Multivariate testing allows the comparison of combinations of variables.
- Multivariate testing is introduced to clients once the optimal layout and design have been achieved using A/B Testing.
- All elements of a page are inter-related; calls to action, headline positions, use of images, colors, text content etc.
- The objective of multivariate testing is to identify the combination of elements that works best together.
- Each and every additional variable multiplies the number of pages that need testing. For example, testing a mere three different elements of your website with three different alternatives requires 99 page varieties.
What should we test?
All of our clients are looking for results. Broadly, they fall into two categories; they need more sales (e-commerce E-Commerce: Products sold and sales made via the internet) or they need more leads.
For e-commerce websites we might test:
- Promotions: test different mechanics or discount levels.
- Product images: test different sizes, styling or screen resolutions.
- Product descriptions: test length and persuasive style of copy.
- Calls to action: test different wording and graphical treatment. 'Buy now', 'Add to basket' or 'Next' may each have varying levels of success.
For lead-generating websites we might test:
- Quote request forms: testing the level of information required.
- Email newsletter registration: testing the use of 'collateral' like sign ups to white papers.
- Business service descriptions: test longer, shorter and different versions.
- Typefaces: test different fonts and point sizes.