Accounting for variance in audience behavior.
“In CRO, every site has its own unique challenges. Even if it may look the same as another, the audience behavior is different on each thus making them different. Learning about what works for one and not another keeps it interesting as no day is the same and no site really is the same.”
Josh Cuttill, Director of Website Strategy & Optimization
The (un)Common Approach
Nearly 30% of large companies now use conversion rate optimization. However, not all CRO solutions are created equal: a flurry of a/b tests without a strategic testing plan based on user behavior data can be a waste of time and resources. (un)Common CRO follows a rigorous model using the scientific method to test hypotheses that can deliver the greatest revenue from increased conversion rates.
( frequently asked questions )
Ideally, CRO would be involved in the initial planning stages! Collecting user data and feedback and assessing a site’s conversion potential before a redesign gives you a solid grasp on how well your current site is working for your target audience and where it could be improved. This information can have a significant impact on the new site’s design, architecture, and technical functions, because the redesigned site will be optimized for conversions and revenue.
If you are already past this point in the redesign process, all is not lost. Usability tests a on new concept or site can confirm the new version’s potential impact on performance within a contained experimental environment. This helps you avoid unexpected negative impacts from a new design and mitigate potential losses.
No, running conversion optimization tests will not have a negative impact on SEO. In fact, a strong conversion strategy will lead to greater SEO visibility. Safeguards must be in place when running split URL tests so that search engines don’t index your test variations, but our (un)Common CRO team works with our (un)Common SEO experts to ensure that the only impact of conversion optimization on SEO is a positive one.
The number of visits isn’t as important as you might think; it’s only one metric in evaluating statistical significance. The conversion rates for the control and the variations are also important. (Basically, the greater the delta between the two, the faster you will reach statistical significance.) Depending on the specifics of the test hypothesis, conversion totals ranging from 200-500 will generally net a winner. And remember, any test that produces new insights into user behavior is valuable, even if the new variation does not win.
Can new, small, or niche websites do conversion testing? Yes. Technically, you can begin testing with any range of visitors. The difference is in how long you’ll need to wait for significant results. A company with 1-5K monthly visits will need to be patient and run fewer tests, while a company with 50K visitors or more can effectively run a high range of tests across the site. In general, if you have at least 20-30k in monthly visits, you can get CRO testing results in a timely manner.