October 27, 2010, Posted by Chris Kullar
27 oct
Most online marketers are familiar with using A/B testing to test simple variables on a page; for example, two different calls to action or headlines on different landing pages to see which one performs better. In addition to determining which elements on a page provide the best desired outcomes, A/B testing can provide a lot more critical information useful in maximizing site performance and maximizing profits.
Key benefits and split testing recommendations:
- Resolve issues and challenge assumptions with hard data derived from split testing. Data is most effective in persuading peers and management to take a recommended course of action; selling ideas based on subjective opinions is a more difficult endeavor. Let objective data be the key driver in implementing new ideas and driving optimal payback in your online marketing and merchandising efforts. Let the market determine the “best” way to implement something on your site.
- Make continuous testing a key component of your marketing efforts; one shot tests yield some useful information, however continuous testing that is part of a company’s online marketing culture will yield even more dramatic and sustainable results. Continuous testing of site characteristics as well as emails enables better understanding of site visitor behavior. A rigorous and consistently applied series of tests will help you stay on top of the changing tastes and preferences of your key demographics and ensure that your marketing efforts are well tuned and optimal.
- Split testing helps optimize your PPC ads as well as your landing pages. The former can help you maximize click through rates (CTR) while the latter enables you to optimize for conversion. Balancing these two factors can help produce optimal ROI.
- If it is not possible or too difficult to set up a true A/B test consider using a sequential test. This test is one where you show one version of a page for a set period of time, followed by a different version of the page for the subsequent period of time. For example, one can test the impact of varying shipping rates in this manner; this would make it a lot easier for support personnel to track the correct shipping rates compared to an A/B test scenario where customer service might have a more difficult time dealing with this scenario.
- Use split testing to think big. Testing dramatically different versions of pages and calls to action often results in the best and biggest improvements in site performance.
Split testing is a key tool in objectively determining a cost effective path for site improvements, selling ideas internally and ensuring your online marketing efforts are properly focused and result in maximum returns for your marketing investments.
