I’ve been listening to some amazing podcasts recently from BBC Radio 4 called In Our Time.
One very interesting podcast was on the ideas and development of game theory. It was fascinating (as always) and led me to wonder if game theory could be applied to digital development in some way.
My thinking took me to the Prisoner’s Dilemma. It’s a fascinating problem where two people are held in separate rooms and questioned over a crime and it is assumed they cannot cooperate. If both keep quiet, they’ll get lesser sentences (on a lesser charge). If one confesses they’ll go free, and the other will get a heavy sentence. If both confess, they’ll both get a medium sentence.
The best approach? It depends on the lightness or heaviness of the sentence. Simply put, the most pragmatic solution is to confess, because you either get to go free (the other one kept quiet) or you both get the medium sentence, which is better than the outcomes of keeping quiet.
How does this relate to digital development?
For many years, I’ve seen very clever digital developers sitting on the sidelines of a project, where the client has decided to take a “safe” option. This option often is more expensive than it could have been, but is not as expensive as going down other routes, such as contractors.
There is an immediate parallel then. Clients have a choice as to who they use. They could
- use in-house and contractor resources which is expensive due to management and day rate, but often gets bogged down in audit trails and project management. Has a much larger cost in terms of maintenance
- use an agency which is often less expensive than the above, but you pay for extraneous elements that don’t necessarily add value. Has a medium maintenance cost, but the company can provide this as well (for a fee)
- use a bespoke developer or small dev house to deliver something, which often has lower costs than agencies and lower maintenance costs as well, but smaller means more risk if they disappear
When websites were becoming important (early 2000s) agencies became the place to work, because the quality of bespoke developers was too random. When the ability to develop these websites became straight forward, then the weighting moved back towards bespoke.
If you had the money, you’d go in-house or enterprise just to make sure it got done. They don’t really matter here.
Then the idea of apps came into being. Again, we had a problem in that the bespoke developers were not strong enough on the whole, so agencies were the only way to really go.
Now apps are much more straight forward and the standard of bespoke developers has been raised, it’s more weighted back to bespoke developers.
Game theory and specifically the prisoner’s dilemma is about making a decision based on the best choice/outcome relationship.
What I’ve seen is that the choice is often not based on whether or not the bespoke developers can deliver, but on other factors. In fact, the right choices change over time in the scenario above depending on perceived quality of freelance developers, newness of the technology, abundance of information around the subject…
It’s the idea of “No one ever got fired for buying IBM”. Game theory is a fascinating thing to play around with and the digital development world is often “Why don’t clients just go with the best developers who often aren’t in the agencies?”
The reason is simple. The risk is too great a lot of the time. Some companies will go for it, especially if they have limited budgets (think startups) but most will lower the risk, whilst knowing that they will not get the absolute best at the end of the day.
It might not be the best, but it might be good enough for them not to get fired.