Connecting Money to Metrics - Guru Interview with Jim Sterne, Founder, eMetrics Marketing Optimization Summit
Did the Old Spice Guy SocMed campaign pay off? What does it take to be a Web Analytics professional? Why don't some people buy when they first visit a given website? How can you track whence they came? What's a good example of video content increasing more revenue?
You'll get answers to these questions and more below in my interview with Jim Sterne, who founded and runs eMetrics Marketing Optimization Summit, the prestigious Web Analytics conference.
Many thanks to Janet Roberts for transcribing and surgically editing this Q&A down to its essence.
Social Media and the Money
Larry Chase: The Old Spice Guy campaign made a huge splash last year. I saw a Nielsen report that there was a sales lift of 107%. But did it make money?
Jim Sterne: The jury is still out on the money because Old Spice ran 2-for-1 offers, the classic direct marketing tactic. Depending on when you read reports and which ones you read, the "actual results" are all over the map.
How Video Content Makes Money
LC: Can you connect the dots between content marketing and the money?
JS: There's a gambling site in the UK that spends a lot of money on video content. Analytics showed that only 6 percent of the site's visitors were looking at this content, and the site owners thought it was a waste of money and were on the point of cutting it.
But, their analysts said the 6 percent that is consuming the content represents 30 percent of their customers, those who are actively bidding, betting and spending money.
The company used that information to put out an email message saying "You're missing something. Come look at our videos." Their profit went up because people were more motivated to consume the product.
You have to be able to determine whether people are actually consuming your content, and, if they are, are they valuable?
The Path to Conversion
LC: You speak of conversion as a journey. What do you mean by that?
JS: With my eMetrics Summit, for example, somebody sees an ad, reads a Tweet, or somehow becomes aware of the summit and comes to the website.
So, I spend money on an ad in LinkedIn, and they click through from there. I don't expect anybody to convert on their first visit.
I judge whether an ad is worthwhile based on engagement. If they come to the home page and bounce away, it's a bad ad. If nobody clicks through on the ad, it's a bad ad, a bad list or a bad daypart.
If they look at the home page and agenda and then look at the speakers and leave, I have a bad feeling about this because they didn't look at the pricing page.
If they look at the home page, the agenda, the pricing and the speakers, or the speakers and the pricing, that's good because they're interested enough to find out the cost.
If they come back later and look more, that's even better because it's more likely that they'll convert eventually.
In between all those site visits, there's a whole lot of stuff going on. They have to check their calendar. They have to get approval from the boss. They have to check with the spouse to get travel authority. Much is happening in the background which is why conversion is so tough to track.
My challenge is to keep track through to the end of the purchase event. Of all those who came and purchased, I have to figure out who came from LinkedIn, Twitter, email ads or wherever.
Why Didn't They Buy?
LC: Why do you need to know where customers fall off the rails?
JS: Abandonment is a point on the path to conversion. If you have seven steps in your path, and they all fall off at Step 5, then you know you have to go back and look at that.
We put up a site for the next eMetrics conference, which has people going from the home page to registration or from the speaker page to registration. Nobody went from the agenda page to registration.
We looked at all the numbers and sliced and diced them. Then I went back to look at the agenda page. Nothing was wrong there, but someone forgot to put the registration button on it.
Making Web Metrics Pay
LC: Out of all the people who are doing Web Metrics and testing, how many are putting a price on each customer or client?
JS: This is why attribution is so important. It's not something people think to do first because it's massively complicated. But in the end, you come up with a chart of profitability for each customer.
Some customers are not profitable because they buy only your loss leaders and overwhelm your customer service. Some are your most profitable customers because they want the convenience, and they don't care about the cost.
Attribution delineates unprofitable customers from profitable ones and helps you figure out how to approach those profitable ones.
Media Budget Optimization
LC: Would you explain your "10%" philosophy: 10% for testing and 10% for measurement?
JS: That's all about allocating media budget. If you're going to spend a million dollars on advertising, why not take 10% and measure how well that's being spent?
Testing can be done two ways. One is the classic A/B split test and multivariate, and optimization is based on that, but the other side is experimentation. You should spend some of your budget on risk-taking.
LC: Calculated risk-taking is probably a separate issue since most wells turn up dry.
JS: ... which is why you need to continue experimenting. You have to drill new wells sometimes.
A lot of people look at changing the call-to-action button from blue to green, or changing the words or changing pictures. Nobody thinks to change the shape of the ad, to change the daypart or make outrageous offers or do publicity stunts.
If you optimize only that one little thing, you'll get only the possibility of that little thing.
Making Money in Web Metrics
LC: I know you can make good money as a Web Metrics analyst. What kind of person succeeds in this field?
JS: Someone who like puzzles, who likes to dig in and figure things out. The applicant you want to hire is the one out in the waiting room who's working on a Sudoku puzzle.
Testing and analytics aren't about tools or data. They're about understanding how to go about solving a problem. You're checking for people's Sherlock Holmes factor.
LC: Do you get a lot of those people from direct response?
JS: Originally, yes, because they were the first people in the company who understood this new concept of analytics and measurement. Now that we've been in it for a decade, there are people who have been in Web Analytics since they got out of school.
I've spent 15 years explaining the web to marketing people, and now I'm explaining marketing to the web people.
About Jim Sterne
Jim Sterne is a Web Strategy consultant, author and founder of eMetrics Marketing Optimization Summits held worldwide. He is the Founding President and current Chairman of the Web Analytics Association.
His latest book is Social Media Metrics: How to Measure and Optimize Your Marketing Investment, published by John Wiley & Sons, Inc.