I'm a partner in the advanced analytics group at Bain & Company, the global management consulting firm. My primary focus is on marketing analytics (bio). I've been writing here (views my own) about marketing, technology, e-business, and analytics since 2003 (blog name explained).

Email or follow me:


« Google Buzz: Right On Schedule | Main | Filtering The Collective Preconscious: Darwin Ecosystem »

February 09, 2010

Beyond #A/B and #MVT #Testing: Optimizing Experiences

I caught up recently with Jeff Eckman and Geordie McClelland of BGC to learn more about their service for deploying, testing, and optimizing direct-response page flows.  The easy-to-use service (I saw a live demo), which is based on ion Interactive Pres & CTO Scott Brinker's LiveBall platform, combines a template-based authoring tool, CMS platform, and testing engine.  A marketing organization creates a subdomain on its web domain, and then works with BGC to develop different landing "experiences" (sequences of pages arrayed along a logical path) on it, instead of a single landing page or static, product-focused microsites.  Then the system tracks conversion rates along and through these experiences and, with full transparency to the marketers, shifts a greater share of requests to the winner.  Meanwhile, the marketers develop new page elements, pages, and paths based on what the more granular feedback about those suggests might make sense.

Contrary to popular wisdom about how fewer clicks drive increased conversion rates, BGC has found that an experience that asks for an "optimum" amount of information in an "optimum" series of screens delivers far higher results: on average 300% higher and as much as 3000+% higher in the 3+ years this practice has been in use.  I asked if their experience across multiple clients had yet suggested an overall rule of thumb, such as, "For products in this price range, three pages and three questions to each page work best", to help jump-start the creative process and save on initial creative costs, but Jeff and Geordie hadn't seen one emerge yet, and were somewhat skeptical that it could / would be worth it to pursue.  I struggled with that answer for a bit, but it's occurred to me that maybe I need to break out of my heuristic-seeking box and accept that in 2010 and beyond, the Process is the Answer (see my earlier post "Personalization Is A Process").  Jeff and Geordie are putting their money where their mouths are on this point:  The service can be offered on an easy-to-try-and-buy CPA basis (after a very modest setup fee).

We discussed technical directions that could broaden the span of the experiences they can engineer and test.  For example, you might imagine that an experience might start within a Facebook application for a couple of screens, and then make the jump to your site for the rest of the process and, hopefully, a conversion.  While they haven't done one of these yet, since a url is a url and its contents can be rendered ecumenically, this should be straightforward.

I haven't seen all the direct and indirect competition yet, but BGC's solution seems fairly unique and easy to use.  In theory you could cobble a GA-based solution together that combines their funnel analysis with their Web Optimizer A/B testing tool, but that would be clunky and likely brittle, perhaps a false economy.  And, not too long ago I wrote about Sitespect's url tunneling capability, but it strikes me that while somewhat similar, the two are currently focused on different parts of the "attract-engage-convert" process.

Bottom line: have a look, especially if you use form-based landing pages to qualify and convert leads for higher-price-point products and services.


TrackBack URL for this entry:

Listed below are links to weblogs that reference Beyond #A/B and #MVT #Testing: Optimizing Experiences:


The comments to this entry are closed.