Join thousands of weekly readers and receive practical marketing advice for FREE.
MarketingSherpa's Case Studies, New Research Data, How-tos, Interviews and Articles

Enter your email below to join thousands of marketers and get FREE weekly newsletters with practical Case Studies, research and training, as well as MarketingSherpa updates and promotions.


Please refer to our Privacy Policy and About Us page for contact details.

No thanks, take me to MarketingSherpa

First Name:
Last Name:
Jan 27, 2004
Case Study

Gift Site Tweaks Design to Raise Sales Conversions from 4.56% to 7.16%

SUMMARY: If your e-retail site shows more than two pages of results for any search or navigational link, watch out! According to the data from this new Case Study, only 75% of online shoppers click to results page two, and a tiny fraction of that bother with page three.

Also includes shopping cart abandonment reduction tactics, plus a creative sample of a "pop-after" coupon that raises sales 20%.
Oh the pain of a too-successful business. During the holiday 2001-season, had such a surge of shoppers that the site crashed repeatedly.

"There were whole days when the site was down," remembers Geoffrey Smith, President of ecommerce & New Business Development.

While Smith knew the site badly needed a tech upgrade to scale for the next holiday season's masses, his secret dream was to tag a series of sales conversion improvement tweaks onto the project. "It's not about growing traffic so much as let's start getting nitty-gritty about optimizing."

First he had to convince upper management to fund the extra project, and choose a limited number of tweaks that would give the biggest bang for the buck.


Smith started by creating a thorough study of the project, including:

o A clickstream study of the past 12 month's traffic showing how visitors were using the site from the home page through the search and check-out process.

o Analysis of what competing sites were doing.

o Price quotes from (and analysis of) several leading vendors for the site upgrade project, including an advanced analytics package to help with future tweaks and tracking.

o Projected ROI based on a moderate conversion improvement, adding .3-.5% on top of the then-current 4.56% rate.

He boiled all of this data down into a presentation for the CEO, with about 30 PowerPoint slides on the highlights, and got the go-ahead in January 2002. After the initial tech upgrades, the design team started tweaking the site in the summer of 2002, and haven't stopped since. Smith pinpointed five specific areas for improvement:

Tweak #1. Home page redesign

Initially was copying most eretailers' home page design, with hero shots and brief sales copy for a group of featured products. But, click path analysis revealed that consumers weren't clicking through the featured items on their way to purchasing something.

Why? Smith suddenly realized that expecting people to buy one of a handful of featured items went against the nature of general gift retail. He explains, "In our business it's a breadth issue, not a depth issue. We have over 1,000 items on the site and no one item represents more than 1-2% of our sales."

So, the very nature of the home page needed to change -- from pitching a few spotlight items to pitching the idea of buying from this particular site versus all the other gift sites out there. "The home page needed to give consumers an idea of who was and what was inside the site."

Then, sold on the idea and brand itself, consumers would hopefully convert at a higher rate when they found that perfect niche gift item deeper inside.

Tweak #2. Thumbnail page improvements

Smith's team had another revelation when they reviewed clicks on thumbnail pages -- pages presenting a grouping of nine small graphics of items in a category. Turns out, just like search engine results, consumers almost never click beyond the first page or two of results.

"75% would go on to the second page, but after that there was a huge fall-off. It dropped to almost nothing for pages three and beyond. And we had some categories that went out 10 pages!"

The designers made three critical changes to cope with the problem:

o Adding more thumbnails per page -- they raised the number from nine to 12, and made the thumbnails slightly bigger with a bit more descriptive copy, including price.

o Moving bestsellers to the first page -- originally the content management system automatically showed products in the order they'd been added to the site. So, newest items were on the very last page. Now the products are organized in order of unit sales plus specific decisions from Smith's team.

o Adding more categories -- Smith figured that since no one wanted to click beyond page two, why bother having a category with lots of pages? So the team began to create sub and sub-sub categories for consumers to surf.

Tweak #3. Left navigation additions

With all the new categories and the lack of specific item promotions on the home page, the navigation bar had to be vastly upgraded so people could find items easily.

Smith's team made the bar very clean and text-based. So as not to be overwhelming, the sub-sub categories are only visible when a visitor has clicked deep enough into the site (or arrived via deep-linking ads) for the very niche categories to be applicable to the particular search.

Tweak #4. Smoothing the check-out process

At 45%, actually had a lower-than-average shopping cart abandonment rate at the time. However, Smith was convinced tweaks could improve it.

First the design team eliminated an entire screen by merging the shipping and billing pages together. Then they tested which should go first on the page - shipping or billing?

Tweak #5. Testing a pop-after

Last but not least, over the Holiday 2003-period, the team tested adding a pop-up to the site that would only appear when a visitor was leaving without purchasing anything. The pop-up offered $5 off any purchase - as long as the consumer clicked back to make the purchase immediately. It's impossible to save to use later.

If the consumer had something in their cart, the return click took them directly back into the cart where they'd see the $5 discount visibly sitting there waiting for them. Otherwise the return click took them to the home page.

Smith asked the design team to give him the ability to disable the pop-after for visitors based on what campaign link they came from. He wanted to test what traffic-sources it worked best with.

Plus, he definitely didn’t want to use it for outbound email campaigns. "People may get our email at work, browse, and then save the email to buy later until they get home that night."

RESULTS's average visitor-to-buyer conversion rate over the 2003 holiday season was 6.28-7.16% with almost 80% of purchases made by first-time buyers (who are much harder to convert than previous customers.)

The site's check out process tweaks reduced cart abandonment rates from 45% to 33%. Asking for the shipping address first produced more conversions than billing address first.

The pop-after was also a resounding success, with a 20% average improvement on sales without it. Smith notes he likes to use it with search marketing the best. "We really tested it with search, turned it off, turned it on, turned it off again. We got more orders with it." He plans to leave the pop-after program in place year-round, with additional testing and tweaking of course.

The site's annual sales are currently around $30 million, from 4.5 million visitors.

One final note -- the site gets extraordinarily low returns, less than .5%. This may be partially due to the nature of gift sites. But, we suspect it's also due to the fact that the design includes a checkbox than shoppers must actively check to confirm the spelling of the personalization they've requested with their order.

Why should you care? Well, consumers are not known for fabulous typing, and you can bet that a personalized gift with a typo would be returned pronto. So this checkbox must be helping avoid the problem.

We think it might be an idea that sites trying to get accurate email addresses should test instead of forcing consumers to type the entire email twice (which is not only annoying, but also lends itself to cutting and pasting with ensuing mistakes.)

Just a thought…

Useful links related to this story:

Screenshots of tweaked pages and pop-after:

WebTrends - the analytics program relies on to track tweak results:

Fry, Inc - Site design and tech operations back-end for
See Also:

Post a Comment

Note: Comments are lightly moderated. We post all comments without editing as long as they
(a) relate to the topic at hand,
(b) do not contain offensive content, and
(c) are not overt sales pitches for your company's own products/services.

To help us prevent spam, please type the numbers
(including dashes) you see in the image below.*

Invalid entry - please re-enter

*Please Note: Your comment will not appear immediately --
article comments are approved by a moderator.

Improve Your Marketing

Join our thousands of weekly Case Study readers.

Enter your email below to receive MarketingSherpa news, updates, and promotions:

Note: Already a subscriber? Want to add a subscription?
Click Here to Manage Subscriptions

Best of the Week:
Marketing case studies and research

Chart Of The Week

B2B Marketing

Consumer Marketing

Email marketing

Inbound Marketing

SherpaStore Alerts


We value your privacy and will not rent or sell your email address. Visit our About Us page for contact details.