Analytics to Identify and Strengthen Organizational Collaboration and Drive Innovation

Companies with many employees face various challenges w.r.t. their size. One of these challenges is to identify key people, skills and information across (and outside) the organization and use them in the most effective way to drive innovation, new initiatives, but also sales.

Situation Today

The natural way for employees to cope with such a challenge is to build social networks (I mean networks of people not software) and collaborate across the organization. This organic social network building happens through various activities such as projects with colleagues, social activities, company events, etc… However, such social networks take time to create, and are typically not that extended. The effectiveness of these social networks is hard to measure (unless perhaps you equip your employees with location trackers). This social network building, I call the qualitative approach to organizational collaboration.

Many companies have deployed technology solutions (tools) to cope with the challenge described above. Companies have personnel directories that show employees’ groups and the official organization hierarchy. Some of these personnel directories allow users to add more personalized information (but this information is not always up to date). More sophisticated personnel directories (or other collaboration tools) also feature timelines of activities/tasks, blogging, integrated search, etc…. .Video conferencing enables people to connect people remotely. Despite these new social tools, email and mailing lists still play an important role in connecting people and disseminate information as well as external social networks and resources. All these tools provide a wealth of information. In essence: collaboration is Big Data.

Do all these social/collaborative enterprise tools help us doing our job better and promoting innovation? From a personal point of view I am tempted to say yes, but much more can be done. My main concern with most of these tools is the lack of analytics features to quickly identify user-relevant information or contacts. New tools – and newer versions of already existing tools – are starting to provide some of these capabilities, but IMO that is still not enough (or not accurate enough) to fully understand the evolving social networks or the relations between people and information (documents, emails, etc…).

What can be Improved

The goal of exposing more of the right analytics to end users would be, for a user, to faster gather insights, new ideas, and enable quicker decision making and eventually translate these insights and ideas to new opportunities, projects and/or costs savings.

To achieve this goal, users should be able to identify patterns in their organization’s data, specifically on threads or evolving thoughts and interactions that can be relevant for their particular projects or questions. In essence, analytics should foster more and improved collaboration with like-minded people, or people that share a common goal. As mentioned earlier, people naturally do this already, but in large organizations it is humanly impossible to scale this effectively and fast, without the help of analytics tools. This type of analytics I call the quantitative approach of organizational collaboration, which I see as complementary to the qualitative approach.

When I look for example at mailing lists or video conferencing, a few questions always pop up in my mind that modern enterprise collaboration tools should be able to answer in just a few clicks:

1. What topics are trending during the last week/month? Perhaps type a topic in and get trend information or have the computer generate topics based on a context analysis of your posts or email conversations.
2. How are the groups and hierarchies evolving over time (who is talking to who)? Can software recommend groups of people that are relevant for me and my projects?
3. What people can be considered as experts on certain topics, based on their posts, replies, published articles, etc…?
4. For particular topics, who are the top contributors and how do they relate to the experts? Are people clustering around certain topics?
5. Who are the influencers/thought leaders, and how do they relate to experts?

From a strategic point of view companies can leverage analytics from social/collaborative tools to answer questions like:

1. Are best practices shared across the organization between the appropriate groups?
2. Is there an alignment between strategy and direction of the company?
   

This is not an exhaustive list and as a software engineer I think that an additional relevant feature for any tool should be the ability to provide an environment to mashup and integrate data by employees, to answer some of these questions.

How can it be Leveraged

Various groups and people (MIT, Virginia Tech, …) do research on this subject and translate this research into strategic insights at the enterprise level. The next step will be to provide the insights to individual employees as well. Enterprise tools with more sophisticated analytics capabilities (many focused on machine learning) are beginning to emerge. Perhaps the biggest challenge is integration of such capabilities across multiple internal and external tools and platforms.

Organizational collaboration is for me not limited to an enterprise environment. Groups with different affiliations who organize themselves as “virtual organizations” to work together towards common goals (for example, Open Source communities or standard bodies) can benefit from this type of analytics too.

To be more successful in capturing the value of collaboration, companies not only need to deploy the right tools, but also need to foster a mashup environment to leverage the organizational insight and tacit knowledge of its employees through analytics.

Special thanks to Marco Valente and Yannik Messerli for the discussions and insight on this subject.