A recently-published research paper concludes that relationships based on general networking are of no real use when it comes to solving complex problems in-company.

Assessing the Limits of Networking for in-Company Collaborative Work

Reaching out to people through the social networks, whether professionally-oriented or not, seems today to have become essential for anyone who wishes to create or extend his/her own personal circle, whatever the nature of that group of people. In the professional sphere LinkedIn claims top spot with over 300 million users, two thirds of whom live outside the United States. This network among others serves as a platform for interacting on particular topics and is variously used for such purposes as enhancing your e-reputation, finding a new job, or recruiting from among existing staff.  This approach has its attractions, but is not effective in all circumstances. Now a study recently carried out by a research team at the prestigious MIT Media Lab – entitled ‘The Strength of the Strongest Ties in Collaborative Problem Solving’ – purports to demonstrate that general networking does not improve team performance in the context of collaborative work. The study uses Mark Granovetter’s definition of ‘ties’ – both ‘strong’ and ‘weak’ – from his work ‘The Strength of Weak Ties’. Following their experiment to assess collaborative performance, the MIT Media Lab researchers conclude that only genuinely ‘strong ties’ enable the high quality collaborative work required to solve a complex problem.

Broad networking useful for sharing information

The experiment was carried out among 80 MIT students. At an early stage participants were asked to rate how well they knew all the other participants. The standard grading scale (0–12) was used, ranging from: 0 = ‘I do not know this person’ to 12 = ‘One of my best friends’. If a student gave another in the group a 12, this was taken to mean that they felt they had a ‘strong tie’ with that person.  In fact 95% of the assessments made were at less than 12, and were deemed to indicate ‘weak ties’. Yves-Alexandre de Montjoye, a member of the MIT Media Lab who co-authored the paper, points out that having a large number of weak ties “does not demand any effort”, fosters information penetration from other networks and can be very useful in certain circumstances. For instance, if a person is looking for new employment, having a network consisting of many weak ties tends to lead to the sort of information exchange – such as forwarding a job posting by email – which carries little or no cost to the people sharing the information.  De Montjoye stresses that such initiatives as “organising huge work events does no more than create weak ties” and claims that creating such weak ties – which are mainly based not on real friendship or genuine affinity but simply on time spent together – “does nothing to help teamwork.”

Only ‘strong ties’ lead to better team performance

When it comes to high quality teamwork on a complex subject “only the 'strong ties' count”, argues Yves-Alexandre de Montjoye. And, according to the criteria of the study, such ties only exist on social networks between very good friends. The Media Lab man believes that the most useful part of the study just conducted is that it reveals the non-linear nature of performance in relation to ties: only strong links – as defined in the paper – are reliable predictors of good collaborative performance. Transposing the impact of these ties to an ‘in-company’ situation, the researchers came to the conclusion that when a project requires a high level of team-member involvement, such as listening long and hard to someone, with a view to collectively solving a complex problem in science, engineering or business, only the strength of the relationships counts as a predictor of success. The performance of a team on a given piece of work can only be correlated with ties that are deemed to be strong.  But then, if the key to better collaborative work depends on ‘strong ties’, why not apply this principle to the management process? This can indeed be done in smaller organisations, confirms de Montjoye. Moreover, strong links will only be newly forged when colleagues listen to and support each other, actions which in themselves help to ensure better overall results.

By Arthur de Villemandy