Multivariate testing is a data-driven approach to online marketing design and execution. Where we once had to rely on speculation in terms of what works, multivariate testing allows you to rely on data generated by site visitors in improving your campaigns. You can drive visitors to convert by simply testing and tweaking layout, copy, imagery, and calls-to-action within your campaigns.
Why Bother with Multivariate Testing?
Many may be skeptical of putting effort into something like Multivariate testing when they might not see results they have envisioned. If you’re not convinced of the benefits multivariate testing holds, could you use any of the following?
- Lower bounce rates
- More visitor engagement on your site
- Higher amount of requests for consultations or newsletter signups
- Increase in sales
While some businesses may not see the inherit benefits of a multivariate testing phase, competitive markets such as insurance, seo, and home loans have seen it almost imperative to their online marketing strategy. Some simple tests could essentially double your revenue as a result of increased conversions.
Simplifying the Process
Of course, you’re not going to start multivariate testing all the sections of your website. It’s best to start with a few key elements that are contingent upon your conversion goal. This not only focuses your attention on the key players but also ensures the traffic levels you need to attain accurate readings is reasonable.
The following parts of a page can be included in a multivariate test:
- Headline and heading,
- Call-to-action buttons (color, text, size, placement),
- Text copy (content, length, size),
- Image (type, placement, size),
- Form length.
Remember not all sections of a website have equal impact on conversion rate. If you include a call-to-action, image, and footer, you may realize one of those doesn’t have as much impact as the rest, which produce winning combinations.
Types of Multivariate Tests
There are three types of multivariate tests to look at:
Full Factorial Testing
In this method, website traffic is distributed equally among all combinations. If there are 12 combinations, each one will receive one-twelfth of all the website traffic. Since all elements receive the same amount of traffic, there’s no discrepancy as to what section performed the best. This method is what generally comes to mind when you think of multivariate testing.
Partial or Fractional Factorial Testing
When it comes to “partial” or fractional factorial testing, only a portion of the possible element combinations see website traffic. For example, if there are 16 combinations, then traffic is split among only eight of those. Only a short length of time is necessary to get accurate results. If you don’t have the time to run a full-factorial test, this option may be worth pursuing.
Originally designed as a method to improve manufacturing quality, the Taguchi method is one of the lesser known forms of multivariate testing. In it, specific assumptions are made in order to decrease the number of combinations needing to be tested. This potentially could help save quite a bit of time while running tests.
The great thing about internet marketing these days is you don’t necessarily have to rely on trial and error to come up with a successful website. If you’re looking to make big advancements to your website, multivariate testing has the advantage of making large changes to multiple elements over multiple pages. It’s the best strategy for large organizations looking to enact rigorous steps around testing.
Note: Photo courtesy of findyoursearch via Flickr under a Creative Commons license.