Leonardi & Treem: Understanding the barriers to success and benefits of enterprise social software
I just discovered some work by Paul Leonardi, a professor of communication at Northwestern University with an appointment at the Kellogg School in the US, who has been researching what he calls 'Enterprise Social Media' (ESM).
Investigating the implementation of a enterprise social network, called A-Life, at a credit card company, he found that after six months those who had used the enterprise social networking site had:
- improved their ability to find information by 31%.
- improved their ability to find people who knew the person with information by 71%.
These impressive statistics reflect what Leonardi considers to be a low starting point but highlights the impact social tools can have in an organisation when they fill a collaborative vacuum. Perhaps also counter intuitively (but not surprising), he also found that younger employees "were generally more skeptical of the tool."
A-Life was based on Jive Software. Jeffery Treem, a co-researcher with Leonardi, provides more detail in this presentation (PDF) at Collaborative Organizations and Social Media 2013 about A-Life from the perspective of understanding the barriers to use of 16 people in the marketing division (out of a total user base of 15,000).
They interviewed the group before and after the introduction of A-Life - overall participation by this group in the first five months was actually very low. Treem associates this resistance to three factors:
- Accountability as a Professional.
- Accountability to an Unknown Audience (the perceived risk of being mis-understood - see below, about the 'leaky pipe' metaphor).
- Accountability (relevance) to Organizational Role.
One of the key lessons from this case study - which I've also see in my own experience over the years - is the importance of:
- Not just using change management in an implementation, but actually allowing future users to participate in the design and selection of the solution. Organisational culture can otherwise eat the virality of social tools.
- Tools must be integrated at a workgroup and team level into local work practices and nomenclature (i.e. talk about KM or collaboration if people are uncomfortable talking about social media).
This applies as much to ESM today as it did to KM and groupware implementations in the past.
Moving back into theory, in a literature review (PDF), Leonardi, defines ESM and calls out two key distinctions from past technologies - visibility and persistence:
"there are at least two affordances provided by enterprise social media that make them distinct from other communication technologies commonly used in organizations: They provide people visibility into the communicative actions of others and the visible traces of those communicative actions persist over time. Because ESM afford the visibility and persistence of communicative actions, they expand the range of people, networks, and texts from whom people can learn across the organization. Consequently, one of, if not the most important, outcomes of these affordances for organizations is increased opportunities for social learning."
Leonardi and Treem, define these and two other 'affordances' in more detail:
- Visibility - Allows workers across levels to broadcast messages throughout an organization.
- Persistence - Displays information across time so that it may be confronted and used by individuals who are not present for the original act of communication.
- Editability - Ability of actors to spend time crafting and tailoring a communicative act before entering it into a technology and making it available for others to view.
- Association - Ability to establish links or relationships between individuals or content.
If you are familiar with Andrew McAfee's SLATES model, this is similar but I think the concept of Persistence is a useful addition - this is particularly relevant when considering the overlapping role of ESM as a system of engagement and a system of record. The one thing I think missing, which I talk about in my equivalent model of 'affordances' is the idea of using ESMs to gain insights & analysis based on activity and information in the network, however Leonardi does recognise the opportunity for social analytics (see below).
He outlines a number of dimensions - as potential areas for academic study - that he categorises using the following metaphors:
- Leaky pipe.
- Echo chamber.
- Social lubricant.
These metaphors can be applied to four business processes:
- Social capital.
- Boundary work.
- Attention allocation.
- Social analytics.
He provides a matrix of proposed advantages and disadvantages for each.
I quite like Leonardi's idea that enterprise social software creates "leaky pipes for organizational communication". However, the literature review suggests some challenges with that model:
- A tendency towards communicating more abstract information (so meaning is reduced).
- The risk of increasing information overload through greater visibility.
Personally I think there are some limitations in these viewpoints because it focuses on direct information sharing rather than indirect opportunities that functionality such as rich user profiles, group memberships and recommendations provide. A good enterprise social software platform should also provide users with tools to help them filter the volume of activity that is visible to them.
In another paper (PDF), he writes that:
"With the swift development of new communication technologies the particular social media we use today are not likely to be the ones we use in the future. We have argued, herein, that an affordance approach, which focuses attention not on any particular technology, but on the types of communicative practices that various features afford, is much more likely to have staying power because it builds theory about the relationship between technology and communication without foregrounding one concept or the other."
I can't agree more here. I've always tried to talk about social media (inside and outside organisation) using pattern language that moves with the technology, rather than being bounded by it. Pattern language also allows me to link capabilities to tasks and business processes, rather than focusing on making organisations social for their own sake.
Overall, I think there is a useful body of work forming here that contributes to how we implement enterprise social software in practice.