Close
Join 237,000 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:
Email:
Text HTML
Join Our Research Team at DMA 2014
Oct 05, 2004
Case Study

How Dirt Devil's Site Gets Unusually High Sales (Includes Check Out Process Test Results)

SUMMARY: Sites that serve many masters -- ecommerce, customer service, offline retailers, warranty registration, market research -- usually suck. Too many goals equals too much mess. Consumers get fed up and leave. But somehow Dirt Devil's site pulls it off, achieving a remarkable 5-18% sales conversion rate. Discover how Dirt Devil's online team turned a terrible site into one you should steal ideas from. Plus, get the results of their tests along the way (including what happens when you *stop* requiring phone and email on your cart order form):
CHALLENGE


"When I started, we spent the first several months just working on the Internet strategy, because there wasn't really one," says Michael Crowdes, Manager Interactive Marketing & ecommerce for Royal Appliance Manufacturing - makers of Dirt Devil vacuums.

The team talked with interactive agencies, some of which suggested ideas like launching a regular Dirt Devil cleaning tips email newsletter.

But although one in ten of US households are in the market for a new vacuum at any time, "it's only for about two weeks. Other than that, vacuums are very much like tires. You don't think about them much if you don't need them."

Loads of internal departments had ideas for the site though.

o The warranty department wanted to save on data entry costs by getting consumers to register online. o Retailer relations wanted to send more consumers to brick and mortars.

o TV advertising said they'd be happy to add an URL to ads if they were sure it would convert at least as well as an 800#.

o Market research wondered if they could run surveys on the site to gauge surfers' vacuuming needs and shopping patterns.

o The parts department wondered if they could sell bags and other spare parts online.

o Customer service hoped an online FAQ could save on inbound calling costs, as well as printing and shipping extra owner's manuals to folks who lost them.

Suddenly, Crowdes' team were juggling a zillion site ideas. Could they create a site that pleased everyone?

CAMPAIGN


First, the team agreed to keep the site as simple as possible by creating one overriding goal that would take precedence ruthlessly above all else. "Our strategy ended up being this very simple document. Number one, we want to make it easy for our customers to do business with us."

So, although the site served many behind-the-scenes masters, its design focused 100% on serving the consumer. How?

Tactic #1. Console-style home page leading to more details

Dirt Devil's site has 100s of pages, but the home page is extremely simple. Instead of presenting links to everything under the sun, the page uses what we like to call "console-style" design, to present one single navigation entry to each of the main groups of visitors.

Once they click through the single navigation entryway that's clearly for them, each group finds a wealth of detailed information specifically for their needs.

To please consumers driven online by TV ads, the home page featured hero shots and offers of whatever products' ads were currently in the heaviest rotation. These consumers could click to learn more, compare products, and purchase.

To please consumers doing research online for an offline purchase, the site offered an easy-to-use yet highly sophisticated geocoding engine that sent you to the store that was truly nearest to your longitude and latitude, not just your zip code. ("You may live next to a Wal-Mart that's technically in another zip and we wouldn't want to send you to a Target that's five miles away.")

To please Dirt Devil owners, the site made it super easy to register a warranty, download owner's manual PDFs, or purchase spare parts.

To please random Web surfers and fans who drifted to the site of their own accord, the site offered an ongoing contest to win a free vacuum.

To enter, you just had to answer a few questions from Dirt Devil's market research team. On the next page, you were invited to sign up to get a one-time email ping when the next contest started. (Remember, the team decided against a regular newsletter. So this one-time ping was the only email you'd receive.)

Tactic #2. Make purchasing easier and more trustworthy

With the super-clean new design in place, the team began a series of tests to raise site revenues and really make it perform.

Test #1. Mingling the Cart and Search Results

"Something drives me nuts about most ecommerce sites," says Crowdes. "You put something into your cart, and then find yourself back at the home page. Nobody ever walks into [a brick and mortar] Target, puts something in a cart, and then finds themselves at the front door!"

So, Crowdes decided to merge the shopper's cart and their shopping search results onto one page. (Link to screenshot below.) Now, when a shopper adds something to his or her cart, the site doesn't whisk them away into a separate cart area that's inconvenient to navigate from. Instead, they see their cart details at the top of the page, and their shopping search results underneath.

Customers can check out or keep shopping equally easily.

Test #2. Making Email and Phone Optional

At first Crowdes tested putting "a fairly sizeable paragraph under where we ask for email address, about valuing trust and that emails will only be used for order confirmation."

But then he began to wonder if consumers weren't fed up with being asked for their email addresses by every site. If a consumer doesn't want to get any email -- or extend any trust -- why should they? So, he tested making email completely optional. You don't have to check or uncheck a box, you simply don't have to enter it at all if you don't want to.

At the same time Crowdes decided to test doing the same thing with phone numbers. After all, in the offline direct response world, requiring a phone number on an order form is a proven response dampener.

Test #3. Free Offer Pop-up to Save Abandons

"If you're on the shipping or billing info page of the cart and you close your browser window -- not just hit the back button -- you get spiffed with a free shipping pop-up. It's got little golden retriever puppies which are a well-known image for the Dirt Devil brand." (Link to screenshot below.)



RESULTS


We love Dirt Devil's site experience, and apparently so do consumers. On average, 5% of home page visitors purchase the featured hero shot vacuum, with that number doubling to 10% when related TV ads are in heavy rotation.*

Crowdes notes, "When our TV ads promoted the vacuum mop, 40% of total purchases of the mop occurred online."

The hero shots are not the only ecommerce drivers though. 25% of home page traffic goes immediately to the parts department, and a solid 18% of these visitors wind up purchasing something.

- 38% of consumers choose not to enter their email address in the check out form when it's not required. If you currently require email addresses for no apparently-good reason, stop!

- 58% of consumers choose not to enter their phone number when it's not required. Ditto.

- When Crowdes flipped the switch to stop requiring both email and phone numbers in the check-out form, overall site sales rose by three percentage points. (Wow.)

"My philosophy is, if those two things make you worry so much that you'll consider not placing an order even though we had trust language and an opt-out box, then we should not require them."

- 41% of contest entrants ask to be pinged via email so they can reenter the contest the next month. 30% of these ask to be pinged a second time. (Note: this proves people may not want to enter an ongoing email relationship -- perhaps you should consider offering a short-term email ping instead.)

- 31% of cart abandoners will return to complete their order when they see the pop-up offer with the puppies for free shipping.

- Roughly 10-15% of Dirt Devil's received warranties now come in online, which is a success but not quite as popular as the team had hoped. So, they reduced the size of the warranty link on the home page, and made the parts department link bigger instead. Go with your winner.

* Note: Hero shot conversion data is slightly misleading because home page visitors range from people clicking from contest news sites (these unexpectedly send a great deal of traffic Dirt Devil's way) to TV ad viewers. So, the 5-10% is a mixture of very low and very high-converting traffic.

Useful links related to this article:

Screenshots of the abandon pop-up and other parts of Dirt Devil's site: http://www.marketingsherpa.com/dd/ad.html

WebTrends Enterprise - the metrics software Dirt Devil uses to analyze traffic and shopper patterns to improve results http://www.webtrends.com

RouteMap from ESRI - the geocoding software Dirt Devil uses to direct visitors to offline retailers http://www.esri.com/software/routemapims/index.html

Campaign Enterprise by Arial Software - the email software Dirt Devil uses to send out requested announcements to site visitors http://www.arialsoftware.com/enterprise.htm

Dirt Devil http://www.dirtdevil.com

Royal Appliance Manufacturing Co http://www.royalappliance.com

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.