Paul Holmes 05 Jul 2006 // 11:00PM GMT
As public relations agencies have grown, often operating in multiple geographies across multiple industry sectors and multiple practice areas, knowledge management has become a critical priority for senior executives. Most large agencies—and many smaller firms—have developed increasingly sophisticated intranets to ensure that they can identify and access relevant expertise on behalf of clients, no matter where it resides within the business.
Knowledge management is a challenge, but knowledge is a not the only—or even the most—valuable commodity that individuals within an public relations firm possess and may, on occasion, prefer to horde rather than share. Relationships are often just as critical to the success of a public relations firm as specific expertise, but they can be much more difficult to track.
“Pretty quickly someone in your organisation could locate a wiring diagram showing how your department is electrically connected to the rest of the building and to the wider world,” says Christofer Solheim, worldwide director of business intelligence at Hill & Knowlton. “But where are the handy diagrams which map human relations? These human connections are at least equally important yet, in all probably, the best you could lay your hands on is a standard organization chart.”
And organization charts don’t tell the whole story. “The formal information hierarchies of an organization are not necessarily the same as the informal hierarchies,” says Niall Cook, who heads the netcomms practice at Hill & Knowlton’s London office. “And they are not necessarily as effective. You need to know who the connectors are, how the information really flows.”
Kate Ehrlich, a researcher at IBM, cites the example of a management consulting firm that lost out on a lucrative contract with a large financial institution. Six months later, management learned that another group within the consultancy had worked on a project with the prospective client and had in-depth knowledge of its business operations. “If that crucial knowledge had been shared, the outcome might have been different,” he says.
The emerging field of social network analysis can help. In a development some believe will add enormous value and others may find mildly Orwellian, new technologies are helping public relations firms and other large companies track relationships—who has them, how many, and with whom. In some cases, they are discovering that the most powerful people in the firm, or at least the most connected, reside at all levels, often in unexpected places.
“Social Network Analysis is a set of survey methods and statistics that reveals the hidden connections between people,” says Ehrlich. “The outcome of an SNA shows where collaboration is breaking down, where talent and expertise could be better used, where decisions are getting bogged down or where opportunities for innovation are being lost. The data give leaders the information they need to take actions: perhaps including making role and responsibility changes that would foster cross-group communications; developing methods for improving trust; using technology to reach others more effectively; or realigning reward and incentive programs.”
Food giant Mars, for example, brought more than 300 research scientists together in Las Vegas a couple of years ago to forge new relationships. In a giant ballroom the scientists—each of them wearing an RFID-enabled name tag—introduced themselves to their colleagues. Each time one of them met someone new, a diagram of their social networks was updated on a giant screen. The company offered prized to the scientists who made the most contracts.
That exercise, reported in CIO magazine, followed a yearlong study of social networks in the company’s research and development division, which was spread over several locations. Senior executives hoped to enhance the company’s ability to innovate and were concerned that their scientists weren’t networking enough with colleagues around the country. To identify existing networks, management mapped the group’s professional contacts, using an online survey that invited R&D managers to name the 15 people with whom they worked most closely.
The company was able to determine which scientists were over-burdened (too many people were going to them for help) and found a lack of communication between its snack food division in New Jersey and its food division in Los Angeles. To remedy that, that company made inter-departmental connectivity an important criterion in performance appraisals.
Social network analysis can also help companies understand the knowledge and relationship capital they risk losing as older employees retire—a particularly pressing issue given the aging population in most developed countries.
“We thought social network analysis was a way looking at how information spreads throughout the organization,” says Ted Graham, who directs H&K’s knowledge strategies globally from his base in Canada. “That’s particularly important in an industry as dynamic as public relations, where people change roles so often, and often take their knowledge with them.”
It can also help companies improve the efficiency of internal communication and collaboration.
“Unproductive relational demands sap people’s time and energy and can bog down entire organizations,” says Rob Cross, assistant professor of business at the University of Virginia and author of The Hidden Power of Social Networks, writing with University of Virginia colleague Jeanne Liedtka and McKinsey’s Leigh Weiss in Harvard Business Review. “Decision makers can become so consumed by managing contacts that their coworkers often cannot get to them in time to take action on opportunities.
“Many of us struggle to keep up with e-mail, phone calls, and meetings-often with substantial implications for both performance and quality of work life. That’s why it’s crucial for executives to learn how to promote connectivity only where it benefits an organization or individual-as well as learn how to decrease connectivity that isn’t needed.”
And some public relations firms—including H&K—are also beginning to monitor social networks among the external stakeholders of their clients, thus developing a better understanding of the way information about those organizations is disseminated, and who the real influentials and opinion leaders are on a company-by-company basis.
Social network analysis emerged in the 1930s as a tool for scientists who were exploring the dynamics of social interactions, and is defined by social networking expert Valdis Krebs as “the mapping and measuring of relationships and flows between people, groups, organizations, computers, or other information- or knowledge-processing entities.”
The pioneers in the field would draw “sociograms” to depict the relationships between members of a network. But the development of new technologies to monitor interactions—particularly those that take place online—and software that can map increasingly complex webs of relationships have made the discipline more accessible and more attractive to a wide range of corporations and other organizations.
“Companies want to have a picture of who the key knowledge brokers are in their organization,” says Cross. “The rise of blogs, online support sites and social networking sites—such as Friendster and LinkedIn—has also helped raise SNA’s profile.”
Technology has had a profound impact on the improved affordability and accessibility of social network analysis in recent years.
“So many communications are electronic these days that it has become much easier to record who is talking to whom,” says Stanley Wasserman, professor of sociology, psychology and statistics at Indiana University and chief scientist for Visible Path, a software company that monitors social networks. “It’s a natural thing to examine these networks and try to make sense of them.”
Solheim cites several factors in H&K’s decision to join the Network Roundtable and to invest time and resources in SNA:
• Collapsing cost: “New analytical capabilities and services evolved which take advantage of the existing media databases across print, broadcast, web and blogs.”
• Greater speed: “Some of these services can extract automatically, at awesome speed, the mentions of brands, companies and people and produce models/diagrams of their inter-relationships.”
• Greater need: As well as the demand for measurement in planning, execution and evaluation, the demand for better research and analysis, a new factor is the influence of websites and especially blogs and other consumer generated media, which can really only be understood and interpreted as networks of influence.
• Protection: “The competencies of PR professionals are of critical importance in understanding the networks of influence and the wider reputation and business implications.”
“The key change is the recent introduction of inexpensive and relatively simple software for the collection, organisation and diagrammatic presentation of human relationship networks,” he says. “This has positive implications for CEOs, COOs and HR professionals seeking to align employees behind a corporate strategy.
“For marketers and communicators, it offers the opportunity to understand better the external networks of influence which determine organisational success and how best to connect to them.
More and more, marketers are required to provide a research-driven, clearly argued, measurement supported case for investment—just as in every other aspect of business. Best practice requires that we stand with one foot in the arts and the other in business science.”
According to Ehrlich, a social network analysis is typically conducted with a group of 15 to 150 people at a time. “The group could be as simple as a discrete workgroup, such as a distributed software development group or a sales team,” he says. “Or the group could be a community of interest, a business unit in a single organization, or the leading companies in a particular industry.”
The SNA looks at relationships types, from reporting lines to communications between individuals to trust levels and even personal feelings (do people in the group like each other). Questions in the SNA might include: “How aware are you of the projects done by this person in the past 12 months?” or “How often does this person provide you with information you need to develop client proposals?”
The SNA might look at the nature of the ties between individuals. Are they direct or indirect? Are they strong or weak? (“Strong ties are characterized by frequent interaction, feelings of closeness, and multiple types of relationships. For example, a strong tie may provide you with emotional support, job-related information, and a person to go see your favorite sci-fi movies with.”) Are they one-way or reciprocal? (“Reciprocal ties are generally stronger than ties that only go in one direction. For instance, a group will generally function better when a key decision-maker is not only sought after for information and advice, but he or she also seeks information from the group.”)
According to Cross and Andrew Parker, an IBM executive and his Hidden Power co-author, there are two types of “central connectors.” “Unsung heroes,” respond directly to requests for information, engage in problem solving, provide personal support and put people in contact with others, often without any formal recognition. But other central connectors hoard power and information. They delight in being “well connected” but guard their connections jealously, viewing them as a source of personal power within the organization.
Cross and Parker also identify several other types: “Boundary spanners,” who create connections between departments or maintain vital external links; “Information brokers,” who communicate across sub-groups of informal networks; the excluded, who need to be brought back into the fold, and the talented, many of whom “just want to be left alone,” and probably should be.
Says Ehrlich, “By looking at who was central in the awareness and information-sharing networks, IBM consultants could show the company which people the team tended to go to for information. By looking at how often people from each group interacted with each other, the SNA might reveal that consultants who had been with the company the longest tend to get information from each other; while newer employees only turn to each other for more information if they are located in close physical proximity.”
At Hill & Knowlton, Soldheim says, a typical SNA might include gathering input from individuals (sometimes via a survey, sometimes by monitoring e-mail traffic data); a calculation an examination of the network characteristics; a report that includes visualisation of the network, usually optimized and focused to illustrate key findings; benchmarking against SNAs of similar subjects, correlated with business performance whenever possible.
Internally, H&K has used SNA surveys to measure how well a particular practice group was collaborating and innovating across a region and then to offer a “personal network diagnostic” to employees. And it has examined individual contacts and employment histories in a large office to discover media, government and client relationships across practices.
Such applications will be most valuable to larger agencies, but even the smallest firm can benefit from applying social network analysis teams on behalf of its clients.
When the first stores about avian flu began to appear, for example, Hill & Knowlton was able to use its software to search thousands of articles, identifying key opinion leaders in the U.S., Canada, the U.K., Australia, Hong Kong and Singapore and producing charts that showed the relationships between those opinion leaders.
SNA has been particularly useful in the issues management arena, helping the firm analyze the way stories spread and make the jump from the blogosphere to the mainstream media; determine the most influential commentators—journalists and the sources—on an issue; and map the external relationships of those within the agency, the client, or the industry to discover who might be in a position to influence the influentials.
Niall Cook says he first got into social networking about three years ago when a client was on the receiving end of some bad press, online and off. “Someone said it would be really useful to have a map of all the connections between people who were commenting on this issue, to understand the links. I started to look at social network analysis, which seemed to be used mostly internally, and thought we could apply the same principles quite easily to look at external networks and relationships.”
Meanwhile, in Canada, Graham was attending a conference on SNA and found himself talking to a consultant from McKinsey. “At first, she was surprising to see someone from a PR agency at a conference on the subject. Then she said she thought it would be interesting to see how outside influencers interacted with one another.”
Graham had been focused on the internal, knowledge management aspects of SNA, but quickly realized that H&K had access to media databases and archives that would make applying the same process to the external environment a relatively simple matter. The most difficult challenge was creating a software package that allowed the firm to extract the names of sources from media coverage in an automated way: H&K finally hired an expert in natural languages to come up with a system that Graham says is “95 percent accurate.”
Public relations people have always monitored such things, of course, but using SNA techniques and software, H&K was able to come up with a map, “a way of visualizing the relationships,” says Cook.
Technology also made the process faster. “We can pull together a set of articles using Factiva and in about an hour we can run through those articles and pull out the names of all the people quoted in them and show the links between those people,” says Graham. “That’s something that used to take people working without this technology days or even weeks to do.”
Later, working with a large financial services client, the firm was able to produce a network map that included all the journalists writing about the company and all the individuals they were referencing in their stories, from company spokespeople to critics. Says Graham, “Often, reporters have their pet sources, people they talk to every time they write on a particular subject. We can tell a client, if you talk to this reporter, the chances are he is going to talk to these people too, so it might be helpful if you talk to them first, and make sure they understand your position or your strategy.”
The firm has also used its SNA to impress new business prospects. It can quickly determine the most influential commentators in an industry sector, for example. “The most influential are not necessarily the ones who write the most articles,” says Graham. “They are the ones whose coverage sets the tone, who are republished or quoted most often. We can identify those people, and we can show clients how a particular ‘meme’ spreads from one medium to another, and where it got started.”
Finally, located somewhere between the internal and the external, H&K has used social network analysis to examine the relationships between account staff and external contacts, particularly in the media.
For example, Cook recently sent an e-mail to others in the agency inviting them to preview the firm’s Reputation Watch research, which looks at high-level strategic issues in corporate reputation management around the world. Each recipient was invited to forward the e-mail to friends and contacts inside and outside the firm, inviting others to preview the same information.
“We found we were very good at inviting each other and not so good at inviting the outside world,” he says. “But it did allow us to identify people who are hubs of information, or who provide a bridge between the firm and the outside world, and those who are just dead-ends when it comes to that kind of information. We could see who were the most active and who were the most succesasful in term of connecting to people who actually viewed the information.”
If knowledge truly is power—and in public relations, that’s surely the case—social network analysis is a powerful new tool for the industry.