Omniture Articles

Monster.com Gains Profits With Revenue Per Visitor (RPV) Testing

Opportunity/Problem:

Monster.com is known for connecting job seekers with leading employers across all industries, around the world. It has 25 local language and content sites in 23 countries worldwide, and is the fourteenth most-visited property on the internet. But Rathin Sinha, VP and GM of e-Commerce for Monster Worldwide, knew that for all the site's popularity with employers, significant revenue upside is available by improving the customer buying experience on the site.

He believed that by testing a variety of treatments on various pages within the site, Monster could significantly increase revenue per visitor.

Results:

Revenue increase from testing with Offermatica is estimated at tens of million of dollars for Monster’s eCommerce channel. The employer home page test resulted in an 8.31% lift in revenue per visitor (RPV) over the default, while the jobs page test resulted in an 11.6 % lift in RPV over the default.

“Although the total revenue lift is attributable to several factors including marketing and advertising efforts, we are confident that we got our money's worth from the Offermatica investment,” Rathin comments.

In addition, Rathin's team has found that the more they learn, the more they can do to improve their business using Offermatica. “We are planning to use it to test and improve our email marketing and other selected direct response marketing areas," he says. In fact, he adds, "We really want to get into a framework where we would be doing this kind of testing continuously.”

Testing Approach:

Working with Offermatica, Rathin's team wanted to increase revenue on the home page and job pages. The first test, on the employer's home page, was designed to see if they could increase revenue per visit by changing the presentation of job postings, promotional copy, and layout. The second, on the jobs page, sought to improve average order value and revenue per visitor by educating users on the value of buying in bulk.

Actions Taken:

1. Employer's Home Page
The first series tested two variables of seven different elements:

  1. promotional copy (default or existing copy vs. stronger offer copy);
  2. copy for the job posting button;
  3. "learn more" link (link present vs. link not present);
  4. "Search and Buy Resumes" button (existing copy vs. "Search Resumes" new copy);
  5. "Search and Buy Resumes" test drive ("Take a free test drive" present vs. not present);
  6. main button design (depth added to buttons vs. no depth);
  7. and layout of top 3 boxes (button at the top of the boxes vs. button at the bottom).

Default Layout: The original, default page included existing copy, "Search and Buy Resumes," and original main button design. [To see the full design details, click on the images below]

Best Tested Layout Recipe: This recipe included stronger promotional copy, "Search Resumes" vs. "Search and Buy Resumes," and different main button design

Although there were 128 possible variations on the combination of elements and alternatives, by using multivariate testing the team was able to test using only eight test "recipes," (or pages with different combinations of elements) to accurately see which combination would be most successful at improving revenue per visitor.

2. Jobs Page
This series tested three different alternatives of four elements:

  1. sell copy (original copy vs. conversational copy vs. bulleted copy);
  2. headlines (default vs. two new versions);
  3. savings listed (savings in percent vs. savings in dollar amount vs. no savings present);
  4. savings calculator (savings calculator in dollar amount vs. savings calculator not present).


Default Layout: The default page included original copy and layout, no displayed savings amount, and no savings calculator.

Best Tested Layout Recipe: This recipe included a new layout, stronger headline, savings in dollars from buying in bulk, and a savings calculator.

Again, while there were 81 possible permutations of elements and their alternatives, through multivariate testing the team was able to test only nine recipes with different combinations of elements.