Friendships born at work and what becomes of them

Surveys on the origin of personal relationships regularly place work among the first contexts cited by employed adults, ahead of neighbourhood and voluntary activities.

That result is consistent with work on proximity: the workplace produces repeated, unchosen and prolonged encounters, a rare combination after the end of education. It also brings together people of varied ages and backgrounds, which few other adult contexts do.

Sociology nonetheless distinguishes the colleague from the friend. A cordial working relationship can last for years without ever leaving the professional setting, and surveys asking about the origin of friendships count only the fraction of those relationships that has crossed that boundary.

Longitudinal work shows that these relationships survive a change of employer less well than assumed, because the context sustaining them disappears at once. Shared topics, shared hours and occasions to cross paths all fade at the same time.

That asymmetry between ease of formation and later fragility is the most characteristic trait of this kind of tie, and it contributes to the measured turnover of personal networks.

The management literature documents the tensions specific to personal relationships crossing a hierarchical line: suspicion of favouritism, difficulty in evaluating, the cost of an unfavourable decision. Relationships between peers at the same level are described as less exposed, without being exempt.

Organisations handle this through recusal rules in recruitment and appraisal procedures, without the effectiveness of those rules having been the subject of published evaluations. Such rules aim at the fairness of decisions rather than at the relationships themselves, which remain beyond their reach.

Work on remote arrangements finds a reduction in unplanned interactions between teams, precisely those that produce the encounters behind weak ties: the conversation by the coffee machine, the shared walk to a meeting, the desk neighbour from another department.

An analysis published in 2021 by Longqi Yang and colleagues, using communication data from a large firm, measures a tightening of internal networks and a fall in exchanges between distant groups. It covers the period when remote work had been imposed on everyone, which sets it apart from an individual choice.

Measuring a fall in cross-cutting interactions says nothing about the quality of the relationships maintained, nor about what the people concerned think of it. Messaging data record exchanges, not the closeness felt between those who exchange them.

The same periods saw other forms of contact develop, and self-report surveys give contrasting results by sector and family configuration, which forbids a general conclusion.