Podcast

How to take your marketing from data-driven to data-inspired

As a CMO, if you’re living in a binary world of marketing, you’ll never find anything you didn’t expect. Digging into analytics leads to questions and insights, leading to even more questions, requiring further exploration of the data. While it can feel as though you’re getting nowhere, the reality is you’re getting smarter as you go, revealing insights that can change the direction of your marketing strategy for the better, leading to more data-inspired marketing, versus data-driven.

In this podcast, Jan-Eric Anderson, chief strategy officer, and Zack Pike, VP data strategy and marketing analytics at Callahan, discuss the differences between data-driven and data-inspired and why it’s important to lean toward being more data-inspired when it comes to marketing decisions.

Listen here:


(Subscribe on iTunesStitcherGoogle PlayGoogle PodcastsPocket Casts or your favorite podcast service. You can also ask Alexa or Siri to “play the Uncovering Aha! podcast.”)

Welcome to Callahan’s Uncovering Aha! podcast. We talk about a range of topics for marketing decision-makers, with a special focus on how to uncover insights in data to drive brand strategy and inspire creativity. Featuring Zack Pike and Jan-Eric Anderson.

Jan-Eric:
Hi, this is Jan-Eric Anderson, chief strategy officer at Callahan.

Zack:
And this is Zack Pike, head of data at Callahan.

Jan-Eric:
All right, Zack, I’ve got a topic we were going to talk about today. I want to tee this up to you and this is getting some experience we’ve been having recently. The topic I want to talk about is the perception of the role that analytics can play and what role they can play for a company, for a corporation.

Zack:
Okay.

Jan-Eric:
I don’t think you would find very many companies that will sit there and say data isn’t important, but if you were to go and survey companies and ask them why are you using data? What is your expectation of data and data analytics in the marketing function? You would probably get a couple of different responses, types of responses you could probably categorize and categorize into a couple of things.

Zack:
Okay.

Jan-Eric:
This is what I think we would hear. One is that I expect it to tell me what’s right and wrong, a very binary view of yes or no, black or white on to validate an assumption, a hypothesis, whatever. The other option is that data is here to help me learn, to provide context, to provide information or deeper understanding around factors in business. Is that fair? These are probably generally the two types of expectations that clients would have for data.

Zack:
Yeah.

Jan-Eric:
Okay. Yeah, yeah. All right, so are there … How important is it to … I guess, is one better than another?

Zack:
I don’t think so. So the two lanes I am, it’s binary, it’s not binary, right? That’s the summary of the two ways. I think that that that is probably dependent on the decisions you’re making and the type of business that you’re making them against. We spend most of our time in the marketing space, and binary data driven decisions in the marketing space are really difficult, unless you’re in a really highly transactional e-commerce environment. It’s all digital marketing. It’s like a factory of leads coming in and purchasing on the website. It’s hard to have it be a binary decision, because … or binary data-driven decision, because customers are buying things in stores. They’re getting hit with channels from all over the place. They might be buying some online and some in store. The attribution path is very different.

Zack:
This reminds me of a podcast that I listened to. This is probably from a couple of years ago from Google, where they were talking about data-driven decisions versus data inspired decisions. I think that while most marketers, most business people would like to be on the data-driven side, where we spend most of our time is on the data-inspired side. It’s the less binary side.

Jan-Eric:
That data-driven thing, it gets back to the assumption that hey, if we have data analytics, it’s going to tell us right from wrong.

Zack:
Right.

Jan-Eric:
Now there are cases where it can be used for that.

Zack:
Absolutely.

Jan-Eric:
What are examples of that? I mean, I guess what goes through my head is that the place where data-driven is most … or this this binary view would be maybe on less significant or smaller, very tactical questions.

Zack:
Yep.

Jan-Eric:
Do people prefer the blue M&M or the red M&M, right?

Zack:
Yes.

Jan-Eric:
That is a binary type of thing. Not a lot of variables going into that type of a decision or that type of an analysis.

Zack:
That is a good way to describe it. The small tactical decisions, the binary world works really well, and that can be a lot of stuff. I mean that can get you pretty far. This is … Actually, as we’re talking, it’s reminding me of a podcast we did a while back where we talked about some of the frustrations CMOs have with data and analytics and their expectations around it. It’s reminding me a lot of that podcast, because in that one we talked about these longterm, big, sweeping changes and wanting to use data to drive the decision around that. That’s where it gets really difficult and much more hairy, right? It’s-

Jan-Eric:
Because you start doing analysis and this is where we get into the data inspired. You start looking at data and at information. Sometimes it just, it leads you to more questions, or oh, okay, I just learned one thing, but that just raises three more questions, which then requires further exploration along the way. You may start to feel like, man, we’re not getting anywhere. We’re not getting closer to a decision. The reality is you’re getting a lot smarter as you go. You’re being a lot more … You’re getting a lot more informed the further you get down the pack.

Zack:
I think part of it is just resetting the expectation, right? It’s, I have … So if I’m in a normal business, I have just any restaurant, a retailer, someone selling through a retailer, I need to be honest with myself about what I’m using data and analytics for. It’s what you just said, the tactical, minute, very micro decisions, I’m probably in a good binary environment. Stuff that I can test really quickly, measure those tests and make decisions and move on. But if I need to make big longterm changes to this company, to the way we operate, to the way our brand is perceived by the customer, to how we approach customers from a communications perspective, that’s where I have to be okay with saying, okay, binary decision, it’s out. It’s not part of this equation anymore.

Zack:
I’m going to take small, incremental steps to get smarter about the business over time and get smarter about this decision that I’m thinking I’m going to make. The way that we’ve done that is we go down the testing path, but it’s a little bit different of a lens into that testing path and it’s having the ability to be flexible about what you’re doing inside that test. I’m thinking of an example that we’re actually working through with a client right now, where we’re testing a very substantial change to the way that marketing has been executed in the past.

Zack:
It’s a massive departure from the way they’ve done it previously. We think it’s the right thing to do, but we’re … Listening to this podcast over time, you would know that we are big fans of putting our money where our mouth is and let’s actually test this theory. Through that test, we had an expectation on how we were going to measure it. It was actually fairly binary, right? We had a typical test to control setup. If the test group achieved a certain threshold above the control group, we were going to call it a success and we were going to move forward and everybody would be happy.

Zack:
But you get into this and you realize, oh geez, there’s some variables that we one, didn’t account for and two, probably couldn’t have accounted for in some cases. So if you’re doing your due diligence on the analysis and understanding side, you start to say, okay, well maybe I need to throw out my original measurement mechanism. This is becoming more of a data inspired decision. I’m using this test now to learn more than I knew before about the business and about this decision we’re thinking about making as I work my way through and navigate the waters of this test in a different way than I planned to in the past.

Jan-Eric:
What I’m hearing from you is that you’ve got to be … there’s a requirement for flexibility in data inspired in that example you were just giving, right? So the requirement to be flexible in how you’re learning and be adaptable to changing conditions, and as new variables are exposed or things that you didn’t anticipate, couldn’t have accounted for ahead of time as you start to learn that, you adjust course as opposed to sticking with maybe the binary view of what the result would have been.

Zack:
Because it’s difficult, because it’s getting us even further away from the silver bullet mentality around the analytics. That this thing is answering my questions. It’s this magical world. Because now, I’m walking in saying, hey, what we thought was going to happen isn’t happening, and now, I have to explain all of the pieces to that puzzle and why it’s not happening and what we’ve learned as a result. So once you get them to understand that, now you start to say, okay, well here’s what we’ve learned and here’s what we think is actually taking place. The reason that’s hard is because the people who you’re communicating to were in that binary world. You have to help them over to the more inspired, less binary world to understand this equation.

Jan-Eric:
I’m wondering, do you feel like that there are trappings of just reinforcing existing bias if you are … if you tend to operate or look at analytics through the binary binary view? So here’s what I mean by that. I come into a situation with preexisting beliefs that I think are … this is the way it’s going to work. We’re going to … More people are going to pick blue M&Ms than red M&Ms. I know that, and so I’m going to come in with this binary view. But what I discover is that I guess if I have a bias for it, is there potential for that? If you’re going into a binary view on this with … If you subscribe to a binary view of analytics that you would be able to reinforce, you’re going to … potentially going to go in and just reinforce the biases that you already had.

Jan-Eric:
In other words, you come in looking for the thing you thought you were going to see in the first place. Whereas, if you come in with more of a data inspired mindset, you don’t carry the burden of having to prove what you already think. You’re really coming in with a mindset of learning what I can from the test.

Zack:
Yes.

Jan-Eric:
So I guess maybe an invisible trapping of being data driven and looking through a binary view is that you may not even recognize your organization’s tendency toward bias and just proving what you think you already know.

Zack:
Right, and even as an analyst, it’s easy to get trapped in that too. In trying to please the people who you’re communicating to based on their bias towards the binary decision.

Jan-Eric:
Man, so just maintaining objectivity and credibility as an analyst. It really is probably much more preferable-

Zack:
Oh yeah.

Jan-Eric:
… to be data inspired rather than data-driven.

Zack:
Yeah, I mean … And it makes … If you think, if you really think about this and you look at the analytics industry in general, it makes total sense. There is not really an analytics product or person who’s solving world hunger, right? It’s like I have-

Jan-Eric:
Well, there might be some.

Zack:
Well maybe, and world hunger is a bad example. But anytime someone tells me about some magical analytics process or tool or something, my response is always, well, why is that not being used in the stock market to optimize what’s going on? Why aren’t they just making their own money with no boss and no responsibilities on their own? Because it’s not actually that clean. It’s not actually that clear. If you can wrap your head around that and just be honest and say, okay, it’s not binary. There’s very few decisions that are binary. Then it’s like this. At least in my experience, it’s been this freeing world to say, okay, I’m approaching these problems now from a unbiased, anything can happen type of approach. It’s like these tests that we run, I think many of our client’s biases towards the binary result … I’m doing some introspection here. I’m talking through this while I’m thinking through it, but we probably don’t do a good enough job of helping them understand that this is a very fluid thing. We don’t necessarily know what we’re going to find out.

Jan-Eric:
Well, yeah, but here’s where that I think it’s hard for someone in a company to not want to have this binary view.

Zack:
Yeah, absolutely.

Jan-Eric:
There’s extreme pressure today, on today’s CMO to deliver results and to be able to say I’m leveraging data analytics to be able to make better decisions.

Zack:
Yep.

Jan-Eric:
The quickest, easiest application or evidence of that would be through a binary result of a test. We tried this email and we tried this email. This one had better open rates. This one generated much more e-commerce than the other one. Look how smart I am. I have used data analytics to create this binary view, and this is the winner. There’s extreme pressure on a CMO today to produce the binary results. The reality is though, the greater value, I think, … Our point of view of this is that the greater value of data analytics is in through the greater learning that comes through the exploration, the curiosity and the discipline to keep digging and keep exploring and keep learning. That really, you end with a wealth of information and understanding about your business much greater than, back to that horrible example, this email or that email, which one made more money.

Zack:
I think another thing I’d say is that is the only way that you’re going to find those unexpected insights if you’re open to that way of doing things. If you’re in the binary world, you’ll never find anything you didn’t expect. The only unexpected finding would be it was the opposite of what I … It was a one instead of a zero, and it wasn’t a zero instead of a one. Where this approach, like some of the things we’re finding, we totally did not expect. It’s really informing the way that we would approach the future with a given client.

Zack:
Even in my past experience closer to the finance world, every time where we took a binary approach, it was the binary answer and we moved on. The times where we took a more fluid, flexible approach saying I’m just going to see what happens as a result of this, and I’m going to rip it apart and look at it upside down and inside out, those are the times where sometimes you found the most valuable insight. The stuff that’s like, holy cow, we didn’t even know that this was a thing, but now that we do, what do we do with it, right?

Jan-Eric:
What I think in summary, what this also reinforces is the idea that analytics doesn’t necessarily have a clean beginning, middle, and end. It’s an ongoing, omnipresent discipline that needs to be applied that can really generate great learnings. Subscribing to the idea of being, trying to push an organization into a state of data inspired versus data-driven does not mean you’re sacrificing learning from analytics or actionable insight. Quite frankly, it takes the lid off of the potential to be able to uncover new ahas through the data and through analysis that may lead you to some great learning that you had no idea even existed. I know that we’ve seen that in our partnership at Callahan and how we approach it.

Jan-Eric:
So I think that in summary for listeners, the idea is that your organization might lean one side or another, be more binary view of data and analytics and being data-driven, what we would describe the data-driven, versus data inspired, which is more of the exploration piece. Our advice would be to, if you lean or have a tendency to one side or the other, while there’s a time and place for everything, if there’s a tendency broadly, it would be a much more beneficial to lean toward being data inspired given all the potential that that it leads to in terms of learnings. Zack, thanks for the conversation as always. It’s very stimulating. I appreciate you being here.

You’ve been listening to the Uncovering Aha! podcast. Callahan provides data savvy strategy and inspired creativity for national consumer brands. Visit us at callahan.agency to learn more.