In Part 2: Understanding Conversion Rate Optimization Data, we learned that as with most decisions in our data-driven digital world, creating an effective strategy for optimizing for conversions starts with data gathering. Getting as much qualitative and quantitative data as you possibly can will help your brand make the right conversion rate optimization decisions.
Once you have all this data, the next step is to synthesize it down and determine how it will inform your conversion optimization strategy. In part three of our six-part Optimizing for Conversions to Create Lifetime Value video series, Margo Andros, Founder at LTV Approach and Brandon Howell, Website Optimization Service Lead at ROI Revolution reveal two top strategies for conversion rate optimization.
https://roirevolution.wistia.com/medias/my98arakp5
Optimizing for Conversions to Create Lifetime Value, Part 3: Creating Your Conversion Rate Optimization Strategy
Before you even get into the strategy piece, there’s a key data gathering element to inform that strategy. We use all of these avenues to build strategies for our clients. Once you have all this data, you can take two directions with it: best practice-based optimizations or testing-based optimizations.
When you do all of this research and look at your data for the first time, you may notice a lot of friction points that need to be fixed. Find them and fix them – they can’t get worse. But what about this vs. that decisions where people have different preferences?
In many of these cases, it’s best to go with your industry’s best practices – especially if you have less traffic to your website that makes it harder to get statistically significant test results.
Testing-backed optimization is always ROI’s preference because it’s real data, not based on assumptions. We like to throw around the phrase “humans are fickle” – you hear that all the time in our office. It’s a gross oversimplification. We see plenty of instances where these “no brainer” optimizations that seem like they’ll definitely be good backfire or don’t work.
“Humans are fickle. You don’t know how your audience will react to something until you test it. And you should test.” – Brandon Howell, Website Optimization Service Lead, ROI Revolution
For example, say you have a fashion ecommerce site and you’re trying to figure out how to get people down your funnel from homepage to checkout. You decide to perform a split test to determine what will be the most effective layout to increase sales from your point of purchase page on mobile. Test A is a one-column layout featuring large, eye-catching images. Test B is a two-column layout featuring smaller images, but more product selection can be shown without the user scrolling as much.
With shoppers having so many different personal preferences, how can you determine which layout should bring more sales? Additionally, different people who work with different channels (e.g. paid search analyst, ecommerce director, UI/UX designer) have different ideas about what’s best. The only way to know is to test your audience.
We ran this test on two separate fashion clients and they both had drastically different results because of their different audiences. For Client A, the single-column layout test was statistically significant and increased their sales. For Client B, the two-column layout test was statistically significant and increased their sales.
Similar vertical, different results, different audience.
Lastly, you should use an A/B testing tool – you may have heard of AB Tasty, Google Optimize, or VWO. There’s also personalization platforms and software out there like Emarsys or Monetate that have some A/B testing capabilities built into them already that you may be able to capitalize on to do a true A/B test that provides statistically rigorous data.
