The Ultimate Bug Fix

To sustain its dominance in the digital era, Microsoft needed to work out the kinks in its marketing machine. Its approach might surprise you.

 By Rob O’Regan

Inside many of the 56 buildings that cut across Microsoft’s sprawling campus outside Seattle, teams of developers toil daily on an endless stream of projects—from short-term product upgrades to long-range, game-changing efforts. Pretty much what you’d expect from the world’s largest software company.

But in the meeting rooms of Building 34, a different pilot project is under way. The focus of this effort is less on code and more on human capital—specifically, a collaboration between engineering and marketing.

For much of the past year, a small team of developers and marketers has met weekly to hash out the next release of Microsoft’s Exchange e-mail server software. Team members have literally written the book on their efforts. The manual—dubbed the “Book of E12″—defines the framework of the next Exchange release. It encompasses everything from the market outlook to the perceived value of possible features to potential customers. At its essence, the book serves as a contract between marketing and engineering, describing what goes in and what stays out of the software.

That marketing would be involved so early in the product development cycle (the version 12 release is not scheduled to ship until the second half of 2006) is a significant change for a company historically dominated by engineers. The Exchange pilot, along with two similar projects, is perhaps the most tangible evidence yet of the far-reaching marketing transformation in the works at Microsoft.

The reinvention of the company’s massive marketing organization, however, is no quick bug fix; executives estimate they’re less than halfway through what may ultimately be a 10-year journey. “I know that sounds like an incredible amount of time, but when you’re a company of 57,000 people and still very much an engineering culture, you can’t have some corporate mandate and expect everyone to dance to it,” says Mich Mathews, senior vice president of the corporate marketing group (CMG), who is overseeing the transformation.

The remaking of the Microsoft marketing machine centers on two main elements: an extensive investment in customer research methodologies and practices, and a comprehensive training program for the company’s marketers—veterans and new hires alike. The goal, executives say, is to institute a consistent “customer value proposition” across the company.

From the top, Microsoft’s executives acknowledge the importance of upgrading the company’s relationship marketing capabilities to improve customer loyalty and extend Microsoft’s already considerable reach into new customer markets. They also admit that they have a long way to go. “We have to be excellent in marketing execution,” CEO Steve Ballmer told a group of marketing recruits in March. “Today, we are not. But we are committed to being excellent.”

The Main Event

During Microsoft’s climb to the top of the software industry, rapid-fire product cycles often happened without much front-end input from the folks in marketing. Engineers would develop new software, pack it with bells and whistles, decide on an acceptable number of bugs and toss it over to marketing for a press release and a launch event.

“The old Microsoft marketing style was that you did an event, and then you waited for the next product release, and then you did another event,” Ballmer said to the marketing recruits. Facetious, perhaps, but the message was clear. Microsoft was not using its marketing function strategically.

Even as Microsoft gobbled up market share, it was gaining a reputation for churning out bloated products with features that users didn’t want, didn’t need or simply couldn’t use because of their complexity. Missing from the R&D equation was any real customer insight. Not a big deal, perhaps, when your core product controls more than 90 percent of a market, as the Windows desktop operating system does. But as the PC software industry matured and Microsoft began branching out into new markets—from online services to gaming to search—the company woke up to a new reality. Real competition emerged among the likes of Time Warner, Sony and Google. Even purveyors of free software, spurred by the grassroots Linux movement, were garnering mind-share against mighty Microsoft.

The company’s marketing function, in particular, was ill-equipped to deal with the shifting landscape. “Marketing always should have been a competency at Microsoft, but it stopped being so some time ago,” says Rob Enderle, principal analyst with the Enderle Group and a longtime Microsoft watcher. “They’re being outmarketed by people who have no marketing budget. That’s all you need to know.”

A Group Grope, and a Common Theme

The mandate for change began in earnest in July 2002, when Ballmer reorganized the company into seven business units that focused on market segments, not products. He also centralized some disparate finance and marketing functions, naming Mathews, a 12-year veteran who came up through the PR ranks, to run the new corporate marketing group. She reports directly to Ballmer.

In October 2003, Ballmer called together his top lieutenants to address the state of marketing at the company. Mathews says the “group grope” helped senior management coalesce marketing’s mandate around a single theme: improving the customer value proposition. “We talked about a common definition of what a value proposition is at Microsoft, and we talked about the process for building the value proposition,” she recalls.

Surprisingly, the engineers who make up the majority of Microsoft’s senior management team didn’t need much convincing. “I expected a bit of a debate, but the discussion was very passionate about the importance of driving growth through stronger value propositions,” says Ballmer. “We were pretty product-centric in our marketing, which meant we weren’t always delivering a higher-level perspective on the value of technology in key areas.”

The engineers were eager for a better understanding of customers, and they acknowledged that a partnership with marketing was the best way to capture those insights. “There were lots of growth opportunities in different areas of technology,” says Andy Lees, the corporate vice president in charge of server and tools marketing. “We wanted to add more discipline into how we could capture those opportunities.”

Microsoft had always placed a high value on customer research, but those efforts focused more on measuring the impact of products and services after their release, not in anticipating how a target customer would use them. Take, for example, the 1997 release of a product for small-business customers. Microsoft’s early marketing efforts around Windows Small Business Server focused on its extensive features and the productivity gains it would provide. But the launch team overlooked two problems: The software was too difficult for the target audience—generally nontechnical users—to set up by themselves, and it was too expensive. “That product really bombed when we introduced it,” Mathews admits.

Compounding matters, there was no real rhyme or reason to the type of research that the business groups were commissioning. “We had a bit of the Wild West approach to research,” says Jeff Hansen, CMG’s director of research. “Any product manager with a buck and a dream could launch a research project. That led to lots of activity, but it wasn’t all driving the important decisions.”

Enter the customer research system. Launched two years ago, the system puts a heavy emphasis on what executives call “informed research”—the type of market analysis and customer insights that drive decisions about products or positioning before products reach customers’ hands.

Staying Informed

When Hansen’s team began developing the research system, it looked outside of Microsoft’s walls for best practices. After inviting a consultant from Procter & Gamble and representatives from a handful of banks and pharmaceutical companies to talk about their research methods, the team realized it needed to get more systematic about its research.

The customer research system framework was developed to address that need. It includes three main elements dedicated to informative research for products still in development, along with the traditional processes for measuring customer satisfaction with existing products. Behind each element lies a series of processes and methodologies. “The framework lets us sit down with the business units to have a more thoughtful, proactive conversation about what research needs to be done,” says Hansen.

The conversation begins with what Hansen calls “essential questions”: Do you know where the biggest opportunities exist? Do you understand not what customers say they want, but what they actually need? How do they react to the prototypes you’re working on? “The answers to those questions give us a sense of where we have coverage and where we have gaps,” says Hansen.

Selling the importance of these give-and-take sessions would be hardest to veteran technologists who were used to a much different product development cycle. “The notion of concept testing or thresholds that we have to meet before something goes into the marketplace was culturally different for our engineers,” says Mathews. “They’d say, What do you mean you’re going to test my product?”

Adding more rigor—in the form of hard customer data—has given marketers an important ally in their discussions with developers. “We are really pushing hard for our marketers to live and breathe the data,” says Ballmer. “I tell them to understand it, digest it, analyze it, form insights and take those insights into their daily work.”

The company’s marketers certainly have plenty of data at their disposal. The website, which has become the hub of the company’s newfound emphasis on relationship marketing, receives 118 million unique visitors a month. Lees claims that more than a quarter of all IT professionals and developers in the world come to the site at least once a month. “There is nothing, nobody, anywhere else that has that level of reach into the target audience,” he says. “It’s an amazing asset.”

Microsoft’s customer research efforts go far beyond Web analytics, however. “It starts with the data and ends with the data,” says John Kahan, Microsoft’s general manager of customer relationship marketing. “But there’s a lot in between in terms of getting the right message to the customers at the right time.” In addition to the website data it’s collecting, the CMG audits 17,000 customers annually to measure satisfaction levels and performs ongoing studies to track customer behavior. It’s also investing in ethnography and predictive analytics. “We’re trying to tie what customers say makes them happy to what they actually do,” says Kahan.

The goal is to analyze all customer segments to identify new ways to increase loyalty and satisfaction. And that’s perhaps the biggest change over the past two years. “We’ve turned the corner on customer satisfaction,” says Kahan—so much so, he says, that the company’s 600 top executives have customer satisfaction metrics built into their compensation packages.

Training Days

Behavioral research and predictive analytics are a long way from PR and event management. When Mathews and her team began their marketing reinvention, they realized there was a significant gap between the skills of existing marketing personnel and the expertise required to execute on the broader mandate.

In the spring of 2002, CMG formed a new training arm, called Microsoft marketing professional development (MMPD), to lead the charge on the skills upgrade. Working with an outside consultant, MMPD created an entire curriculum around eight core competencies, ranging from value proposition design to segmentation to customer insights. It created an online self-assessment tool for those competencies that is now built into all marketers’ performance appraisals.

The team also partnered with Northwestern University’s Kellogg School of Management and Duke University to create training and development programs targeted at various levels of expertise. Today, all incoming undergrads and MBAs hired by the company spend their first two years in a two-day-a-month program called Marketing Leadership Recruiting, which functions as a “mini-MBA” on Microsoft’s marketing practices. More experienced marketers hired from other companies attend a two-day “boot camp” on the Microsoft campus to learn the basics of Microsoft marketing.

The toughest task continues to be retraining Microsoft’s existing marketing personnel. Marketing directors who are skilled at PR or analyst relations are asked to embrace a much broader mandate around relationship marketing. “Many don’t have that muscle,” says Mathews. “Some of the new kids coming in, frankly, are more seasoned in relationship marketing than the managers they are working for. That’s been an interesting dynamic.”

Most seasoned managers are eager to learn the new competencies, says David Hamilton, MMPD’s director. “Our marketers are so thirsty,” he says. “They might not like what the competency appraisal tells them, but at least it’s not a subjective discussion on how you rank versus other people.”

For these veteran marketers, Microsoft has worked with Kellogg to develop a custom curriculum based on the eight competencies. “They used a classic marketing technique: Find the need, find the gap and fill it,” says Mohanbir Sawhney, a Northwestern Kellogg School of Management professor who helped Microsoft develop the curriculum. “We worked backwards to design the training element for very specific competencies.”

The results have been impressive. Ninety days after the initial training program, the team compared those who had taken the training courses against the control group. “We found that we had significantly moved the needle on the competency we were targeting,” Sawhney says.

“I have not seen such a systematic approach to marketing resource development, at least in technology companies,” he adds. “What is unique is that marketing, not HR, has led this process.”

The training is just beginning to penetrate the ranks of Microsoft’s massive marketing organization. Only 12 percent of the company’s 5,000 marketers have gone through at least one of the formal training programs. “We want to move that number up by tying training much more closely to developing competencies,” says Hamilton.

The commitment from senior management should go a long way in driving those efforts. “Leadership needs to realize that they have something to learn too. Microsoft has done that,” says Sawhney. “Ballmer spent four hours in one of the training classes I gave. When the CEO talks the talk and walks the walk, people start to pay attention.”

The Long Journey

With the elements of its customer research system and training programs in place, CMG last June kicked off its three value proposition pilot projects: one for Exchange; one for the next version of Windows; and one for a new, unannounced Office product. These teams already had shown a good working partnership between marketing and engineering, thus increasing the chances for a successful pilot.

“We want our marketers to be working hand-in-hand with engineering,” says Mathews. “You can imagine the arguments that are taking place. It’s a great way to ensure that we are really differentiating.”

The Exchange team is furthest along. Its 55-page “Book of E12” features a series of user scenarios—how a user protects his system from junk e-mail, for instance—that the engineers developed working with marketing’s customer research. Features that don’t address specific scenarios are dropped from the design.

“The scenario aspect is a good forcing function to justify every feature in the product,” says Jeff Ressler, director of product planning for Exchange and the team leader for the pilot. “Even though we might think development teams are set in their ways, there is flexibility. There is a craving on both sides for information about what customers want.”

The impact of the value proposition pilots won’t be fully measurable until the products reach the market beginning in 2006. In the meantime, CMG has begun retrofitting its new methodologies to existing products that are underperforming. Applying the value proposition framework to its Windows Small Business Server software—the product that bombed in the late ’90s—convinced the team to cut the price and offer a pre-installation option. Sales rose “like a hockey stick,” Mathews says.

For now, Mathews will take these short-term victories. The larger battle, to win the hearts and minds of 57,000 employees worldwide, is far from over. “We’re beginning to see success,” says Mathews. “But in all honesty, it’s far too early to claim victory.”

The biggest challenge, everyone agrees, remains cultural. “You have to think differently about what you’re doing,” says Enderle, the consultant. “It’s hard for the superstar engineers you’ve hired to admit they don’t know something. Right now they’re saying the right things, but I’m waiting for them to start doing the right things.”

From CMO magazine (a publication of CXO Media, Inc.), June 2005

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: