We asked 20 of the world’s top conversion rate optimization experts to share their top tip for making your eCommerce website convert. The responses were amazing and we assembled them into the SlideShare below. Their advice includes topics related to A/B testing, usability, metrics, cart abandonment, design philosophy, customer experience & more.
Our 20 Conversion Rate Experts:
The Thank You/Receipt page is one of the most underutilized pages. Your customer is at their most excited right when they purchase a product, take advantage of this. Another major perk for testing Thank You/Receipt pages is you are not adding extra friction to the cart. Checking out is one of the most anxiety-inducing processes a customer goes through. Yes, cross-sells and up-sells in a cart work well, but they can become a distraction. Get the purchase, then push for more with dynamically produced content that is relevant to the purchased product.
One of your biggest conversion leaks as an ecommerce site is going to be your cart. People will abandon it. So when you get these users to come back to your site through remarketing and email marketing, consider redirecting them back to their cart. From what I’ve tested, this simple trick can typically provide a double-digit conversion lift.
Focus on the people who will buy. That might seem obvious, but I’m amazed by how many businesses are trying to convert 100% of their traffic. They see that they’ve got massive numbers of visitors coming from Pinterest, they see most of those visitors bounce, and they think, “Crap, we have to do something to get those Pinterest people to stick around!” But, in the meantime, legitimate prospective customers are arriving at their sites in much smaller numbers-let’s say 10% of visitors-but the site owner doesn’t focus on them because “there aren’t that many of them.” Focus first on the people who are likely to buy. Even if it’s a small slice of your traffic pie.
If you’re A/B testing (and you should be!), choose a single metric on which to base the success of your test. Just because you have the ability to choose multiple success metrics doesn’t mean it’s the right thing to do. In fact, it’s common to get mixed signals from multiple metrics across your test variations, and then making a ‘clean’ decision becomes difficult (i.e., “Which metric should I pay attention to?!?”). If you’re in the e-commerce space, Revenue Per Visitor is typically that single metric. It may feel uncomfortable to use just one metric in your tests, but doing so will make your optimization decisions so much easier!
Start optimizing closer to the money. Your actual gains in terms of revenue are going to be way bigger. Get your checkout flow to rock first, then cart page, then the product pages—so you slowly move further from the money. Leave your home page last: 20% up lift on your checkout page vs 20% uplift in your home page: checkout page win is gonna be way more absolute dollars.!
I don’t really have any top “tips” because I don’t think it can be boiled down so simply. What I can say is that I think more businesses should be looking at the overall customer experience, both on a page-by-page basis, but also as a whole experience. This means the journey from PPC click to landing page, this means the trip from landing page to product page, product page to cart, etc. But this also means what happens after? What are the retargeting ads like? What are the reminder emails like? How do all of the pieces fit together to tell a larger narrative that is both personal & rewarding?
We’ve heard a lot about companies like Zappos or Warby Parker, and while it’s easy to look at these companies and boil it down to “good customer service” or “they let you try on a bunch of glasses at home” the truth is, there is no part of the shopping experience with either of these companies that feels like it was rushed together for the sake of building a “funnel.” What they’ve really done is constructed an interactive narrative & their customers are taking part of an experience – like a good book or trip the the amusement park – and they’re not just in the business of merely selling shoes or glasses.
So, if I were to give a “tip” it would be to always be asking, “how can we make this a better experience?” Take nothing for granted, and always strive to make everything better.
Everyone’s doing CRO. Everyone who has a site, is running an ad, or is executing an email campaign is optimizing. All of these activities are done with the SOLE intention of getting someone out there to buy from us. The winners have one advantage: they are using more information for their day-to-day decisions. It’s a considerable advantage. What is this “information” they have access to? It’s talking to lots of customers and noticing patterns. It’s watching lots more customers through analytics. It’s doing a test to see if an idea is good for the business. CRO is really about generating data—information—that tells a business owner what they should be doing. He who has the best data wins. How much data did you generate today?
Test more and test less—it’s not a paradox. Most companies don’t test nearly enough and those that test a lot often test lots of variations. Test for impact first. Isolate the variables that matter by testing. Lots of traffic gets wasted on testing many variations of variables that don’t move the needle. Find out what does move the needle then test for variations.
How do you know what matters? Develop your hypotheses using personas and buyer scenarios. Can you tell a story from your customers’ point of view where the variable seems to truly matter? Then test that! When you get good at this then test a whole lot more. The more smart tests you do the more you’ll learn and the more you’ll improve.
If you’re an ecommerce site bidding on terms like “next day delivery,” please stop pushing me to your generic homepage. Instead send me to a landing page that confirms my search intent (next day delivery), and then—and *only* then—asks me to self-segment by flower/chocolate/apology type. That would be a delightful wind of change.
My top tip is to apply segmentation to your CRO, either before you run the test (exclude/include segments) or after when you analyze. The reason being that geography, referral source, referring keyword, new/returning visitor, device type, etc. can all have different conversion responses to a test, which may be more effective to target differently (depending on the test hypothesis). If you take the average lift and apply it to all segments, you may actually be reducing results in one or more cohorts.
People shopping online have no attention span, and compulsively (now habitually) skim and scan pages for cues to move forward. If you want to get their attention, you need to get in their way, not try to take them where you want them to go. Develop a new persona, call him Ferris the Ferret.
Bryan Eisenberg, The Godfather of CRO | @TheGrok
Scott Cook, founder and Chairman of the Executive Committee of the Board of Intuit, is a former Procter & Gamble veteran with extensive background in traditional product development, marketing and market research. He often shares how he has shifted his company to be more agile to deal with ever changing complexity that today’s increasingly high paced and chaotic marketplace demands.
He reminds executives that “the job of today’s leader is to remove the barriers to experimentation.” Work with your organization’s team to come up with “leap of faith assumptions” and find “the evidence to support those assumptions with tests and experiments.” If someone so strongly rooted in traditional marketing methods can work hard to change their corporate metabolism and culture then there is still hope for every company out there, if they work at it.
Ask questions, listen to your customer and do things that don’t scale. When I started my first affiliate site I was 9 years old and knew nothing about business. When I first started receiving sales I didn’t stop to try to think why it was happening I immediately started emailing my customers and asking them why they chose my product. I didn’t try to automate it, I didn’t take any shortcuts, I wrote each email personally. I learned that day that my site though tailored for women was being visited by men. I immediately changed the look and feel, tone and content to fit my market. Without taking the time to ask questions and talk to my customers I would have never gotten that product/market fit early on.
It’s easy to think that gorgeous design alone can help conversion—which isn’t the case. Design is, of course, an important trust signal but understanding your customers and presenting them with a compelling offer / call to action is much more important. I’ve often made the mistake of thinking a pretty face lift will increase conversion when ultimately it was a clean, no-frills design and compelling message that was needed.
Start with the end in mind and make sure your checkout experience is seamless. Every single visitor on your website has to go through your checkout flow in order to become a paying customer. At this point in the conversion process, every little lift equals more money in the bank.
For eCommerce retailers, a continuous testing strategy is critical for making great website decisions. If you’re just chasing after the so-called “best practices” from blogs, seminars and consultants, you could be making big mistakes that hurt your profit. For example, in testing the product page template for WineExpress.com, we found layout and design changes that lifted revenue per visitor by 41%. In other words, before that test, they were missing that 41% revenue!
Don’t just focus on improving the obvious ecommerce things like your shopping cart and product page, or copy case studies. Take a step back and really figure out your unique value proposition, in particular the reasons why someone should use your site instead of your many competitor’s sites, and then make sure you are clearly promoting this (never just presume your visitors know what it is!) This should be shown on your homepage in the form of short bullet points, and ideally a small version in your header or sidebar, so visitors will see and understand it no matter what page they arrive on.
Another way to ‘think outside of the box’ is to ramp up your efforts to collect visitor emails and send autoresponders to bring them back to purchase. This works very well for service driven websites, but can also work well for ecommerce stores. Give visitors a free incentive like a 10% off first purchase coupon, or a free ebook relating to the niche you are selling products about. Incentives like these provide value to the visitor, and ensures that you don’t have to rely on your website converting them they first time they arrive. You can even experiment with using exit intent pop-ups to show incentives – these are now increasingly common on ecommerce sites and work very well for increasing opt-in rates.
Always accompany “what” with “why.” From headlines to call-to-action (CTA) buttons to analytics reports, people generally care little about the “what” if the “why” remains unclear. Transform your features to benefits, your CTAs to specific value, and your data to actionable insights. “What” paired with “why” naturally paves a story that facilitates decision making for your customer and your business.
Process, not Tactics! Working with hundreds of eCommerce customers on their A/B testing and optimization efforts, I have realised that having a process for increasing conversions gives a sustained competitive advantage. Many companies start with tactical A/B testing where they have certain ideas of their own (say let’s try adding a trust widget on checkout page, changing colors, shifting nav bar, etc.). They test these low hanging fruits in a few months and then they struggle to come up with more ideas. On the other hand, some companies start with a goal to increase conversions and then develop a process around it that involves user survey, analytics research, ideation, prioritisation, and execution. These are the companies that execute new tests every week and are able to optimize conversions effectively.
Karl Blanks, Chairman and Co-Founder at Conversion Rate Experts
Declutter. Not everything on your site will be helping conversions. Identify page elements that might be hurting conversions—or simply wasting visitors’ attention—then test removing them, either with A/B split-tests or multivariate tests. “Decluttering tests” are easy to run because you don’t need to create anything new.