A weekly Google Hangout dedicated to discussing content marketing, search marketing, SEO and more.
Topic: Create a Data Driven Content Marketing Program
This week, our guest Steve Farnsworth joins the conversation to help marketers know how to effectively measure content marketing efforts against goals. We talk about getting the right data and knowing how to analyze it to focus on determining what’s working and what’s not.
Steve Farnsworth, CMO The Steveology Group
Erin Robbins O’Brien, President at GinzaMetrics
Karen Scates, Manager Marketing & PR GinzaMetrics
FULL VIDEO TRANSCRIPT
Karen: I just want to welcome everyone to this special edition of our FOUND Friday, for the first time airing on Blab. The episode is going to be available as a podcast on our SoundCloud account as well as a video on our YouTube account after this. So, lots of ways to see us.
This week, Steve Farnsworth – hi, Steve – is joining Erin and me to talk about creating a data-driven content marketing program. But before we get started, let’s just do a few introductions. Steve, why don’t you start?
Steve: I’m Steve Farnsworth. I am the CMO at the Steveology Group where we create B2B content for demand generation. I can get found on Twitter @Steveology.
Karen: Awesome. Thanks. Since we’re new to Blab, Erin, why don’t you go ahead and introduce yourself?
Erin: Sure. I’m the president of GinzaMetrics. I can be found on Twitter – personal Twitter only at @TexasGirlErin and work-related stuff @GinzaMetrics. Ginza is a search and content marketing analytics platform.
Karen: I’m Karen Scates. I’m the manager of marketing and PR at GinzaMetrics. I just want to remind everybody, if you like what we’re saying, you can just go ahead and give us a little applause here and we’ll be competing for those, I’m sure.
Steve: Yeah, to win something at the end?
Karen: Ooh. Maybe like a cocktail, I don’t know.
Erin: We’re all winning those. Thank you. I don’t care whether I get applause or not.
Karen: The one with the least number of applause gets the most cocktail maybe.
The most recent Content Marketing Institute trends said that 45% of marketers don’t know what a successful content marketing program looks like. That just struck me and surprised me, so I we thought we’d talk today about what’s going on. Are you guys surprised by that number?
Steve: Not really. We say, “How do you measure a program?” I think part of the problem is like saying, “How do you have a successful website? What’s the measurement? Is your website successful?” It’s a really broad question.
All the pieces that make up content marketing program, it’s a collection of, ideally, dozens of programs you’re doing. Each of those programs has to stand on their own so you have individual KPIs. What I think that probably speaks to more loudly or broadly is that people aren’t necessarily truly tracking the improvements doing campaign to campaign. I think that’s probably what’s going on.
Erin: Similar to Steve in the “not surprised” situation, but I guess maybe the surprise is I’m surprised I’m not surprised. I just figured at this point, we would have worked a little bit more towards some cohesive ideas of what good campaigns look like or what good programs look like.
It feels a lot like we’re still very disjointed in terms of marketing channels, marketing methodology, tactical decisions versus strategic decisions, who we expect to be strategic in certain areas where we place emphasis and budget on things. There are so many different moving parts that I think a lot of times what happens especially depending on the size of your organization and the longevity of your organization. How long has this department been around? What did they try to piggyback onto things? What did they try to split off into its own group? How have you hired agencies, consultants, different things?
I was in a meeting a couple of weeks ago. In the call, where seven different agencies dialed in and what you could hear was a lot of people not really sure who owned what. Or taking credit for things, like four different people took credit for. There’s a lot to tackle when it comes to: why don’t we know what a good marketing program looks like.
Steve: I think two falls here. One is I think that management still is learning what being a technical [technologist] looks like. That’s all very new for folks. I think there’s between ignorance in the rank in file about good KPIs and understanding in which of these KPIs really come from management at some level. You have to say, “Here’s our goal. How are we going to measure it when we’re making those goals or moving that chain?”
A lot of times, I know people who are still measuring social media outreach and activity by followers. This is a question I’ve had multiple times just in the last month from people who make a lot more money than I do. I swear to God. That was one of the clear KPIs. That’s just silly. You’re measuring followers on LinkedIn to measure your success in marketing or just purely by shares or purely by the eyeballs that read it.
Those are all interesting KPIs to have but is it leading to secondary like a proxy KPI where if someone reads this, do they tend to subscribe or so they tend to ask for the next later stage content or some other piece that’s proxy to. Most of the goals will be around lead generation. If it’s not, you need to have pretty good conversation about why that’s not.
So if you don’t have a proxy piece that gets downloaded or some deeper engagement, I don’t think we ask for that. We tend to stick to the light ones. How many eyeballs did we get? How many pageviews do we get at some level?
Erin: Where is it all that getting rolled up to do? You’re counting things, you’re measuring the stuffs, and then you hand off numbers and reports to somebody, and then what gets done with that later? My biggest fear all the time is whenever anybody asks for new analytics they’re like, “Hey, can you add this into the dashboard?” or “We’d like to measure this.” I’m like, “I have no problem building something that will do that for you. I need to know what you plan on doing with it. One, so I can build it the right way, so you can actually use it. And two, who asked for it? What are they going to do with this information once they have it? Because that always means a lot of different things when you’re talking about outputs and insights, and then how will it be integrated?”
One of the big things that we talk about all the time is data normalization. Everybody is measuring things but they’re measuring them in a lot of different ways which means when you get to these bigger insights and more like business intelligence (BI) layer of analytics across an entire organization to make overall smarter decisions, you’re not even comparing apples and apples. You’re not comparing apples to any other fruit at this point. You’re just somewhere off in the vegetable patch because you can’t re-aggregate things because like you said – it’s what’s the core goal. For most people, it’s rather new. At some point, it’s money. Money is a goal and that means conversions. If you can’t tie what you’re doing back to conversions, what are you doing every day?
Karen: We’re talking a lot about KPIs. Why is KPI such an issue? Why can’t people settle on KPIs?
Steve: Most people just don’t know what to ask. It’s always embarrassing. It’s always hard to put yourself out there and be measured. That’s scary because you might fail. So you have to have a certain culture of understanding that this is we’re going to shoot for a number in the dark and measure where we are and see if we can increase that number. That’s what we’re really doing. What people really think it is is like, “If I don’t get 100 downloads and I only get 13 downloads, that’s my job.” I think some of that is part of doing it.
It’s a really huge piece. People don’t always know what the goals are. Honestly, when you have the right goals, you tend to pick things that are so high that they don’t connect to any kind of departmental or organizational goals. Or it’s so low that it’s irrelevant to what they’re going after. So there’s this huge disconnect.
I have to say that a really large part of my time in consulting and probably a more critical piece is actually sitting down with CMO and the director of lead gen and the content marketing manager. They’ve done personas, they’ve done this kind of stuff, but they haven’t really sat down and really looked at the data and said, “Who are we really talking to? Does our content really map to these people? How are we going to measure that?”
That seems like [9:04 inaudible] for conversation. If you ask if they did it, they’ll say yes. If you start digging into it, you see a disparity between what they say they’re doing and the knowledge that you need to go to drive a good program and then set appropriate KPIs.
Erin: This is one of our conversations we have all the time from a search perspective. The KPI thing is hilarious. It always falls into some category which is, they don’t know what goals to set. They don’t know how to set them because they don’t know what all goes into them. They have conflicting KPIs across departments or, as you mentioned, they don’t know what the larger goals are and how to create some sort of connective tissue between their job and this larger goal.
We have this conversation with search all the time which is essentially, it’s funny when somebody says, “I want to build a website and I want to build a website about X. I want you to make it rank.” I’m like, “Why don’t we back it up a little bit and you find out what are people looking for? What is it that you do that people are [10:02 inaudible] and how are they talking about it? What are the issues associated with that? Why don’t we build a site and build content around things that people actually give a darn about?” As opposed to the other way around which is this idea of “I’m going to build it and then I want you to essentially fix it after it’s broken and make it findable.” Why don’t we just start out with findability in the first place?
This goes back to why don’t people have successful things? We’re not surprised. That’s the sad part. Why are we not starting there?
One of the things that you mentioned, Steve, is having these conversations with higher level people and they think that they’re doing some of this is, we have this conversation a year ago, ironically, and I had said that if you are in middle management or lower level tactical implementation and doing things, if you can’t get the people above you to tell you what organizational and departmental goals are, you’re working with jerks. They should tell you.
Steve: Absolutely. I think Carmen essentially had a really [11:14 inaudible].
Karen: Carmen just came in and she says that she thinks some make the mistake of reporting the KPIs that won’t get me fired or won’t shrink my budget. They’re playing it safe because, as Steve said, they’re like, “Oh, I can’t really set a real KPI because then if don’t meet it, I’m going to get fired.”
Erin: But at the end of the day, not converting and not making money is what will get people fired and eventually that comes back to you. Say I got a million Twitter followers and we didn’t move one more widget. At the end of the day, that’s still noticeable.
So if you’re like, “There’s only so long you can really do that,” what happens is we get a lot of number stuffing and then passing the buck. You’re hide things under layers that make it look like it’s not your fault or not your department’s fault, and then it gets shelved off somewhere else. The problem is a lot of times nobody follows the money. Nobody follows the buck to figure out conversion. That goes to something like attribution modeling like the discussion…
Steve: The KPIs, yeah.
Karen: The other thing is, is this setting the goal and then forgetting about it? We were talking earlier, Erin, about setting your personas up and then going, “Oh, we’re just going to write what the C-Suite wants to hear or what we want to talk about.” This problem of maybe going ahead and setting KPIs and then waiting for a year to measure it, how often should people be looking back at those KPIs and measuring their progress towards goals?
Erin: [12:50 inaudible] explodes?
Steve: There’s a good framework to think about measurement and there are three components. One is dashboard. I think you need to have a dashboard so you can basically get stuff daily. So you see what’s going on, the raw numbers, kind of what’s going on, if there’s anything that needs your attention.
The next level of opportunity is campaign analysis where you’re going back and looking at a campaign and seeing how you did that measurement. It should really be taking place a couple of places along the campaign and definitely at the end you’re reporting. That should be a fixed element.
I think the last piece – this is big. Generally campaign reporting doesn’t happen a lot. But the last piece is optimization where you’re looking at all that data and looking at, now it’s on maturity, what were the hidden opportunities we’re finding in this data? Where can we actually increase it?
There’s also danger. Sometimes I think this is the problem people start looking at. “Look, we got eyeballs for the ten best whatever article.” You need to make sure that the data you take from that optimized is targeting the people you want to speak to.
Erin: Man, there’s so much there. The persona thing is hilarious too. I don’t want to throw anybody on the bus but I’ll just speak in generalities. When it comes to persona targeting and creating things like KPIs, the persona thing is funny to me because what happens a lot is it’s like, we have these three personas and somebody wants to hear about this. What we did was we hired an agency or somebody internally, created a campaign, everybody decided “Fine, that’s the campaign,” because some executive said, “Yes, I like this.” Then they’re like, “But we have to make it fit these three things. So we’ll just add a line in there for moms and then we’ll just add a line in there for people in metropolitan areas. We’ll change this one little word.” I’m like, “That’s not persona targeting! That’s not what that means.” They’re like, “We did it and then we’re wondering why our KPIs are all out of whack and things look super jenky.” It looks jenky because when you said persona and when somebody actually was a persona, you guys are really far apart on what that actually meant.
With KPI measurement timelines, again some of this is like organization size. Looking at every KPI across every marketing channel every day for some places is really difficult but the counter argument to that is you probably shouldn’t be doing a lot of marketing activities that you can’t really manage, get useful insights on and optimize accurately on a regular basis.
If you have 50 marketing things going on at one time and you are an army of two or three, then what’s potentially happening is you’re getting a lot of information in that you could be using but you’re spread so thin, you can’t actually hone down and figure out where you could be doing things better. Sometimes the idea is crunch it down a little bit, pick a couple of things, really work on that.
From a campaign perspective – we talk about campaigns all the time – campaigns can be anything. It can be general awareness, it can be seasonal, it can be product-related or feature-related, it can be persona-related. You should be able to compare campaign versus campaign, campaign month over month, campaign quarter over quarter, campaign this year versus this campaign last year, campaign versus competitor performance for those same keywords or organic performance or social media monitoring and listening for that same campaign. With campaigns, it’s the idea that you can break things down and then measure them over and over again against each other, against similar things.
From a campaign, you can figure out: this persona’s campaigns are these five things. This other persona’s campaigns are these. And I can cross-compare within persona groups, across persona groups. If you’re not setting up your marketing to be able to do that, you’re missing out on so many cool nuggets of information.
Steve: I think that one of the challenges that people have is where to start. I said in the conversation that multiple companies should have all this stuff well underway. There’s about a billion dollar company – true story – and they weren’t really doing any kind of content measurement, which just blew my mind.
I think if you’re going to start looking at the data, people go – I get the question about – everyone wants benchmarks because that’s how we think. The reality is benchmarks are irrelevant to you because you have unique needs so you need to figure out what’s important to you. That’s a lot of work. Marketers don’t want to really get into that.
That’s a framework to think about that stuff. If you start from zero, what you really want to do – it’s like going to the gym. How many pushups did you do? How long did you run? That kind of stuff. That’s really what you want to start doing.
I think that there are three phases of grouping your different metrics. First phase, you can focus on traffic generation. Second phase is going to be engagement. The third phase is going to be conversion.
You’re looking at measuring traffic. You need visitors, pageviews, backlinks, source traffic – those kinds of things. But when you move on to measuring engagement, I think that’s the first thing. You can get those locked down to see where those numbers are. There are a couple of metrics you probably could use and that’s a good place to start.
But then you look at engagements, your next level or next phase. I would put in bounce rate, time on website, new or returning visitors, sessions versus pageviews, shares by content.
The last one will be when you start getting that leverage, that high level maturity, which is measuring conversion, the last phase. That’s when things start paying off. The first phase I think is almost like all the hard work. But now you want to look at opt-in rates, click-through rates, number of leads, any ROI if you’re able to track that. That’s a simple way to tackle your problem. You start looking at that data – your point, Erin – to pass data and see how you can compare.
Karen: And measuring content effectiveness is still one of marketers’ top five challenges according to CMI. Simplifying that to begin with maybe is best but I think getting those good numbers at the end, like you said, and going with the conversion, Steve.
Erin: A lot of people want to start out with conversion. They want to start with that idea of this thing converted at X amount. The problem there without doing the rest of this stuff is you don’t know why. You don’t know if it was a fluke. So we go back to the medium, method, and message thing all the time. Was it what you said, was it how you said it, or was it where you said it? Was it some combination of those things or was it who you said it to? These are all considerations that you have to look at when you’re trying to figure out content effectiveness.
So if you’re not set up to actually measure things in a way that allows you to understand that, to me, the point of measuring it is completely moot. Why do I care about why a specific thing converts so much if not only to toot my own horn and take credit and keep my marketing budget? Then secondly, to know what should I be doing more of or less of or differently or how does this combine with these other things? Or what does this look like in the competitive landscape or how does it shift over time? How do I leverage something that we had previously to make something better? There’s so much to that that when you just start with conversion, you’re missing out on so much of what leads up to it that actually is the real meat of the problem.
Karen: I have a comment from confluent forum.
Steve: It’s a good comment, too.
Karen: He says, “True, Erin, but that’s why you need to track and test your messaging, and then tracking them into GA [21:23 inaudible] conversions.”
Steve: Yeah. Tags and that sort of thing. That’s a really good way when you’re sharing content in different places. Google Analytics is a good way to track what kind of messaging is working. If you do Google Analytics, that’s a good way to track that stuff.
Erin: One of the things that we find a lot too is that a lot of different departments across larger organizations, not everybody in every department that may be more tactically focused has access to this information or has the ability to go talk to the people who are running their analytics provider or know what to do or certain things are set up but there’s a blocker in between there.
One of the more common things we run into when we’re talking to big organizations is a lot of people are taking credit for the same content. So some content will get created and then you’re having a fight between social, PR, all these different folks who are chiming in. So then the idea about measurement is lost on claim.
Karen: I think that’s about all the time we have for today. Steve, did you have any words of wisdom?
Steve: I don’t have anything to offer other than what we talked about. I think Carmen’s comment is pretty good.
Karen: Yeah, Carmen has a new comment. She says, “My favorite Google Analytics tool is Split Testing. There’s magic.”
Steve: One of the things with marketing, when you try to advise a client about a title or a name or something like that, there will be all these arguments that will just become silly at some point. So now if I’m trying to do a PPC ad or something like that for content or a webinar or whatever and the client goes, “I like it to be mysterious. Let’s just call it the biggest thing you don’t know about. I think they will click on it.” I go, “You know what, that’s fantastic. Let’s test that and see if that works. Now I’m going to test the ones that I’m suggesting and we’ll see how that comes out.”
Erin: I’m a really big fan of running experiments on things mostly because running the experiment is not my idea versus your idea. It’s what’s right, what works. When you look at it that way and you call them experiment, you’re not saying, “My way versus your way,” or “I’m going to win.” You’re saying, “Let’s try these things out and we’ll go with what works.” It doesn’t matter to me because what works is best for all of us. There’s no way I’d lose.
Karen: Awesome. Great conversation, guys. Thanks for joining us, Steve.
Steve: Thanks for having me. It’s been way too long since we’ve all been on camera. Erin, you look fantastic. Obviously, life is treating you well, girl.
Erin: Thanks, you guys. It’s so sad. We have to get together and do this more often. I have to see Karen every week. So, Steve, it’s good to have you in the mix.
Karen: She has to look at my face all the time. Thanks, guys.
Steve: Thank you for joining us.
Karen: Thank you for joining us.
Erin: Bye, everyone.