Podcast

How we organize messy data

Callahan Agency | June 18, 2018

In marketing, we have a real problem with messiness of data. There are thousands of systems that we can execute through, whether that’s email, or media, or paid search, or social. Even in just the social space, there are thousands of tools to manage that stuff. Every tool produces its own little data set, their own stream of data, and nothing talks to each other.”

In this podcast, Zack Pike, VP of data strategy and marketing analytics at Callahan, explains how to organize messy data and select the appropriate marketing analytics tools for your brand.

Listen here (or subscribe on iTunes,  StitcherGoogle PlayGoogle Podcasts, Pocket Casts or your favorite podcast service):

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 Jan-Eric Anderson and Zack Pike.

Jan-Eric:
Hi. I’m Jan-Eric Anderson. I’m vice president of chief strategy officer at Callahan.

Zack:
I’m Zack Pike, vice president of data strategy and marketing analytics at Callahan.

Jan-Eric:
Today we’re gonna talk about one of Zack’s favorite topics, marketing analytics, which is actually one of mine too. Truth be told. Zack, who leads our marketing analytics discipline, our team, essentially came in and established this. You basically started from scratch, built something from nothing. It has ramped up way faster than we could have hoped in terms of how clients have engaged with it. When we go into clients, when you and I walk in together to talk to clients, and we’re talking about marketing analytics and this topic, their ears perk up, the hair on the back of their necks starts to stand at attention.

Zack:
Yeah.

Jan-Eric:
It’s clearly a topic of great interest for our clients.

Zack:
You know, in marketing, we have a real problem with messiness of data. There are thousands of systems that we can execute through, whether that’s email, or media, or paid search, or social. Even just the social space, there’s thousands of tools to manage that stuff. Every tool produces its own little data set, their own stream of data, and nothing talks to each other.

Jan-Eric:
Well, then I can speak to that for sure. I mean, with my background in media, you know, we do a lot of media reporting.

Zack:
Yep.

Jan-Eric:
When we’re reporting on paid social, we have one look of a report and if we’re doing paid search, we’ve got another report, and if we’re reporting on a TV plan, or a mobile display plan, we have these different reports. If we’re really on top of our game, we can pull it all together and get it nice and concise into a single report.

Zack:
Right.

Jan-Eric:
It’s hard to understate the, or overstate rather, the amount of labor that it takes to get that done.

Zack:
Yeah.

Jan-Eric:
It’s quite an effort.

Zack:
Labor is a good description. It’s very labor intensive to deal with all this data.

Jan-Eric:
Number crunching.

Zack:
Exactly. You know, in my background, what’s always been really successful is having what I can as analysis-ready data. I know you and I talked about this when I started and …

Jan-Eric:
What does it mean?

Zack: Right. Right. Analysis-ready data is having data in a really efficient structure that makes it easy to ask a lot of questions of it. That’s the biggest problem, especially in large organizations. When you’ve got a group managing the data, often times it’s not in a structure that makes it easy to ask question A and then question B, and then question seven, and then question 45. You can’t anticipate everything.

Jan-Eric:
Right.

Zack:
That was really the plan when I came in, was … I know there’s a lot of data. I know this agency already has a lot of data floating around. We’ve got to get it into an analysis-ready format before we ever start to answer questions with it. I mean, it was really that simple when I came in. I think it’s been successful so far, but there were three key things we had to do to get to that point. Right?

The first one. We had to have the technology to be able to handle that data. In the marketing space, not only is the data really messy, but there’s a lot of it. That’s not stuff that you can manage in excel. That’s not stuff that you’re gonna keep on a desktop somewhere. You know, probably the first, keep me honest, month or two, maybe into three months, we worked on the technology. It was getting a good solid database set up. Something that we could use across clients and keep everything secure and private and meet all those requirements.

Then we had to be able to visualize that data because people like to look at stuff and if you can’t see it, it’s hard to make decisions off of it. There are a lot of other technologies in there to make it efficient, but storage and visualization are two very, very important pieces. The other piece of that technology world is collecting the data. It’s one thing to put it into a database and store it. It’s one thing to build charts and graphs and interactive dashboards off of it, but if I can’t get data in there efficiently … cause the clients aren’t gonna wanna pay for the labor to do that. That was a big piece of it as well.

We have a lot of API connections into all of these services so that it doesn’t take a human to collect all of that data every day, week, or month for reporting. That’s really the first step to getting analysis-ready data. It’s a good piece of technology. In fact, most time it takes a collection of technologies, which is what we use.

However, where people get into trouble, is they think technology is the answer, right? We go out and buy Tableau software, or we go buy a Google cloud and since we think that that is going to solve our problems, but there’s two other very important pieces to that equation. The first one is having a plan around the data. I mentioned analysis-ready data was kind of the intent of all this stuff.

You have to have a plan on how you’re gonna get there, otherwise you start collecting data from all over the place. You know, if you don’t really have an idea on how you’re gonna use it, or the questions it needs to be able to answer, you start putting it into your technology in a format that, when we start asking question number 47 or question letter Z or A, it makes it harder. It takes a lot more time.

Jan-Eric:
Well, and speaking to the plan. I mean, we would talk quite a bit about what we’re trying to do and where we think we’re going. I don’t recall at any point you said, “Well, you better get it right now, because once we get started, we can’t change things down the road.”

Zack:
Right.

Jan-Eric:
I think the intention here is that you have an idea of how you intend to use the data, what are the types of views you’re looking for, what are the types of decisions you seek to inform.

Zack:
Yep.

Jan-Eric:
Right? Having that type of an idea and turning that into a plan based on where we think we’re headed, this is the kind of functionality we need to have. This is the way we need to structure data. These are the types of data sources that are required to …

Zack:
Right.

Jan-Eric:
Lead us to where we wanna go.

Zack:
Building in the flexibility to be able to handle the things you can’t anticipate. To your point you just made. The marketing space changes every day. It’s like something new comes up. There’s a different way to do things. If you’re in an environment where, once you put it in it stays that way and it never changes, and you can’t be flexible, you’re never …

Jan-Eric:
It’s the path to being obsolete.

Zack:
Right. You’re never gonna win in the analytics space in marketing. That works in other industries, but it doesn’t work here. Yeah, that was a really important piece. I mean, you can probably speak to this better than I can. We’ve, even just in the past four months, had to change ways we’re doing things, right? We have different clients coming in asking for different things. It requires a different approach to the data that we didn’t anticipate.

Jan-Eric:
No question about it. I mean, different categories, clients that compete in different sectors, in different industries have different things that are important to them.

Zack:
Right.

Jan-Eric:
Also, different variables that impact their business. Here’s an example, we’ve spent a ton of time talking about is, weather.

Zack:
Yeah.

Jan-Eric:
The impact of weather that it has on different categories, and the way you look at weather …

Zack:
Yeah.

Jan-Eric:
is something that’s kind of, a need for a variety of [00:08:30] categories. If we were to start to work with a client then, that needed to have weather as part of their analysis, to have that flexibility to be able to bring that in is really, really important.

Zack:
Right. Right. Yeah. Absolutely. I mentioned that there were three things. I talked about two. Technology and the plan. The most important piece of this equation … Again, I’m gonna keep going back to the technology, that it does not solve your issues, is you have to have someone, or a group of people who really understand data. I call ’em data engineering type mindsets, that have to be in there.

It’s people who have dealt with this mess before, who have run into the problems that you’re gonna run into, who have had to work through and find solutions to the craziness that’s gonna happen when you start taking Facebook data and putting it with AdWords data, and then putting that with media data, and then trying to put weather on top of that, and then trying to put  Walmart revenue on top of that. There are a lot of issues that come up in that equation.

If you haven’t dealt with it before, haven’t dealt with something like that, it slows things down substantially. You know, the good thing about marketing analytics is there are some amazing things we can do with it. The bad thing is that there is no playbook. There’s no resource you can go to and say, “Hey. Here’s exactly how you do it.” Right? There’s consultants, and there’s agencies and companies that do this all the time, that you can tap into, but if you’re someone who’s trying to get something started in your organization, there’s no playbook you can go to, just to follow steps to do it. You’ve gotta have some smart people who can problem solve …

Jan-Eric:
Right.

Zack:
once that data starts flowing in. You’ve gotta have an organization that’s receptive to dealing with those issues as well, and helping that person think through different ways of doing thing. I mean, we’ve sat at my desk and gone back and forth, trying to find answers to questions and asking different questions of the data running into roadblocks days before. It takes that to, one, answer those questions, but two, figure out the holes in your data strategy that are gonna get you to the analysis ready data that you’re really going for.

Jan-Eric:
Yeah, I would add onto that, and talk about that, having a strong data analytics partner in place, what that enables is, it enables a team of curious-minded people to ask and continue to ask more interesting questions. It’s inevitable. It’s human nature. You walk into solving a problem with an assumption of what the answer is. What I have found so rewarding about working with you in this environment, with having this analysis ready data is that we can continue to ask more and more questions, but not only ask the questions, we have the power to find answers to those questions through data.

Zack:
Yeah.

Jan-Eric:
And having data that we can connect in, overlay with different sets of data can be overlaid together. It’s been rewarding. It’s very fulfilling personally …

Zack:
Yeah.

Jan-Eric:
to work in an environment like that. Frankly, I think it makes the work better.

Zack:
Yeah. Yeah. I think super fulfilling for me too. When you see internal stakeholders or a client making decisions off of good sound data, cause a lot of times it’s not sound in the marketing space. It’s pretty cool.

Jan-Eric:
Yeah.

Zack:
It’s really awesome.

Jan-Eric:
Clearly it’s a hot topic, and an area where a lot of marketers are, and advertisers are, needing help in this area. For us as an organization, I think we are delivering a better service back to our clients, leading to their satisfaction, because of what analysis data enables.

Zack:
Yep.

Jan-Eric:
It enables us to explore, to ask more interesting questions, but not only to explore and ask more questions, but to be able to discover what the answers are …

Zack:
Yeah.

Jan-Eric:
which inevitably will lead to a more decisive set of action and recommendations that the agency is making back to clients.

Zack:
Yeah.

Jan-Eric:
It’s been very liberating. I feel like the shackles have been removed. I look back at the days when I considered marketing analytics to be basically glorified media reporting.

Zack:
Yeah.

Jan-Eric:
Desperate reports that were really masked as “analytics.” I put air quotes around that. We’re down really into the business of truly doing marketing analytics.

Zack:
Yeah, yeah.

Jan-Eric:
It’s been great.

Zack:
Well, let me ask you this question. Do you feel like most stakeholders are getting their heads wrapped around that, most agencies, or do you feel like this is still kind of an out there thing?

Jan-Eric:
On the agency side, I think that agencies frankly struggle to figure this out.

Zack:
Yep.

Jan-Eric:
I think that agencies are good at evaluating their own work.

Zack:
Yeah.

Jan-Eric:
The precedent in that is through looking at how well creative might be liked.

Zack:
Yeah.

Jan-Eric:
What kind of engagement social media campaign received, or what the unique reach was of a TV campaign. These are media metrics. These are creative campaign metrics.

Zack:
Yeah.

Jan-Eric:
I think that as an industry, advertising agencies have gotten into a trap of looking at their own work. It’s a little bit of navel gazing …

Zack:
Right. Yeah.

Jan-Eric:
and critiquing their own work. I’ve not run into very many clients though that wanted to buy a campaign, a creative [00:14:30] campaign. They didn’t wanna buy an award-winning campaign. They didn’t wanna buy media impressions.

Zack:
Yeah.

Jan-Eric:
They didn’t wanna buy Facebook likes.

Zack:
Yeah.

Jan-Eric:
What they really wanted to buy, whether they said it or not, was more business.

Zack:
Right. Exactly.

Jan-Eric:
That’s what this marketing analytics component has brought in. It has shown a light. It has illuminated a path to connecting business results back to what we’re doing. At the end of the day as an agency, we’re making stuff that’s intended to drive business results. Now, I keep saying this, the shackles are off. We can put all of our attention on driving business for our clients. I think that’s really the linchpin on driving the client satisfaction.

Zack:
Yeah, yeah. I agree. If I can kind of wrap this up with a couple of final points. It feels hard, but getting started somewhere is better than just throwing our hands up and saying that this is too difficult. I’m just gonna stay in Excel and keep doing my normal media stuff. Asking questions, trying to gather more data, is going to help anyone figure out where those holes are, and then find solutions, and find the hole and find the solution. That’s how you get started.

Jan-Eric:
Yeah.

Zack:
It’s something that any organization can do. Personally, I think every organization should be doing it, because of the reasons that you just mentioned.

Jan-Eric:
Your humility breaks through because you make it sound easy. Clearly it’s not easy or tons of people would be doing it. Thanks for joining us. I’m Jan Eric Anderson. He’s Zack Pike. We’ll talk to you again soon.

 

Thanks for listening to our Uncovering Aha Podcast. Callahan provides data savvy strategy, and inspired creativity for national consumer brands. Visit us at Callahan.Agency to learn more.