It was just 9:30 a.m., and the day was off to a terrible start. Heather Yates, vice president for business development at MedNet, walked at a quick clip down the hall of the company’s modern Birmingham, Alabama, office space, her face clouded with concern. The company, a website delivering health information free to consumers, generated its income through advertising, mostly from pharmaceutical companies. Now, Windham Pharmaceuticals, MedNet’s biggest advertiser, had asked to change the rules by which it had done business for the past four years. Moreover, Mahria Baker, Windham’s CMO, had told Yates that this wasn’t just an exploratory conversation. Windham was seriously considering shifting its MedNet ad dollars to Marvel, a competing website with which Windham already did some business. Yates, who had been with MedNet since just after the company was founded in 2002, felt blindsided and, at the same time, resigned. “We have some legwork to do,” she thought to herself.
“We can’t afford to say ‘No,’ and just walk away, and we can’t just ask them to stay with us because we’re good people. We have to convince them that our set-up is worth their ad dollars. And we have to move quickly. Our other advertisers won’t be far behind Windham.” She had asked Baker to fax over a copy of the results of Windham’s latest advertising campaign, and had promised to call her back the next day, as both companies needed to finalize their budgets. Then, immediately after they had hung up, Yates had called Bill Bishop, MedNet’s vice president of consumer marketing. “Can you clear some time for me right now?” she had asked him. “Windham is thinking of pulling their ad dollars from us and taking them to Marvel.” Now she was on her way up to Bishop’s office, two floors above, with the fax from Baker and notes from her conversation in hand.
Industry Background and Company Origins
This case, though based on real events, is fictionalized, and any resemblance to actual persons or entities is coincidental. There are occasional references to actual companies in the narration. Copyright © 2007 by Harvard Business School Publishing. To order copies or request permission to reproduce materials, call 1-800-545-7685, write Harvard Business School Publishing, Boston, MA 02163, or go to http://www.hbsp.harvard.edu. No part of this publication may be reproduced, stored in a retrieval system, used in a spreadsheet, or transmitted in any form or by any means—electronic, mechanical, photocopying, recording, or otherwise—without the permission of Harvard Business School Publishing million in profits. (See Exhibit 1 for 2006 income statement.)
The accessibly written, easy-tonavigate, and vividly presented content was developed by 24 trained journalists, doctors, designers, and administrators. Additional materials came from the faculty of a prominent medical school, news agencies, a photography service, and an active community of visitors that used social media tools such as blogs, community chat, and virtual reality to communicate medical information. (Visitorgenerated media was reviewed by medically trained journalists.) The award-winning site was considered the best health website for trusted, evidence-based, consumer health information. Advertisements on MedNet proposed specific and immediate solutions to health concerns. MedNet had 4.3 million monthly visitors, but new competitors had flattened its audience growth during the last quarter of 2006.
Now, in the first quarter of 2007, MedNet faced competition both for visitors and advertisers. Nonprofit and governmental websites competed with MedNet for visitors by providing similar content on mainstream medicine. The websites of the U.S. National Library of Medicine and World Health Organization weren’t nearly as easy to navigate as MedNet, but they were comprehensive. In contrast to MedNet, these two websites provided information on alternative therapies as well as on scientifically based solutions, albeit with with carefully worded disclaimers. What’s more, employees of large corporations could increasingly turn to customized health websites on their own company intranets. The theory was that if internal health websites could help workers quickly identify health problems (prompting overdue doctor visits) and promote general good health, the employers could reduce their portion of employee health care costs.
For-profit health websites posed different degrees of financial competition for MedNet’s advertising revenue and audience. Recently, so-called condition-specific sites that focused on particular problems, such as Cholesterol.com, had emerged. (Yates was confident that Cholesterol.com was already drawing pharmaceutical advertising dollars away from MedNet.) An indirect competitor, ClinicalTrials.com, marketed only experimental procedures. Its audience was smaller than MedNet’s and the material was difficult for the layperson to understand. ClinicalTrials.com received a fee for each time a visitor it referred enrolled in a clinical trial.
Then there was Alternativehealth.com, a long-time, popular player in the “health space.” It provided information about scientifically “unproven” therapies and procedures such as herbal remedies, vitamin regimens, and massage. Its audience was larger than MedNet’s and its advertising sales more robust. Due to a recent lawsuit concerning its content, Alternativehealth.com had begun using disclaimers—with no apparent impact on its audience size. Due to the alternative health consumer’s distrust of pharmaceutical companies, the website did not compete with MedNet for advertising dollars. Still, MedNet had to keep Alternativehealth on its radar.
Methods Used to Calculate Advertiser Payment
Yates’s thoughts raced through the company’s competitive landscape as she waited for the elevator. In her short phone conversation with Bill, he had
told her to take a little time to review MedNet’s original value proposition to its advertisers. What they needed to do was re-justify their approach, if it was possible to do so. But, he had cautioned, they were compelled to keep an open mind. “Think through the facts,” Bill had said. “Why don’t you come up here in about half an hour. I’ll start to mull over our options as well.” Yates thought back to MedNet’s roots. Back in 2002, MedNet’s founders had made some key choices regarding revenue generation. MedNet could, in theory, sell content to site visitors, like an online magazine, charging a few dollars per article or an annual subscription fee.
On the other hand, if the site could draw advertisers, and if advertising revenues were strong enough, the company could provide content free of charge—which is what most web users expected. An advertising revenue model was made possible by sophisticated web analytics: technology that tracked the behavior of each site visitor—pages viewed, links clicked, and so on. This software made it easy for advertisers to calculate their return on advertising investment (ROI). The obvious candidates to buy onscreen advertising space from MedNet were pharmaceutical companies, which for over a decade had promoted their drugs aggressively to consumers.
As it happened, MedNet was launched at a time when many other consumer health care websites were going out of business, leaving pharmaceutical firms looking for web promotion outlets. MedNet seized the opportunity to build relationships with these advertisers. In deciding how best to generate revenue from advertisers, MedNet chose traditional banner advertising, charging pharmaceutical advertisers such as Windham Pharmaceuticals on a cost-perthousand impressions (CPM) basis. (One advertising impression meant that one visitor requested from a Web server a page that had a specific advertisement on it.) Measuring impressions was the closest way to estimate the number of people who actually saw an online advertisement. By pursuing an impression business model, MedNet was fully “monetizing” its available inventory of “eyeballs” (site visitors). An independent auditor verified the company’s impression counts each month.
Yates reached Bill Bishop’s office and pushed the door open. Bill was on the phone, but he waved her to a seat. “Two minutes,” he mouthed at her. She nodded, and sat back. She thought about what she knew about Marvel. Marvel was essentially a large search engine that had decided to follow the alternative advertising model: contextual, or pay-per-click, banner advertising. Under these terms, advertisers paid website owners only when visitors actually “clicked” on an advertisement to learn more about an advertised product. The key metric to measuring this kind of online advertising campaign was the click-through rate (CTR), measured as the number of clicks divided by the number of ad impressions delivered. Advertisers considered website click-throughs (and telephone calls to a call center generated by a newspaper advertisement) to be the equivalent of customers interested in potentially making a purchase.
Yates thought back to 2002. No sooner had MedNet’s founders opted for a pay-per-impression model than advertisers began resisting that pricing structure—but mainly from general-interest websites, where the majority of impressions came from visitors uninterested in their products. Advertisers based this perception in part on the percentage of click-throughs that ads yielded; the click-through rate on a general-interest site tended to be half as high as on highly focused “destination” content sites like MedNet.
In 2006, MedNet.com therefore could still command a $100 CPM ($100 for each 1,000 impressions) contract from its advertisers—10 to 20 times what general interest websites might charge. Similarly, Alternativehealth.com’s advertisers paid for impressions only, and not for click-throughs. But Marvel, a hugely successful search engine, turned the table on its competition in the fall of 2006 by declaring it would provide impressions for free and charge advertisers only for clickthroughs. Because Marvel had a vast audience (19 million visitors per month), charging for even a small percentage of click-throughs would pay off handsomely. If the site sold advertisements in enough categories, including the pharmaceutical market, Marvel could bring in huge revenues. By late 2006, some advertisers began to ask other sites to charge only for click-through “sales leads” like Marvel did. One drawback to this popular revenue model: reports of increasing “click fraud.”
Advertisers’ competitors were fraudulently clicking on advertisements to drive up advertising costs. Not only was Marvel offering MedNet’s long-standing advertisers like Windham different financial terms, but it also competed for visitors interested in healthcare. Visitors often came to MedNet by way of a search engine such as Marvel, although such search engines served as a starting point of inquiry, not a serious source of trusted medical information. Mahria Baker’s challenge stuck with Yates: “At Marvel we get all our impressions for free, and we pay $0.54 for each click-through. At MedNet we pay for every impression, and by my calculation we pay $3.33 for each click-through.
Granted, we’re not averse to getting impressions—anytime that anyone sees your logo, your slogan, and your product’s name, you are theoretically doing your brand some good. But here at Windham, click-throughs are really what matter. They separate accidental observers of our ads from the serious prospects who proactively seek more product information and may buy our product. I can’t justify paying six times as much for a click-through from one of your visitors.” Baker had paused a moment, then added, “Heather, help me here. Is there another way of looking at this that I’m missing?” “Yes, there is,” Yates had replied, “and if you let me call you back tomorrow I believe I can show you what you are missing.”
MedNet’s Audience and Visitor Behavior
Bill Bishop hung up the phone and turned to Yates. She spread out a copy of the results of Windham’s latest advertising campaign, and the two of them pored over it. (See Exhibit 2 for Baker’s data.) Many search engines and general-interest websites had large audiences that returned to the sites regularly, in a predictable pattern. By contrast, most visitors to targeted health websites such as MedNet came only when “in crisis.” However, when they did come, they stayed long and explored avidly, clicking around to clarify symptoms or determine the best course of action for a pressing health problem. They often researched unrelated symptom areas as well, in order to help family members, or out of curiosity. These visitors then returned during the next crisis, although some did become repeat visitors.
MedNet visitors clicked on more pages and advertisements than generalinterest web surfers did (see Exhibit 3). In addition, health website visitors tended to buy more products from advertisers when they did decide to purchase. (See Exhibit 4 for a study of results and frequently viewed web pages on MedNet.) If the product advertised was not available over-thecounter, then the visitors would urge their physicians to prescribe the medication that they’d discovered in the advertisements on MedNet. Windham produced Vesselia, a prescription medication that reduced cholesterol and plaque in a patient’s veins with fewer side effects than competitors’ offerings.
High cholesterol was one cause of heart disease, and it was attributed to both genetic predisposition and lifestyle choices. Keeping cholesterol low could be a long-term issue for many patients, requiring months, possibly years, of daily medication. Each patient who began a series of treatments would use the medicine for an average of 12 months. To encourage customers to request a prescription for Vesselia from their doctors, Windham provided coupons on its website that customers could print out and redeem at a pharmacy.
For instance, when a customer clicked on a Windham ad at MedNet’s website, he was taken to the Windham website. Windham’s computer system could identify that the customer came from MedNet and insert that information into the Windham coupon bar code within fractions of a second. A different coupon code was provided to those web visitors who came to Windham from Marvel Search. (Coupons with yet another barcode were sent by postal mail by the Windham telephone call center to respondents to newspaper advertisements.) When patients redeemed the coupons at a pharmacy, the pharmacy returned them to Windham. Windham could thus attribute drug sales to the relevant advertising venue. On average, patients took three months to redeem coupons for Vesselia after Windham had first placed the advertisements. The current campaign would be considered closed at the end of February 2007.