An Expensive E-Commerce Game: Product Page Roulette
Like so many online shoppers, we sometimes have trouble making buying decisions here at Conversion Sciences. However, most online shoppers don’t have access to the special tools we have here in the lab. This is a story about how we used our optimization tools to tell us which website we should buy from. There are […]
Like so many online shoppers, we sometimes have trouble making buying decisions here at Conversion Sciences. However, most online shoppers don’t have access to the special tools we have here in the lab. This is a story about how we used our optimization tools to tell us which website we should buy from.
There are some important lessons here for you if you work on an e-commerce site.
Roulette is a game of chance. Your product pages shouldn’t be a game of chance. Product Page Roulette is a geeky game we invented to tell us which sites are gambling and which are changing the odds by testing.
Let me give you some background. We are the Crime Scene Investigators of e-commerce websites. We have access to a variety of cool decision-making tools in our lab. As in an episode of CSI, these tools help us find evidence of visitor dissatisfaction, extract the key insights, and test for opportunities to increase revenue.
Our newest Conversion-Scientist-in-Training prefers to work at a standing desk, and this sounded like something we all should consider doing. (We need to fit into our lab coats.) So, she went on the hunt for standing desks.
She summarized her findings in a to-the-point email. That’s the way we roll here.
The links took us to these product pages:
|These are the product pages of the desks we liked best. Click to enlarge.|
Now, none of us in the conversion lab are experts in the selection of stand up desks, and so we found ourselves with a paralysis of choice.
Finance liked Option 1 for obvious reasons. It had the lowest price.
HR liked Option 2 because of its flexibility.
Design liked Option 3 because it was “kind of Steam Punk.”
The developers were happy with their chairs and ignored the discussion.
Choice is often an e-commerce conversion killer. When the marketplace offers a spectrum of prices and a spectrum of features for a product, visitors (and scientists) can become paralyzed. This is where the less logical aspects of the purchase decision will gain influence. The site that is easiest to understand, that presents key information most clearly, or that simply feels more friendly, will win the business.
Instead of giving in to our emotions, we decided to collect some data. We wanted to know which site was most likely to be picked by a typical human. We clearly aren’t typical humans.
To estimate which page is most likely to be a visitors’ choice, we used one of the tools we keep in our digital lab, an eye-tracking simulator.
Eye-tracking studies use special hardware and software to track where on a page their test subjects are looking. The process is time-consuming, and the equipment can be expensive.
Eye-tracking simulators estimate what an eye-tracking study would tell us without all of the time and expense. These tools look for features that tend to get the attention of the human eye. Contrast, images and white space are just some of the things taken into consideration.
We had some credits due to expire at Feng-GUI (are these credits really perishable?!). We decided that the page that Feng-GUI predicted would perform the best would be the winner of our dollars. Your visitors are voting on your website with their dollars, and so will we.
Product Page Roulette was invented.
Product Page Roulette
Product pages are the pages that generate a key e-commerce action: Add to Cart. Clearly, product pages are very important to the success of such sites.
We captured the top 687 pixels of each product page and put the images through the eye-tracking simulator. We chose 687 pixels because this is the amount of the page that a typical visitor might see in their browser. Some will see less, others will see more.
Here are the results:
|Eye-tracking simulators tell us which parts of a page get the visitors’ attention in the first few seconds of a visit.
Click to enlarge.
It looks like there are storms moving across these pages, but the red, yellow and green areas are estimating attention, not rain. The red and yellow areas show where a visitor’s eyes would spend the most time in the first few seconds of their visit. Green defines lighter areas of attention. Parts of the page with no color are probably not being seen in these crucial early seconds.
This doesn’t tell us who the winner is. However, when we combine these reports with what we know about effective product pages, we can discern our winner.
What To Look For In A Product Page
The job of the product page is to provide everything the visitor needs to make the decision to “Add to Cart.”
The page should draw their eye to the product image.
|Product images work best when they draw the eye.|
Images are usually magnetic to the eye, and all three got some attention to the image. National Business Furniture made good use of badging (“Take 10% off”). However, the “Watch Video” logo on the site competed with their product image. It had better be a good video.
Pricing & Relative Pricing
The page needs to provide pricing clearly. If the product is discounted, it should be clear.
|The pricing block of a product page gives the details of the purchase.|
None of the pages got much attention paid to their price blocks. National and Rakuten included a crossed-out list price to highlight their discounted price. World Market goes for simplicity.
Shipping, Tax & Specials For This Product
The pages must answer questions of final cost. When product pages don’t, they generate high abandonment rates in the cart. People will add something to the cart and start the checkout process just to find out what the final bill will be.
|The details of shipping and specials should be addressed before the cart.|
Rakuten got a great deal of attention to their Free Shipping icon. It’s done in a unique color. However, above that it says, “12 new from $607.92.” So, what’s the real price?
Attention fell on World Market’s “Save 25% on all furniture” banner. They disclose their $35 delivery surcharge right next to the Add to Cart button. This may be a negative for some visitors, but this is probably reducing their cart abandonment rates. If something is material to the purchase, even if it’s “bad news,” don’t hide it.
Trust & Proof
The page should communicate that the business is safe to buy from. Logos, guarantees, return policies, ratings and reviews communicate safety.
|Credibility comes from your brand, from other brands and from the opinion of others.|
National nails it with bright-red stars and a link to reviews. However, the red-on-tan stars clearly don’t draw much attention. Rakuten does little to highlight their return policy, so it is probably missed by visitors.
World Market plays on the brand recognition they get from their retail stores to establish trust. Their logo is the only red thing on the page and clearly draws attention of visitors. Do you have a competitor with a familiar brand? It is an advantage.
The Add To Cart Button
The call-to-action plants the idea in the visitor’s mind that they should do something. Thus, pages with high conversion rates make this one of the most visible items on their pages.
|Visitors should never have to search for the Add to Cart button.|
The Add to Cart button on the World Market product page sucked attention to it. It’s raining hard on that button. Their use of white space made their call-to-action button jump off the page.
Rakuten places their Add to Cart button on a gray background, ensuring it fades into the digital woodwork. National buries the button at the bottom of the page (and below the fold for many viewers).
And The Winner Is….
In our analysis, World Market did the best job of getting attention to the key parts of their product page: the product image, trust symbols, and the all-important Add to Cart button. The placement of a sale banner in the middle of the page is unusual, in our experience, but seems to be working for them in the attention department.
Most of the attention on National’s page went to a button featuring a product video and the image of a related product. Video may indeed be an advantage for them, but they missed on almost all of the fundamentals.
Rakuten may have scored better if they didn’t have such a distracting left sidebar. Their product image definitely gets noticed, but their call to action does not.
Why World Market Is Winning
With a little detective work (which is what we do) we found a clue to World Market’s apparent success.
They optimize their pages for conversion.
(You really should go check on your competitors’ sites with Ghostery. Like, now.)
Crazy Egg tracks where people are clicking on any page, and also reports how far visitors are scrolling down a page.
Optimizely is a split testing tool. It allows World Market to try out changes on visitors and see which changes generate the most revenue.
If we assume World Market is actively using these tools, we might infer that the elements of their winning page were identified during their optimization efforts. We would also expect them to have a higher conversion rate and to be earning more revenue-per-visitor than their competitors.
What’s on your website?
A Word About Eye-Tracking Simulators
In fairness, the eye-tracking simulator we used simulates what your visitors find in the first few seconds of their visit. While it has proven to be a valid measure of page effectiveness, it does not predict conversion rate.
It is entirely possible that World Market has the lowest revenue-per-visit (RPV) of all of these choices. If that’s true, it won’t be for long if they are using their testing software.
If you’re not testing your e-commerce product pages, you’re playing Product Page Roulette with your competitors. It can be an expensive game. There are tools available that can predict how visitors will behave when they visit your e-commerce product pages. If you can win at the game, your e-commerce business will win the revenue game.
Opinions expressed in this article are those of the guest author and not necessarily MarTech. Staff authors are listed here.
New on MarTech