has all these resources, they are struggling with the attribution as to what role does something like paid ads play. So if you’re a small company, you want to be more careful and you want to be a little bit more conservative, um, because you may, you may, um, you may run into things that may be harmful, you know, had a client that they ran. They came up with the idea of a, of a discount for their service.
Ruben Ugarte (35:06):
And it was only about 10% of, of ’em off with the top pro uh, product price. But any test that they just rent it and they just actually deployed it to everyone. It was a full rollout and it worked, uh, the, the discount got more customers during one month, let’s say January. But then what they realized is that their numbers in February that affect it because they’re really, they really kind of capture a lot of the, the potential February conversions in the January. Um, so it, it was really more harmful in the long run than the short run. So things like this can happen. So do you want to be more careful and having well, they see companies maybe don’t give right as much intestine. It’s not so much the lack of then or doing it enough is that they’ll design the test property, right. They’ll have the right sample sizes, the right conditions for what success means. So they run it. They’re told test themselves as a success and they just, okay. You know, our AB testing tool set, it’s valid, it’s confidence. Let’s roll it out. But I do think there’s a little bit more thinking that needs to go before we can do that kind of rollout.
Jim Rembach (36:13):
Well, that’s very helpful, but at least it gives people an understanding that it is a process, right? And the fact is you must do it. I mean, we have, we have to be able to try different things in order to measure their effect. Because I mean, you, you tell me, I mean, what could a 1% change mean in regards to conversion?
Ruben Ugarte (36:35):
There’s, there’s that math where, you know, if you were to get better by 1%, every day in about 70 days, it’d be twice as good. Now, maybe for companies, they might not get 1% better every day. But if you do that, let’s say every couple of weeks, then by the end of the year, you could be doubling conversions. And that could really double revenue and double profit and so on.
Jim Rembach (36:54):
It’s huge. It has some huge potential data can definitely help you in a number of ways. Um, no pun intended that you could have never imagined, imagined, and relying on your gut is not the way to go. It just isn’t. You have to always test yourself because what you knew prior to March, 2020, this is the same Ruben. I had fun with you today. How can the B2B DM Dan get in touch with you?
Ruben Ugarte (37:25):
Well, they can check out my website that is, uh, Rubin in regard to.com. Uh, you’ll find links to the book, that data Mirage, you know, I think there’s a lot of different ideas there. If you’re an individual entrepreneur, a coach consultant, I think there’s a few things that will help you better understand how to deal with your clients. And if you’re a small company, then you’ll understand what’s maybe the right approach for your stage, uh, which is I think vastly different from what a, you know, a fortune 500 companies doing. So I would say those, all those things can be found at the websites, uh, in anything else, blogs, free content. It’s all about
Jim Rembach (38:02):
Ruben. Thank you for sharing your knowledge and wisdom, and we wish you the very best. Thanks. It’s a pleasure.