All design involves manipulation. Crafting solutions to deliberately influence our decisions and actions. But is there a line that shouldn’t be crossed when it comes to using people in an experiment to test if their emotions can be altered?
Big data continues to be a hot topic and we are increasingly seeing data-driven decisions and processes replace expert opinions in everyday activities. Indeed, one designer quit Google with the following comment:
I had a recent debate over whether a border should be 3, 4 or 5 pixels wide, and was asked to prove my case. I can’t operate in an environment like that.
Trouble is, Google was able to prove that using data over instinct when deciding between 41 shades of blue for text-based links led to an annual increase of $200 million in advertising revenue. But whilst some decisions may be purely data-driven, most remain dependent on how the data is interpreted. And interpretation can be heavily influenced or manipulated by the environment, politics and language used.
A recent psychological research study showed that playing a game with a different avatar influenced behaviour afterwards. Those who played with the Superman avatar (context: hero saving the day) were kinder in later decisions than those who played with the Voldermort avatar (context: evil world destroyer).
In 2004, an experiment conducted at Stanford University (recently reported in The Atlantic) showed the influence of language on game play. Using the classic Prisoner’s Dilemma, one group were told they were playing ‘The Community Game’ and one group were told they were playing ‘The Wall Street Game’. Two-thirds of those playing ‘The Community Game’ chose to co-operate and share the rewards. Two-thirds of those playing ‘The Wall Street Game’ chose not to and focused on personal gain.
A simple shift in language can influence decisions and behaviour. Often with participants even realising. The subject of behavioural economics is not new. But combined with big data, its role in deliberately influencing decisions will continue to advance.
- Why Google has 200 million reasons to put engineers over designers – The Guardian, February 2014
- It matters which avatar you choose when gaming – Harvard Business Review, February 2014
- These two words will make you more selfish – The Atlantic, October 2013
- The Behavioural Insights Team – UK Cabinet Office web site
Flickr image: Optical illusion kindly shared by The Lex Talionis. It’s impossible to see both states of an optical illusion simultaneously. You have to make a choice about how you interpret what you think you see…
To be effective and productive in the modern world, organisations should not rely solely on hierarchical organisation charts to explain how work gets done. Networks help highlight individual contributions
The image above is a classic traditional organisational hierarchy. A manager responsible for making the decisions, supported by supervisors, each leading a team of people tasked with carrying out the decisions.
One of the biggest flaws within hierarchies is the tendency to treat all individuals at the each level as identical. In the example above, we have a decision maker, a group of supervisors: the blue dots A, B and C), and a group of do-ers: the red dots 1 to 9. (Yes I’m back with the pictures of dots again – goes with the name…) There are all sorts of challenges to the effectiveness of this system, not least trying to operate in an environment that doesn’t observe the rules of hierarchies and let’s everyone make decisions. But that’s for another post… This one is exploring how a network makes it easier to identify individual contributions and raise productivity.
Let’s rearrange our dots as the actual network that functions within this organisation:
The numbers and letters represent the order in which each individual was hired. The organisation chart does not tell us anything is different between the first person hired or the last. But the social network does.
Teams A and B are highly inter-connected. Team C is not. Supervisor C has barely any connections out of his/her reporting line and team. And we could guess that hires 7 ad 8 were made by C. All have come from outside the organisation and have yet to build up their network. The most useful member of Team C is no.9 because they have direct connections into both other teams and can more easily tap into their knowledge and expertise. But most interestingly, they have a connection with their manager’s manager. I’d guess No. 9 is heir-apparent to C’s job. C should be planning their next career move.
No. 1 is the longest-serving hire and well connected but not as well-connected as newer employees. Looking at the connections, the manager (purple dot) must be a more recent hire than no.1 because they have made connections with No. 4 and No.6 so they are not averse to communicating direct with team members yet do not interact with no.1. No. 1 is on the way out. Their career at this organisation has peaked.
No. 6’s career is on the up. Connected with people in all three teams, connected to all three supervisors and connected to the manager. Even if No. 6 knows nothing, they have access to everyone who knows something. The alternative scenario is that they are the person who knows everything, and everyone seeks them out when they need help. Either way, 6 is highly valuable to this organisation. Yet the organisation chart would suggest they are just a junior role.
Organisation charts make it far too easy to lump everybody into a single group – the level they are currently placed at within a hierarchy. And if you are near the bottom, you are expendable because the larger number of people at your level, the bigger the assumption that you are easy to replace. Imagine the company needs to reduce headcount due to financial difficulties? The common method in large organisations is to simply require all teams to reduce their headcount by the same amount. So teams A, B and C each lose a person, facing demoralising uncertainty and disruption in the process. The more productive approach would be to eliminate Team C, but keep no.9 and move them into one of the other teams. You’ve reduced the headcount by the same number, saved a bit more money because Supervisor C would likely have been on a higher salary and not impacted the two higher-performing teams in the process.
The image above visualises both approaches. On the left, the lowest performer from each team is removed. Look how sparse the connections now appear between the three teams. On the right, the lowest performing team is removed but the highest performer from within it is retained, and that isn’t the supervisor. The connections between the remaining individuals is much tighter, across the two teams. How much likelier is it that they will be able to support one another? None of this would have been evident by just studying an org chart that treats everyone at each level equally.
Of course this is a vast simplification of just one of the differences between networks and hierarchies. Hierarchies do have their benefits. They help us organise large volumes of information in ways that are simple to understand, making sense out of what would otherwise seem chaotic. But they do tend to create inequalities – it’s easier to reward the few at the top than acknowledge the many below – and their weakness is in failing to appreciate the messy realities about what is really going on. We are beginning to develop the tools to better understand and work within networks, enabling us to make sense out of the chaos without having to create a hierarchy in the process. Organisation’s that tap into this new found capability will out perform those that don’t.