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:
May 18, 2006
Case Study

Must-Read Results From Online Lead Generation Campaign Tests -- Which Banners, Landing Pages & Forms Work Best

SUMMARY: Whether you're trying to get busy business execs to click on your banner or you'd like (many) more prospects to fill out online forms for webinars or white papers, review this Case Study immediately.

For the past 90 days, one marketer has been testing b-to-b creative versus something completely different. The latter has won tests so often even he is surprised.

Yes, includes before-and-after creative so you can see what worked and what bombed. Share with your designer today. (Note: We rarely label stories "must read" so this is a fairly big deal.)

"I just started this job 90 days ago," says Mikel Chertudi, Director Online Marketing, Omniture.

Just as with new US presidents, in the world of business-to-business marketing, you're hugely judged by the success of your first 100 days "in office." Yes, even if your offering has super-long sales cycles, you probably will be judged (perhaps precipitously) by your 100 days' lead generation results.

Chertudi was under even more pressure than most people because his new employer produced high-end Web analytics. So, he had to prove not only that he could market well but also that he could use analytics to improve results -- immediately.

The goal: to generate loads of qualified new business leads within 100 days via webinar and white paper offerings; plus, to create a set of lead generation marketing "rules" based on tests.


Chertudi began his testing with baseline current creative for each aspect of the lead generation process. In each case, he split traffic and A/B tested his way to learn what worked best. His tests included (link below to samples of all):

Test series #1. Banners advertising webinars

Chertudi ran a series of quick tests for banners appearing on Omniture's home page (which receivedsignificant traffic) and then rolled out everywhere with the winning banner formula. Based on what he'd learned in other studies, he tested the following three elements:

a. Image art -- Chertudi knew that people look at people. So he wanted to spiff up his banners with a photo of a person. But, what would work best -- an image of a generic female executive (a model from clip art) or smaller pinhead headshots of real-life webinar presenters? The former looked more polished, but the latter was more realistic.

b. Short vs. long copy -- Chertudi was pretty darn certain that short (under 50 words) bulleted benefit copy "You'll learn " that described the webinar would work far better than no more copy than "FREE REGISTRATION" in big letters. After all, his prospects were busy, experienced executives whose time was valuable. Plus, there are so many free webinars out there these days, that it's not a big selling point (or is it?)

c. Call to action buttons/hotlinks -- Would hotlinked words work better than a button image? Would a descriptive hotlink "Download the SuccessKit" work better than a shorter, all-caps command "DOWNLOAD NOW"?

Note: after discovering which banners worked best for a particular campaign, Chertudi then quickly matched all creative along a prospective clicker's path to match that creative as much as possible. This meant the headline, colors, typeface, images and even click button of the banner, landing page and even thank-you page (viewed post-registration) matched as completely as possible. (Link to sample below.)

Test series #2. Landing page templates

Chertudi's first landing page template looked an awful lot like 99% of the webinar registration forms on b-to-b marketer's Web sites these days. It looked like a site page -- with the header and top navigation links you'd expect on any page of Omniture's corporate site.

The headline was very brief, simply explaining this was the form to fill out to get whatever offer had been advertised. "Sign up for webinar" or "Download white paper," etc.

Plus, there was a bit of clip art (a stockphoto headshot of a happy executive) and alternative contact info for visitors who wished to phone or email rather than filling out the form.

Next, Chertudi tested each element of that page to discover what would move the needle to convert more visitors to registered users. He tested:

o removing navigation links
o replacing the company site page "header" with a logo
o replacing stock art images with thumbnails directly related to the offer
o removing alternative contact information
o adding more copy detailing the offer
o simplifying layout by reducing the columns from three to two

Test series #3. Registration form fields

Last but not least, Chertudi also ran tests on the actual registration form fields to determine how much and which types of information he could ask prospects for before they hit their form pain threshold and abandoned the form rather than filling it out.


Let's just say it looks like Chertudi's job is safe for the time being. He hit his lead generation goals well before deadline. Key learnings:

Banner tests --
Relevant, real-life images beat stock photos significantly: 34% to be exact. It seems that pretty images are far less important than "true" images are when it comes to getting people to click.

The ultra short, all-caps copy won in both tested positions (much to Chertudi's surprise.) Seems that simplicity is the winner for banners, and even top execs don't mind all-caps commands for an interesting proposition. Last but not least, the buttons won over text-links for banners.

Even for small-seeming creative tweaks, the results could vary hugely. The worst performing action line, a hotlinked line of text reading "-> Download the SuccessKit" gained a 1.46% lead rate. The best performing action line, a button reading in all caps, 'DOWNLOAD NOW' got a 4.17% click rate.

Landing page tests --
Chertudi's test results prove that if you are sending prospects to a landing page that looks like a typical page on your site with a registration form STOP and test something else immediately.

For his tests, prospects were significantly more likely (over 100%!) to respond to a landing page that looked like it was 100% landing page -- isolated from other pages on your site. In fact, the prospects didn't expect the landing page to be "part" of Omniture's site (even if they were clicking on a banner ad inside that site). Instead, they strongly expected a landing page to be "a part" of the banner or other ad creative that they had just clicked on.

This lesson has BIG implications. You need to develop and test your offer landing pages in isolation from your site yet in tandem with your ad creative. It's the reverse of what most people do these days.

Chertudi's other landing page lessons were that fewer columns work better. Plus, relevant, specific images of the offer improve results over generic images. And extraneous links, phone numbers and email addresses reduced response to the form.

(Note: If you need to drive phone traffic, you may want to ignore that finding while making sure that your advertised number leads directly to a live person who'll convert the lead immediately. General office numbers and voice mail or phone trees probably should be abolished for landing pages.)

And, while short copy works better on banner ads that get super-short attention (the shorter copy tested 14% better), business prospects may prefer far more detailed copy on landing pages. (Again, this is the precise reverse of what we see most b-to-b marketers doing in email and other ads. They give longer copy up front and almost none on the landing page. You should test reversing this for better results.)

Lastly, Chertudi discovered to his delight that as long as you keep a registration form fairly short (he now asks for brief contact info plus three questions), you can add questions without hurting response.

Although he does not ask for street address or fax number (we assume both tested badly without enough upside usefulness to keep them), he does ask for job title, department, and Web URL. These latter three questions "don't depress responses at all."

Useful links related to this article:

Creative samples from Omniture's tests:

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.