The best CRO tactics don’t overcomplicate the experimentation process and make it easy to track success. In this post, I’ll highlight four areas we have found to be rich with CRO opportunities that are easy to get started with, analyze what works, and grow from there. Let’s start with where the digital experience has always started: the browser.
The one-day exercise to improve conversion rate +7%
Load speed and user interaction responsiveness are ground zero for CRO. Why? Because for every second delay in load time, you’re losing 7% in conversions. For example, a site doing $100,000 per day in sales would see $7,000 in net new sales for every second shaved off of the site load time.
If you can find low-hanging fruit that shaves off just one second of load time, you’ll be rewarded immediately with ROI for the easiest improvements. Google Pagespeed Insights is an excellent place to start, and an easy way to get a numeric score to gauge where you are today, where you need to be, and to measure progress.
Don’t get overwhelmed: The report is going to spit out a massive wall of data but you only need to take it one step at a time. Here’s a one-day exercise to get to the important actions.
- Start by breaking up all the findings of the report up into chunks. Gather your development team and during an hour session, go through each task and determine if it’s not applicable (for a false positive), low effort (less than an hour), medium effort (less than a day), or high effort (more than a day).
- Go back through the list a second time but this time for any task with an effort estimate also indicate if the change will be low impact, medium impact, or highly impactful in terms of improving end user experience.
- End the meeting. You now have a great list of specific tasks you can, one at a time, to improve performance and thus conversion.
Because site designs are unique and there are many technology stacks in use your performance recommendations are going to vary. However, quick wins will often be found with deferring CSS, image format optimizations, and pre-connecting to third party origins. Medium effort wins are going to involve JavaScript optimizations like minimizing main-thread work and reducing execution time, avoiding excessive DOM size, and serving static asset with an efficient cache policy.
If you want to go a step further and truly get to ludicrous speed, go headless with our guide.
Upsell all the things and experiment with your promotions
Of all the trends mentioned in our post, upsells are the most productive CRO tactic we are seeing across all of our clients. Specifically, we are seeing great performance around post-purchase upsells. Essentially you complete the checkout/order and show the customer the order confirmation page but also you keep their credit card order open so that they can easily add a few products you suggest on this page to their order with just one click.
The key to success here is the reduction of friction (no other work beyond one click) and of course showing relevant products. This is where your testing comes in. Test the format of the order confirmation page, the specific products suggested, the CTA for buying additional products, and the messaging about how they don’t need to add their payment information again. Test, optimize, experiment, rinse, and repeat.
But don’t forget about promotions!
For a digital merchandiser, promotions are an easy way to get stuck in your habits and assumptions for years. You learn what promotions you run in what season and stick to the plan over and over again. But promotions are a rich area of CRO and can offer big swings in revenue performance.
For example: A prominent client of ours in the cosmetics space increased their average order value nearly 40% when a promotion was changed from one free sample to two free deluxe products with each $50 order. How did they determine how well this modest promotion tweak would perform? They didn’t. They simply tested the damn change — many changes — and found one (and others) that performed better.
The future of checkout is getting rid of checkout
If a Blu-ray could talk and you interviewed one in 2014, I bet you would hear blind optimism about the future of Blu-rays. In 2014, U.S. sales of physical media were over $10 billion. That’s over 300 Blu-rays an hour. But fast-forward to today and streaming is killing physical media, with annual physical sales down over 50%. It’s only going to get worse from here.
What does this have to do with checkout? Well, take a step back and admit an uncomfortable truth: The checkout process is an absolute pain. Creating a new account is a pain, creating a new password and worrying about security and storing it is a pain, entering in your address again is pain. Is your shipping address different than your billing address? Forget about it. The solution? Get rid of checkout. Think I’m crazy? I’m not. When you take mobile shoppers and combine Apple/Android pay you don’t get 1 + 1 = 2, you get 1 + 1 = 10. Mobile embedded payment options, even more than offsite payment methods like Paypal, reduce customer friction so dramatically that customers start to bail on checkout if native mobile payment options aren’t available.
Baynard Institute estimates that 26% of US online shoppers have abandoned an order in the past quarter solely due to a long or complicated checkout process. That is one order out of every four. Reduce friction and you improve conversion. Sometimes inverse relationships can be quite beautiful.
OK, now that that is out of the way, let's look at the good stuff: Rich hunting grounds for CRO.
The first area to consider is testing dynamic checkout buttons. This is pretty simple. Instead of just an “Add to cart” testing how well a generic “Buy now” button performs compared to a branded buy now option.
Here is one simple A/B test idea to try:
Try different variations, measure impact to conversion carefully, rinse, repeat, and refine.
In ecommerce search is not a cry for help; it’s a cry to spend more 💸💸💸
For a SKU dense site, search offers users a quick and easy way to search for products by name. And search matters: According to Econsultancy, visitors who use onsite search are 1.8x more likely to convert and can generate as much as 13-40% of a store's overall revenue.
First and foremost make sure search is simple, front, and center. See how well Kylie Cosmetics does it:
Search offers many opportunities for discrete tests. Some ideas:
- Try different personalization plugins to see if relevancy can be improved
- Try instant search results as you type
- Try promoting featured products at the top of results
- Try optimizing search on mobile
- Try different styling and positions for the search button
- Try different facets and filters on results
- Try swapping in different search engine plugins to see if one performs better
- Try improving search results speed
- Go to your top three competitors and search for real products and buy them (at least all the way to order confirmation) - what worked well, what doesn’t?
- What is different about competitors' search that, abandoning all of your assumptions, you should try and validate with data if it’s better or worse?
Make an educated guess of where to start and relentlessly track, analyze, and let the data guide you. (But make sure you 100% trust your data first). Balancing patience and timeliness in your experimentation will lead to that double-digit growth in 2020.