Norm Erosion
I work at an engineering consultancy. An engineering consultancy sells project to clients and makes money by then billing per man-hour for the staff assigned to complete those projects. Some engineering consultancies are little more than body shops/temp agencies, providing cheap short-term staff to other firms and littler more than that. The higher end firms provide whole projects which the client never would have done on their own, either due to bad culture or lack of ability. The firm I work for is the latter kind.
One problem engineering consultancies often have is co-location. Clients usually want the very expensive consultants they pay for to work from their offices. The firm wants to please clients but it also want happy staff and a strong internal culture and collective identity. Such a culture/identity is hard to build if people are in different places and never see each other. Hence my firm chose very early on to commit to having staff work from HQ at least two days a week on Thursday and Friday. The logic was simple. The long term gains from a strong culture drastically outweigh the short term gains from slightly more contracts or happier clients. I say was because that logic no longer seems to hold.
A year ago or so, I was working on a large project. It was one of the largest projects in the firms history and at the time brought in around 20% of total revenue. In terms of team size, revenue and customer base, that one project was larger than most startups or product firms. At some point, senior stakeholders at the client demanded that we work from their office 4 days a week. Our management initially refused, then negotiated and finally agreed. When they did, there was a big fuss. Meetings were organized to understand our concerns. We were offered free breakfast and lunch on the extra day. There were talks in the firm about the effects this would have and weather the tradeoff was worth it. Most of all, everyone recognized that this was unusual, and would only be done in these kind of extraordinary circumstances.
Fast forward a year. I’ve just started on another project. It’s for one of the most well-funded startups in history. (30m seed). We work from the client site 4 days a week. There’s been no discussion of this or recognition that it’s abnormal.
It’s interesting just how quickly norms can erode. This is only one example, and one data point is not enough to infer a general trend, but it’s still interesting because it illustrates a few of the conditions which make norm erosion especially likely.
- Tangible, short term rewards for ignoring the norm ($$$ from having a major project extended) vs Intangible long term harms (weaker culture, less employee satisfaction, less networking)
- Individual gains, socialized costs. Various individuals would be directly responsible for the collapse of a major multi-million dollar project had they said no to a client. No one would be held personally responsible for gradual cultural erosion.
- Norm adherence -> "temporary, short term, situational" non-adherence -> general non-adherence. The norm didn’t snap in one day. Rather it gradually went from a norm to a sometimes followed norm to nothing. This is my impression of how most change happens.
The core problem issue is that most work is done by groups and one of the problems with groups is that they’re not aligned. What is best for an individual group member is not always what is best for the collective. Consider the scenario where
- We have a choice between A and B
- 95% of the time it is better for the group to choose A
- It is better for the individual making the choice to choose B
If the individual making the choice lacked any self interest, we could trust them to make the best choice in each situation, choosing A in the 95% of cases where it’s optimal and B in the other 5%. On the other hand, if they’re self interested, they’ll choose B as much as they can get away with. Norms are one way to prevent this. If choosing A is a norm, not doing it is obvious, attributable to one individual and usually requires justification. All of these things raise the cost of norm-breaking actions and so make them less likely.
I’m not sure what my conclusion is here. Maybe that it can seem rational to approach situations on a case by case basis rather than using general rules but it’s usually not unless you’re in a very aligned team.
Dissent 