Advancing the Debate on the Open Plan Office

When a new piece of research hits the workplace community, you know pretty quickly as everyone starts talking about it, and clients email us asking for our views. The latest headlines are are in regard to some research on the open plan office.

The research recently published by Ethan Bernstein and Stephen Turban, ‘The impact of the ‘open’ workspace on human collaboration’ certainly made fascinating reading and solidly reinforced our views about what is important when implementing workplace change. Naturally we can only go by what is published, but it seemed to us that there are some interesting aspects that were either not addressed or weren’t covered in the paper and we’d like to raise these here.

What are you trying to achieve from an open plan office?

Probably due to confidentiality, there is no in depth understanding of what either organisation was seeking to achieve through increased collaboration between members of staff and teams. There has for some time been a general unspecified notion that “collaboration is best” and workplace designers have avidly provided spaces for people to collaborate within. When talking to users, they are often baffled as to what it is they are supposed to be collaborating on? If there is a solid reason and you can explain it to them, then all is well. If the general message is “we want people to work with each other more”, then don’t be surprised if nothing changes just because you change the layout.

Things need to make sense to people and if they don’t, then they are bemused and carry on doing what they’ve always done. So, for AWA, the clear lesson is to explain the change that you want people to make, in a way that makes sense in terms of what they are trying to do in their jobs and what the organisation is seeking to achieve.

Does removing the walls deliver collaboration?

The next troubling aspect is that the office design described in the research studies seemed to take the simplistic view that people will collaborate if you take the walls and barriers away. The real thing that happens when you take walls away and implement an open plan office, is that people are able to collaborate more easily if they choose to. They have better visibility of other people, get a lot more contextual information about what is going on and can see when colleagues are available.

However, we all know that most people have a range of tasks to fulfill, and that simply providing them with a desk in a sea of other desks is unlikely to meet their myriad work needs. Depending upon their role, people may need quiet workplace conditions in which to concentrate, or spaces to work informally as well as formally with colleagues. They need great places where they can work seamlessly with people both in the same location and others in different locations by using collaboration technology. A well-designed workplace also provides opportunities to bump into colleagues and catch up; to see people they might not otherwise see; and to learn something about what others are doing.

Assigned seating but no other places to work?

In the case studies quoted by the Bernstein & Turban research, staff continued to have an assigned desk in the new regime – something that many people don’t have when they transition to a different way of working (e.g. – activity based working). That implies that there may have been less space available to provide other work settings for different tasks. We can’t be certain, but it is stated that the open plan was delivered “at the expense of privacy and concentration”, so it is a reasonable assumption that there wasn’t a range of other places for people to use. And because they had an assigned desk, they effectively had no choice about where they were working each day. Our experience shows that choice is an important part of the buy-in to new ways of working.

In fact, there is no information about the lived experience of people working in the environments studied in the research – all we know is what volume of interactions people had with other people who had volunteered to participate in the study.

What about interactions with people outside the study group?

This brings us to another point – the participants represented 40%/45% of the workplace population respectively in the 2 studies. What does this tell us about the rest of the population?  The researchers said there was no evidence of bias between the participants and the rest of the population – but surely there is a difference simply in the fact that some are willing to participate, and others aren’t – whether that shows a positive bias towards the study or otherwise?

Further, there is no mention of the “Hawthorne effect” whereby the act of being observed changes the behaviour of those being observed. Perhaps it was felt that over time people would forget they were wearing the sociometric badges?

The findings of the studies are based on the interactions between participants – so interactions between participants and non-participants were not captured. As such this doesn’t show the totality of interactions between all the staff – only those in the study (both before and after the change was implemented). There are many explanations as to why people might not be interacting so often with other study participants after the change, but what is troubling is that the target was to increase the quantity but not the quality of interactions.

And who’s to say that the number before the change was the ‘desirable’ level? If the number of those interactions was higher or the interactions were longer – does that imply that they were superior in nature and outcome to shorter, less frequent interactions? Could it be that people got to know each other better and then relied on IM and email more often?

Quantity vs quality of interactions

We feel that the lack of information about the quality of interactions is the biggest area of concern. By looking only at the quantity, it forces us to conclude that the number is the important thing – without having any idea whether the discussions / interactions achieved any goal or provided each participant with what they needed.

The researchers state firmly that face to face (F2F) networks and email networks respond differently to changes in the built environment – with one change failing to predict changes in the other. This raises another question which is around the things that impact behaviour. There is a clear direction from the research conclusions that the built environment (the open plan office) was directly responsible for a reduction in F2F interactions (something assumed to be bad). No account appears to be taken of other things that might have been going on within the organisations.

Senior management in the study companies reported that according to their performance management systems, performance had reduced, and this is entirely laid at the door of the open plan office! What were these measures? Could the reduction be attributed to other factors? What caused the researchers to conclude that there was a causal link? There’s no suggestion that anything else had an impact on the reduction in performance. Really?

To be honest, if the changes were made without a significant and successful change management support programme, it’s reasonable to anticipate at the very least that staff felt that their psychological contracts had been ripped up.

And finally

In pursuit of the best use of workplace resources and the best workplace experience for the occupants, neither traditional cellular offices or traditional open plan spaces provide a productive environment. The move towards a mix of spaces designed to cater for a range of tasks and preferences (often called agile or activity-based working) offers more opportunities to cater for both solitary and collaborative working styles. That said, whatever you move to, there must be a change programme designed to identify what new working practices and behaviours are desired, to deliver what the organisation is seeking to achieve. Old habits don’t die easily, so people need support to get the most from the opportunity.

In conclusion, our view is that the Bernstein & Turban research provides some really interesting food for thought, particularly making use of technologies that haven’t been available in the past. It opens up lots of potential for detailed discussion about what is required in terms of workplace design (specifically the open plan office) and change support if the much sought after additional collaboration is to be effectively delivered – and that is NEVER done just by taking down walls.