At last week's Adobe Omniture Summit in Salt Lake City, Utah, Ancestry.com's Head of Marketing, Josh Hanna, presented his company's marketing story in a panel titled "Inuition Is No Longer A Marketing Strategy: The Plight of the Data-Driven CMO."
Hanna discussed his company's marketing strategy and data's role in his marketing organizations day-to-day with AdExchanger.com.
Click below or scroll down:
- On the data-driven CMO
- Can TV and online work together?
- Ancestry.com’s marketing budget and where display fits in
- On cross channel attribution models
- Re‑targeting and its effectiveness
- Brand versus DR, online versus offline
- Tips for the “data challenged” CMO
JH: I've been with Ancestry.com for nine years in various roles, product management, content, valuation and marketing. And I've never thought about it any other way. I have always been data-driven as has Ancestry.com. I've always hired those types, too, and I think it's a cop out for people to say that half of marketing is art. Maybe a small part of it is, but for me, it doesn't have to be. It should be much more scientific, and we've just always held ourselves to account that way.
Certainly, there is a place for brand marketers that are more creative than analytical, and we have some of that in our organization. But even those people I still try to hold accountable. I mean, PR rolls up through me. This is not some fluff.
Well, we have to make it work, because it is so important. Our demo is first and foremost, influenced by television. We are an older demo. And yet we are online marketers as well. We have very robust display advertising, paid search, affiliate and email marketing and the rest. Ultimately, I think we're still in the first or the second ending here on understanding the mix.
We know that our online efforts account for over half of the sign ups that we get and they are attributed to multiple touch points. So we have to pick apart and try to assign value on everything from a display activity to an affiliate, to paid search, to natural search. And then you introduce television into the mix - you don't have that one to one relationship - you can't tie those eyeballs back to an affiliate program.
I'm pleased with where we are. But I think we have many miles to go. We are looking at all of our channels to understand how we attribute value to each of those. But, we take a great deal of comfort looking at that responsiveness from a 30 second spot or a 15 second spot and the immediacy of the impact.
I'll give you the headline figures. We spend about a hundred million dollars as marketers and most of that is working, media dollars. So, there are other items that roll into our subscriber acquisition costs that we report as a publicly-traded company.
Looking at display advertising, we see it as two different functions: it's brand awareness, in a way that television is brand awareness and DR.
Secondly, it’s re-messaging, which is just clever creative executions. To me, it's not much more sophisticated than that. We just need to be sophisticated about the frequency caps…
Regarding display budget, off the top of my head, television is 50 percent of overall and display is a good chunk. Other parts of our budget includes search where we're under no illusions. We're not building a whole lot of brand awareness and driving category work. We're capturing, I call it “The Google Tax” - the tax at the end of the funnel that we're paying for. And so we don't think of search as anything other than what it is. We'd love to figure out a way to continue to scale our display budget.
People who know me know that I'm never satisfied. I think it would be irresponsible for me to talk about spending a $100 million and say that we’ve got it nailed and we're satisfied.
There’s going to be a real premium on understanding interaction in my view. I think we're better than most, but I don't give us an A+ on our attribution and working media mix. There's always going to be a refinement because you start talking about moving even a million dollars from one place to another it can have a profound impact on your business and as a publicly-traded company, subscriber acquisition cost is looked at closely.
Isn't everyone going to love it if we can show subscriber growth with SAC coming down quarter to quarter? That's my responsibility to try to make that happen.
Let’s call it 20 percent of our display activity, but that would change based on, say, a home page takeover which can blow out your budget. But, it's an important part of our business. We take a lot of comfort in it because we say it's super ROI positive. And it is, but you’re re‑targeting so first of all you paid something to even have them engage with the site probably.
So to attribute something - maybe it's television or something else - you need to understand there were other dollars spent. So you don't want to delude yourself into thinking similar to paid search. It's not a marketing strategy to just focus on paid search. You’re just capturing the interest that's there.
It's an important part of our mix but we're not going to massively grow the category and the brand by continuing to optimize just that. That's just execution.
I will sound like a contrarian in a couple ways.
First, I'm just too new to this to have ever thought about things rigidly like brand and DR. I really don't think about it that way. I think we have our cake and eat it too. We look at our television and other activity driving higher brand awareness, but we also better see a near term result. It's not going to be good enough to just be brand.
And, we are definitely contrarians in that we have actually moved a bit of our budget into television over the last couple of years. We've seen the investment payoff. In absolute terms, we're spending more online year on year. But, because the business is growing as fast as it is, our marketing investment is growing, and on a percentage basis, we've made a greater investment in T.V.
So I guess I'm the wrong guy to ask to talk about when are television dollars going to move online because we've actually been going the other direction with success. So I think of course there will be a point at which we're pretty saturated but we're a long way from full brand awareness and being able to extend our brand. I mean, we're not in peak television and we're not on broadcast - and we have enthusiasm for that approach.
If there are “data challenged” CMOs out there and they know that they need to get their arms around “big data,” are there some tips that you could give? How do you get your arms around the data if you're a CMO as the data continues to propagate and influence decisions?
To me, it starts with understanding your KPIs. For Ancestry.com, it’s very easy to understand our KPIs. We’re a public company measured quarter to quarter on subscribers added, subscriber acquisition cost, churn and revenue per unit.
But, beneath those KPIs are 400 more - from brand awareness to conversion metrics and retention and engagement. Nothing focuses the mind like understanding what you're being judged on and instilling it in all of your team. There is a lot of numerical focus for the entire team. And there is no ability for someone to be judged on an expensive, pretty, Super Bowl ad that makes everyone feel good. It has to come back to the numbers.
Also, you need to start creating dashboards that reinforce your marketing activity. We’re over the top sometimes where we literally look at a ticker minute by minute while a television buy is running. We can see then when our product placement happened as we're watching sign‑ups as they come. And it's helpful even on the QA side. If there's a sudden dip, we know there's a site issue, for example.
There are many metrics to track and understand – you just need to know which ones are the key ones.
Well, it's not a tip, really. But, I throw resumes in the bin that don't have a lot of numbers and percentages on them. I don't know how people would pitch themselves as a data driven marketer and not talk about their successes. So, that's one thing that I look for - someone is numerically focused. And, I take the interviews process very seriously and ask analytical questions around complex problems - management consulting type questions.
By John Ebbert