October 13, 2006
Case Study
SUMMARY:
Marketers for print publications are lucky online because they have heaps of keyword-rich content that draws search engine traffic. Unfortunately, most of that traffic goes sliding off again without converting.
Here's a real-life Case Study of a B-to-B subscription marketer who tested landing page tweaks to increase conversions from both PPC (paid search ad) and organic traffic.
Yes, includes before-and-after landing page screenshots:
|
|
CHALLENGE
“We were very happy with our traffic, but we were never very happy with our conversion rates,” says John Brady, referring to organic search metrics for Business & Legal Reports (BLR) prior to 2006. A few took subscription trials or bought something, “but most just seemed to disappear.”
When considering the search conversion problem, Brady, an owner, with brother Bob Brady, in the B-to-B publishing company that Bob founded in 1977, had to consider the site as a whole -- because every publicly available page of your site is a de facto SEO landing page with incoming traffic to some degree.
Brady changed the goal of the search-facing site to growing his list -- generating email opt-ins, registered leads first and foremost. He could sell them on the back end after that. With that in mind, what page tweaks would work best to convert search traffic into lists for further marketing?
CAMPAIGN
Brady's team began the revamp project by focusing on paid search landing pages for two reasons. First, traffic to those pages is far easier to control than SEO pages because you pay for what you get. Second, since you pay, you want to maximize that ROI.
Those components: More compelling headline (a stronger benefit, pegged more closely to the key words that generated the visit); drastically shorter copy; reduced navigation options (a nearly total focus on the free-trial offer); addition of a large photo; and a different registration process (a button clicking through to the form in place of a form embedded in the landing page).
In stage two of this multi-year marketing test, the team applied their lessons learned to organic entry pages. (Note: Unlike paid search landing pages, these are regular site pages; otherwise, a spider wouldn’t be able to find them.)
They determined that a three-week test would have enough conversions to be statistically reliable. Obviously, traffic and conversion levels for a particular page are what matters, not a particular time period. A higher traffic B-to-C publisher could have gained the same data in a matter of hours.
Brady identified the three search terms that generated the highest traffic for their largest sub-site, http://HR.BLR.com, then tracked all conversions generated by that site’s SEO control pages for three weeks.
Then they served only the test landing pages and tracked all conversions generated by the same three search terms.
The organic test pages differed from the organic control pages in five ways:
A. Compelling headline copy.
This included a stronger benefit pegged more closely to the key words that generated the visit. (“Confused about how cell phone laws might affect your employees?” versus the control’s “Cell Phone Laws.”)
B. Drastically shorter copy.
Although the test page’s main copy was not appreciably shortened, subheads were eliminated, as was copy offering specific advice (e.g., what policies to set regarding employees’ use of cell phones on the job).
The advice paragraphs were replaced with copy promoting BLR’s human resources library database and tools as a one-stop informational resource.
C. Fewer navigation options.
Extraneous links -- such as ones leading into “additional sources,” BLR’s catalog offerings, a site tour and an online help service -- were eliminated so that the only options were to register for the free trial subscription or leave the site.
“When we removed those other links, we made sure that we also tied the pages together for improved navigation,” Brady points out.
D. Embedded registration form.
The control organic entry pages had a “Free Trial” button/two-step process, so (in a reverse of this aspect of the paid search tests), the test SEO landing pages featured an embedded registration form.
E. Sales copy above the order form.
Removal of the extraneous link areas made room for a box above the embedded order form featuring bulleted sales copy that was customized to relate closely to the chosen key words.
RESULTS
The findings from the two paid search test campaigns in late 2004 and late 2005 contributed to free trial signup rates jumping from 3.11% to 5.19%.
To Brady’s delight, for the SEO entry pages free-trial signup response rates leapt by 150%. The control landing pages’ format generated a response rate of 1%, while the test landing pages’ format saw a 2.5% response rate.
Brady attributes the increase in the number of visits to the test entry pages to the simpler, improved navigation structure, but he credits the more focused sell approach for the jump in the free-trial conversion rate. “In the successful landing page version, no one wanders around. They either act on this page or they don’t act at all.”
The new template was rolled out across BLR’s four sub-sites beginning in June 2006, along with an expansion of the targeted search terms. Paid conversion rates are still being monitored and analyzed, but so far look the same or better than those for the control landing pages.
MarketingSherpa's Search Marketing Benchmark Guide has shown that organic conversions generally are very close to paid conversions if you have an average conversion rate.
Useful links related to this article
Creative samples from Business & Legal Reports search tests:
http://www.marketingsherpa.com/cs/blr/study.html
The Conversion Zone - Brady's new company that helped with the testing:
http://www.theconversionzone.com
Business & Legal Reports:
http://www.blr.com
