The Maxymiser blog has moved!

You should be automatically redirected in 6 seconds. If not, visit
and update your bookmarks.

Monday, June 02, 2008

Is Your Uplift Sustained?

As our clients and readers of this blog will know by now, multivariate testing offers a proven way to discover the most effective content online and generate more revenue through websites. There are always easy wins to be had through testing, however, there is also longer term sustained uplift to be gained through careful campaign design and planning.

Achieving sustained uplift should be considered the ultimate prize in any optimisation campaign. A 5% uplift in conversion rate which impacts the bottom line for the long term could deliver significantly more value than 30% uplift on the response to an element of seasonal content.

In our experience uplift resulting from multivariate testing generally falls into one of three categories:

Temporal Uplift - where winning content includes promotions or time-sensitive elements. Examples include limited special offers where the value of uplift is realised during the period of the offer.

Extrapolated Uplift - where form, structure and information architecture in one area of a site are tested and the winning content elements can be applied across all pages employing the same template. In this case, a small percentage in uplift can have a dramatic effect on overall business performance

Sustained Uplift - where ongoing benefit is produced by implementing winning content. Tests which include structural improvements to high traffic pages and multi-stage processes such as check-outs typically provide sustained uplift.

When deploying multivariate testing, as much consideration should be put into the expertise in campaign design that you are able to leverage from the provider as is put into the selection of the tool itself. Temporal uplift can return some jaw-dropping numbers in a short time period but it is the Extrapolated and Sustained uplift that will provide the long term impact on your site performance.

Mark Simpson

No comments: