Whirlpool Corporation has a very sophisticated talent management process, using traditional talent pool tools such as the 9-Box to rate and calibrate top talent. The 9-Box is a matrix that many organizations use to evaluate the talent pool. The x axis of the 9-Box assesses leadership performance and the y axis assesses leadership potential. Employees are assessed by a committee or group as to where they fall on the matrix of high to low potential and high to low performance. The rating is based on the assessment by a group of more senior leaders based on hierarchical interaction. In my experience, the 9-box process tends to identify and assess the “usual suspects” of leaders that are on the short list of high potentials as assessed by the senior leaders. One problem is that senior leaders only have snippets of observation and often miss novel or renegade leaders that are held in high esteem by their peers. The 9-Box was made famous by GE and is seen by many talent managers as the preeminent talent assessment tool. We began to postulate additional ways to identify top talent, especially in a world of social networks. To test our theory, we conducted research on the social networks in twenty of our top strategic areas. Our findings were startling compared to the top talent rated by our traditional means. There was little correlation between those individuals who rated as central to the social clusters, the knowledge brokers, and those rated as top talents through the 9-Box. It made us curious about identifying talent and why those who seem to not only make things work, but who simulate social interactions against key strategic issues, are not more represented in the talent management results. Social networks may not replace 9-Box or other traditional methods but it does provide another lens to talent.

Context

Whirlpool is a hundred-year-old company that markets and sells appliances throughout the world with 70,000 employees. It is the world’s largest appliance company with sales of $18B. In the last ten years, Whirlpool has survived the war for talent by adopting leading edge tools and disciplines in talent management. Yet, even with such advanced talent systems, Whirlpool found that a continuous flow of adequate talent is an ongoing strategic issue.

Triggers

The trigger was an experiment we conducted in a leadership development workshop. Our hypothesis was that if the 9-Box is the primary tool in any organization to assess leaders, it misses real-time performance of some leaders as perceived by their peers and subordinates. Inherent in the hypothesis is that a 360-degree tool that observes leader performance by a crowd of people who work with the leader in real time in every imaginable context, often outside the eyes of most senior evaluators, is an additional talent assessment that sheds new light on talent in organizations. This 360-degree tool can be elicited from social network assessments. We were seeing increasing importance of social networks in the workplace and wondered if we could carve out a unit of analysis that was not hierarchical or identified by an organization chart. As part of the strategic planning process, we asked the executive committee to identify the top twenty pressing problems of the enterprise and to assign a multi-level, cross function and regional team to address the problem. Examples of the perennial problems included: Creating a fabric care blue ocean; First launch value creation; Post-launch value creation; Regional business unit competitive gaming; Global Water Business Strategy; Attracting and developing top talent; China strategy 2008–2012; Home improvement sector blue ocean; Consumables business growth; Advancing supplier relationships; Cash generation/Next Gen; Material cost productivity; Freight and warehousing global strategy; Total lean enterprise upgrade. We designed a leadership development workshop/work-out to address these pressing problems. Some of the teams had worked together and had a history; some of the teams were newly formed for this event. For the existing teams there were varying levels of performance from needs-improving to take the strategy to the next level. Our criteria were that many of the team would have day-to-day accountability for the problems and that some leaders, who were outside the problem, were also assigned. We looked at global problem solving teams assigned to the twenty top strategic issues in our company, and we wanted to see what a sociogram or social network map would show us about how work really gets accomplished. A sociogram is a graphic representation of all of the links a person has within a context. It is created by asking the people in a social network a series of questions about who and how they interact with everyone else in the network. The sociogram consists of dots that are people and lines that are the interactions. Often the length to the line will convey the quality of a relationship, the direction of the line will convey the direction of communication and dots that have a significant number of lines leading to them are often knowledge brokers within a social network. Links in the social network are based on the amount of time, emotional intensity, mutual confiding and reciprocity. Depending on how social network information is collected, it can also show aspects of the relationship that include: quality of the interaction, the flow of information, the introduction of a new idea and how it migrates through a group. The sociograms quickly surface the knowledge brokers—individuals who are the glue that holds a social network together. Knowledge brokers can usually get to anyone in an average size network in two to three steps. They also represent the fragility of the network; if they would leave, a hole would be created. In most cases, knowledge brokers are good, but there are times when they are hoarders of information or power brokers. In this case, we postulated they were knowledge brokers and wondered how they were rated in the talent pool 9-Box. Our hypothesis is that most of the knowledge brokers were not rated high the 9-Box, if they were considered at all. By contrast, the sociogram on the right is the post launch teams. What stands out more than anything is that the cells that launch products are not communicating with one another? Before the teams came to the workshop, we sent a questionnaire to them with twenty questions. The survey and sociogram was conducted and completed by Raytheon. The questions included items about who you connected with the most, if that connection was good or a strained relationship, who you share new ideas or innovation with, and how long you have known these people. It took two to three weeks for Raytheon to process the questionnaire. On the first day of the workshop, the team was exposed to their sociogram with the help of an expert in this field, Holly Raider from The University of Chicago. In almost every case, the teams were intrigued with their sociogram and found new learning about how they relate. For example, the Freight and Warehouse team is divided into three locations. There were very few connections across locations. They worked in building silos. This was a new insight to them and they spent considerable time resolving. The post launch team, as shown above, realized they did not share best practices. As a result, they set up a best practice sharing form for all post launch teams. They also insured that incumbent team members join new teams to impart tribal knowledge. What we saw was something far more intriguing. We conducted the analysis on twenty teams and over 500 senior leaders, and we started to see a view of talent not seen before. This study was insufficient to give us key “peer reviewed” results but it did show us promise. We asked the head of the talent pool system to compare who was ranked at the top of the 9-Box with our list of key knowledge brokers. We found very little overlap. We found that many of the key knowledge brokers (good ones) or nodes in each social network rarely showed up on official talent pool lists of top talents.

Key Innovations & Timeline

We are just starting to remodel our leadership and talent model to include social networks as a key development mechanism. We need more concrete data to make the adjustments to our process but we are in the process of comparing the talent pool results to the sociogram results. We believe that both tools will be important but at this stage, we are in the beginning phases of designing sociograms into our talent process. The head of Human Resources and the heads of training are working to develop the structure and processes that will use social networking as a way to identify experiences, feedback, mentoring, training and development. At present, we are also researching to see how sociograms are used in leadership development. Many definitions of leadership center on the ability to influence a group. Networks are based on social influence and therefore, may be at the heart of leadership. For example, there are four key types of social networks in organizations: Communication, Task, Friendship, Advice or Mentoring. There is also some research that suggests that looking at social networks in the context of leadership development can help organizations with their talent pool: Smith (2006) found that individuals who engaged in behaviors leading to the creating of bonding types of social capital experienced higher levels of organization commitment and job satisfaction as well as higher performance ratings. Burt (2005) found that individuals who span more holes between networks receive better performance ratings, higher salaries, and larger bonuses, and find better jobs, faster. They are based on getting better information, better timing and better referrals. Spanning social networks provides individuals with a vision of options that would otherwise be unseen (Burt, 2005).

Challenges & Solutions

We need internal mechanisms, or more cost effective ways to create the sociograms. We used an external source to create the ordinal studies but outsourcing is an expensive alternative. There are many challenges in conducting the sociograms. While the expense and newness of the field for leadership development present challenges, there are more intangible ethical issues. For example, in one of our sociograms, there were people that no one identified as a partner. They were an island on the chart. We chose not to show the names on the overall chart but gave each person their feedback individually so as not to embarrass anyone. We also had to do some detective work to figure out if the knowledge broker was good or bad for the team. In one case, we found that the person was a knowledge hoarder. We did this through a series of interviews, not as scientific as the sociogram. Additionally, if we decide to use the sociogram in leadership development, we will need trained coaches. We will also need to decide how the individual data is managed, who sees it and what it means. We are also conducting a literature search to see what the research tells us about social networks in identification and development of talent.

Benefits & Metrics

Using a peer review or crowd sourcing lens in addition to the traditional top-down lens would provide an alternative to assessing talent. In addition to a 9-Box rating and feedback, individuals could receive sociogram feedback about how they were rated by their social network and the role they play in the network. The network could rate their contribution, much the way a supervisor rates a subordinate’s performance.

Lessons

The lesson is to experiment in unconventional spaces with new tools and to re-assess and ask what it means for the conventional tools you are using. Another lesson is to sometimes let new ideas percolate and see where they lead you.

Credits

Nancy Tennant, Tamara Patrick-Whirlpool, Kimberly Thompson-Whirlpool, Holly Raider—University of Chicago

Materials

Materials: Sociograms http://en.wikipedia.org/wiki/Sociogram 9-Box http://hrweb.mit.edu/system/files/9-Box+Grid_0.pdf Resources: Raytheon Social Network Analysis: http://www.raytheon.com/businesses/stellent/groups/rps/documents/content... Key References Social capital and intentional change; exploring the role of social networks on individual change efforts. M. L. Smith. The Journal of Management Development. 2006. Vol. 25, Iss.7; pg 718. A Role Based perspective on leadership as a network of relationships. A.G. Sheard, A.P. Kakabadse. The Journal of Management Development. Bradford. 2007. Vol. 26, Iss.4: pg. 331. Burt, R.S. (2005). Brokerage and Closure: An Introduction to Social Capital. Oxford University Press, New York, NY. Higgins M.C. and Kram, K. E. 2001. Re-conceptualizing mentoring at work: A developmental network perspective. Academy of Management Review, Vol .26, pp 264-88.

You need to register in order to submit a comment.

Sort By: Latest Oldest Likes Replies

November 27, 2012 at 9:52am

Marie Wallace

Great story Nancy. Really enjoyed seeing the practical application of theories that I work on in the lab :-)

December 1, 2012 at 5:35pm

Yin Cheng

Hi Nancy,

Wonderful experiment!

I am wondering if it is useful to differentiate knowledge brokers vs employees with capacity to take on bigger roles in the future.

If goal of talent management is to identify who today can take on bigger jobs in the future given the necessary development and experience then perhaps the key questions are to post to the social network are
- who in your group do you see as potential senior management if given the necessary development and experience.
- what make you say so? Give details of person in action in your group.
- For these individuals, where do you think the organisation can next rotate him to deal with challenges that the organisation face?

It would be interesting to see how results from above correlate with 9 box results.

Do drop me an email if you wish discuss further. Will be happy to exchange ideas.

Cheers!

June 10, 2010 at 1:39am

Matt Shlosberg

Nancy!

Thank you for posting this great idea!

Some questions/comments and food for thought:

1. This idea reminds me of Gary Hamel's Natural Leadership Meter. He uses a slightly different approach but it seems like he is trying to solve a similar problem. Take a look at his proposed solution:

http://www.managementexchange.com/content/leader-meter-finding-natural-l...

2. You've mentioned that you need to find an internal cost effective way to process sociograms. I've seen a couple of tools out there in the semantic technology field that will allow you to do just that. Hewlett Packard's Jena, IntelliDimension, and Oracle's semantic storage (RDF) are the first two that come to mind. They are able to process graphs like the one used by sociograms. Also, it seems like it takes a long time for Raytheon to produce them. In today's global war for talent and real time social networks, it may become critical to be able to process this data in real time rather than wait a week. I realize I may be making a big deal out of something small, but I thought I'd give it to you as food for thought.

3. There's a study out there that talks about bias in the talent selection process. It identified that employers are biased when they look at candidates' resumes and ask about previous experience. The study showed that prior performance isn't always a good predictor of future performance due to factors outside of the candidate's control, such as the difference in the organizational culture, education, opportunity to showcase skills and make decisions, etc. My question is, how do eliminate bias when using your system? For example, it seems to me that your approach may be biased towards natural leaders. What about people who aren't natural but can be built? Is there a bias towards extroverts rather than introverts in the fact that you look for people with certain connections or communications styles? How do you know that people selected by your system will have other key leadership competencies, such as integrity, ability to coach, and self-awareness? Also, is there an assumption here that naturally built social connections already exist in your organization and they can point to the source of talent? What if you have people that haven't had a chance to showcase their skills or haven't had a chance to do so? Also, I wonder how much of these sociograms are produced due to the organizational culture vs. people's competence?

Overall, I think it's a wonderful idea. It would be interesting to see how you can improve it further by eliminating biases. Also, I wonder if you still leave a lot of untapped talent out there by using this framework. Are there any hidden gems in your organization that this framework can't identify, such as those who haven't completed the questionnaire or those that haven't been given a chance to really showcase themselves? It seems like Raytheon's model is to show social networks within your organization, not to identify the source of talent.

Thoughts?

April 16, 2010 at 12:18pm

John Chesen

This analysis raises the question: are there two types of leaders in a dynamic and innovative organization, the type identified by the 9-Box and the type of "super connectors" identified by analysis of social networks? Both types are essential for different reasons, although the personality and leadership profiles of each are significantly different (as suggested by low correlation in your analysis). If two leadership types exist and both are essential then each group should be nurtured in different ways because each probably has, on average, a different intrinsic motivation vs. extrinsic motivation mix and responds to different incentives.

November 17, 2010 at 4:03am

Stefan Lindegaard

Hi Nancy,

Great to see a company that looks into the role of "informal" leaders...

I can imagine that you will have some challenges having this approach gain broader acceptance at Whirlpool, but I think the times are with you. Once we better understand the value of a strong networking culture and it's impact on innovation and beyond, this kind of work will be highly appreciated.

Stefan

November 17, 2010 at 5:24pm

Ellen Weber

Very interesting approach to unleashing and developing talent Nancy, and thanks for the wonderful insights here! Wow - Your post adds hope that people and talents can refresh tired and broken systems.

I’d like to hear a bit more about your “ detective work to figure out if the knowledge broker was good or bad for the team.”

Would this interaction not also be a delightful segue into building the criteria together so that all helped to create criteria to follow, and all helped to remain accountable to that criteria and to share and learn from their efforts to do so.

Your statement, “ … we found that one person was a knowledge hoarder,” reminds me of the way we were taught, and the way many still operate for what they see as competitive advantage. What motivation could be offered to help workers move from hoarding to sharing in a trusted setting?

Would you agree that qualitative research is as valid as numeric quantitative studies – when operated with normed reliability and validity, and negotiated standards? Yet you concluded that “We did this through a series of interviews, not as scientific as the sociogram.” Could you elaborate a bit more on why that was so?

In my work to certify leaders in brainpowered strategies for leading innovation, I find that mentors become learners at times and learners also mentor as they learn. Have you found that to be the case as you progressed with talent development?

Again, thanks Nancy, for the fascinating and engaging work. I will be deeply interested in your continued progress. Thanks for sharing!

December 9, 2010 at 11:57am

Joshua Letourneau

Nancy, it's great to see Whirlpool executing SNA to this degree, particularly with a focus on overall performance.

Building the Sociograms doesn't have to be expensive - after all, you're seeking "subjective awareness", not a magic bullet.

In 3 to 6 months, you can run another graph and see how your interventions are (or are not) working.

We should talk!

Josh Letourneau
jl(at)knightbishop.com
www.KnightBishop.com

January 20, 2012 at 5:56am

Tony Nelson

Hi Nancy The search for people who really add value in an organisation the size of yours is vitally important. Your innovative use of different lenses sounds like it may reap rewards in time. You might want to check out Prof Lynda Gratton's work on boundary spanners (see 'Hot Spots' Prentice Hall) who use social networks in unfamiliar settings to create new knowledge and innovation.

January 24, 2012 at 2:51am

Stephen Remedios

Dear Nancy, I like the point you make about the different networks at the work place and need to harness all of them to ensure holistic leadership development! I'm personally of the view that we are headed into an era of unprecedented networking and in that context I couldn't agree more with a number of things you have said in your wonderful story.