Finding the Consumer Pathway
What do you track and how do you make it actionable? We're deluged with data, what do we do with it?
Moderator: Greg Smith, COO, Neo@ogilvy
Panelists: Alexandre Douzet, Co-Founder & VP of Marketing, TheLadders.com
Dion Sullivan, e-Commerce Marketing, Atom Entertainment/Shockwave.com
Matthew Roche, Founder & CEO, Offermatica
David Herscott, President, MEA Digital
What do you track and how do you make it actionable? We're deluged with data, what do we do with it? The panelists started with some presentations to illustrate what the problem is.
Alexandre Douzet of TheLadders.com sees the need to "filter" and use a 2-step process: free and premium. There's a lot of optimization to segment active vs. less active candidates. TheLadders.com goes through multivariate testing to see what works, using A/B testing. They've integrated keywords from Google Ads into the pages. In one example shown, they got 2.6 percent vs 3.48 percent conversion, a 32 percent lift. The impact is tremendous. They are also implementing "Persuasion Architecture", a more user centric approach to optimization, segmentation of user base using Personas, with nonlinear targeted "funnels" to bring people in/
Shockwave works to decrease customer churn and increase subscriber loyalty. They measure and respond to customer behavior via incremental improvements and product innovations. Optimizing the conversion funnel: acquire, convert, retain. First you drive traffic via SEM, SEO, affiliate marketing, behavioral and feed. Next, convert web site flow and conversion via website optimization, usability testing message testing, community. Then retain via communicy and eCRM. You need to do a good job of explaining "what it is" on the home page, then provide clear pricing information and allow the consumer to understand what he or she is buying andd committing to. Key performance indicators drive incremental improvements. On site levers include referring URLS, unique visits, conversions, trial churn, login frequency, paid subscribers, avg. lifetime value, winbacks. Off site levers include getting customer feedback and competitive analysis. A key thing is that Shockwave has a free 10 day trial. If action isn't taken within 72 hours, action drops off. You also need a research strategy Determine research objectives, determine research methodology, then build research tools. The test plan is to develop alternative sub models, all with different characteristics and price points to learn which most gamers prefer, identify which entices gamers to sign up for free trial, identify which model increases long term value.
The funnel is fairly tight. The goal is to get the visitor to the end of the funnel. As the funnel gets longer, the ad needs to hold the visitor all the way through. A real funnel has many steps. Long term progress will come from improving the relevance and engagement of the online experience. You do this by intelligently changing content to suit the source of a visitor or their behavior. ESPN understands the sports fan: ads on Google Mail have completely relevant content. Engagement unlocks new value. Kayak.com keeps somebody doing something they want them to. A poor example was cited: engagement using Flash for a minigolf game won't get people to lock in a mortgate rate.
High-impact segmentation uses 4 sources: source, keyword, action, geotargeting. Good place to start and anybody can get started with these. The best way to start is to use big slices.
Why do "gut" when you can go with objective information? You need to do quantitative testing wherever possible.
The panelists were asked about secrets which were surprising. Here are some of their answers...
The shortest form doesn't always work. Let the net decide: I don't know better than the consumer. People ignore the fact that overanalysis can lead to paralysis--it's not how much you have but how effectively you use it and take action steps forward. Douzet said he was surprised how long a lead can last-- TheLadders leads can last 18-24 months. Obviously, it's a different segment than most.
Look at what engages people on the site. What's the "give to get" ratio? It's amazing how many clients focus on buying traffic but don't look at what they're doing with it. Priorities also change over time. At the beginning TheLadders focused on traffic, now it needs to focus on conversion. If you can take the guesswork out of the equation, you can do a lot.
When you speak about consumer insight, focus on the ultimate objective. What sort of relationship do you want to develop? Herscott is not a believer in focus groups, but it's important to get insight. David Herscott of MEA Digital did a test for a client and showed many different versions. The least likely version won. Putting a Verisign trust icon on the page is virtually a guarantee of conversion. For BtoB clients, provide white papers as an added value.
Look at your direct competitors. Also look at other sites you admire. TheLadders benchmarked vs. dating sites: Match, eHarmony, etc. Why the colors, what's the call to action? Dion Sullivan spends a lot of time researching sites, content and how they market themselves. Zappos.com is selling the Nordstroms customer service experience online. It's not B2B vs. B2C. With a lot of traffic, you can do multivariate testing. If you're getting 30-40K visits per month, start with A/B splits.
The panelists all agreed: you can't predict, you need to test as much as possible.
