Why game theory decides where your Saturday night takeaway comes from
by David Dwyer on 03/10/2019 102 Reads
Over the past few years, Inspire has been approached by a number of takeaway businesses from the same geographic territory.
They all wanted the same thing: to increase the number of orders they can fulfil.
But they all had the same problem: Just Eat, the online platform that allows customers to search for, order and pay for, takeaway meals, but does not:
Had moved into the area and is now gobbling their profit.
And like most, if not all, disruptive tech entities, Just Eat, Scoffable, Deliveroo, UberEats have an inherent competitive advantage in the “local territory near you” marketplace – one not based on higher cost-effectiveness or better programming, or online marketing, but on the quirks of human nature.
Just Eat et al’s advantage is based on the tendency for otherwise rational humans not to cooperate, even when it might be in their best interest to do so. The interesting aspect of this is that the scenario is actually heavily based in Economics, Mathematics and Game Theory.
John Nash's "Nash Equilibrium", (ED: you might recall the name as he was portrayed by Russell Crowe in the movie "The Beautiful Mind"), as well as Robert Axelrod's study in the 1980's on the "Evolution of Cooperation", (what is the group learned behaviour over time?) both tackled the same problem and came up with their own conclusions.
I've expanded the example to fit the scenario we want to share, it goes something like this:
10 blindfolded prisoners are sitting in the same jail cell;
Each is serving a 10-year sentence;
The prisoners are told by their jailer that if they all raise their hands, each of their sentences will be reduced/commuted to two months;
The prisoners are also told by their jailers that if one of them chooses not to raise their hand, he can go free.
But there’s a catch.
The prisoners are also warned that if two or more of them chooses not to raise their hands, then all their sentences will be doubled to 20 years.
The sensible thing for a prisoner to do would be to raise his hand – to benefit themselves and the group.
But as long as an opportunity for personal benefit is offered, the temptation remains for every prisoner to leave theirs down. As all the prisoners have this temptation, human nature suggests they are all likely to put their own personal benefit first.
Thus each prisoner leaves his hand down, and all their sentences are extended.
Axelrod's "Evolution of Cooperation" study found that over iterative instances, 2 key elements emerged were for the behaviour to evolve then a sufficient deterrent or incentive needed to be in place to counter the self-interest.
In short, the chance for personal gain is more appealing than the certainly of gain for the group.
And here’s where your Chicken Tikka Masala comes back in.
As I mentioned at this start of this blog, Inspire has been approached by various takeaway businesses in “local territories” like yours – the prisoners in our extended analogy.
Each currently pays between £18,000 and £20,000 annually in commission to Just Eat.
In the last year, Just Eat have increased their charges, from around 14% and 18% of the value of a single order sent to a takeaway business to prepare, as well as many new disruptive tech players entering the market.
In some areas Just Eat also offers the option of delivering the product to the customer, but there’s a price to pay for that too: the jump to a staggering 42% commission.
With costs like those, it’s no surprise that takeaways are all looking for a way to cut their online partner out of the loop.
They each wanted us to create for them a strong online presence to wean their customers off of their online partner.
We could have done that, of course, but seeing the bigger picture, we recognised that there were issues within the marketplace which meant that ultimately – even with a strong online presence – each takeaway was still going to need to market itself heavily online.
For a start, it’s not easy for a small takeaway business to go up against the likes of Just Eat. To reach the same audience would not only be very difficult, but also expensive and time-consuming, even with the advantages provided by a democratised web e.g. Google Map Pack, Search Engine Ranking, Google search ads and social platforms e.g. Facebook Ads.
The other issue is the fact that most of the other takeaways in the area were already also using Just Eat to sell their food.
And here’s where the plight of the takeaways becomes the Prisoner’s Dilemma.
If all the takeaways in a particular local territory were to leave Just Eat or online partner and create their own individual online presence, they would all benefit.
With no takeaways to send orders to, Just Eat would no longer be able to service the area.
With no commissions to collect, Just East would lose interest in that territory. It would likely stop advertising in the locale, and likely focus its energies elsewhere.
But that only stands if all the takeaways leave Just Eat. If only one – or even a handful of them – continued to use Just Eat, those that remained would effectively do a land grab on the available takeaway orders for that day.
In that case, Just Eat's marketing in the area would not stop, and the lure of being involved with Just Eat for individual takeaways – whatever the cost or downsides – would not subside.
Just Eat would continue to have an interest in that area, and would out-price and out-value whatever an individual takeaway could do, leaving them back where they all started.
We’ve already seen something similar happen.
When all the takeaways agreed to turn off Just Eat every Sunday and use their own marketing mediums, the result was pure Prisoner’s Dilemma.
Seeing the opportunity to benefit themselves, rather than stick together for the benefit of the group, one or two went rogue in order to snaffle up all the day’s orders for themselves.
So what to do? Axelrod's "Evolution of Cooperation" tells us that learned behaviour – like that of the prisoners deciding whether or not to raise their hands – evolves.
It also tells us, crucially, that the evolution of that learned behaviour can be influenced through either a sufficient deterrent, or a sufficient incentive.
The result: the learned behaviour on the part of the takeaways who actually turned off Just Eat for the day was according to the "Nash Equilibrium" that playing fair is not just pointless but also not an optimal or profitable solution and the approach should be abandoned.
What they all should have done – rather than staying in Just Eat or going it alone – is created a guild – and here’s why.
A ‘“a local territory near you” Takeaway Guild’ could collect a contribution of, say, £5,000 from each member.
Handing over that sum to the guild would not only mean that the participating takeaways would avoid paying Just Eat’s much higher commission – a strong incentive – but would also be a large enough sum for them not to want to lose if they were booted out of the guild for flouting the rules – a strong deterrent.
The money collected from guild members would be pooled and used to gather online takeaway orders for all the restaurants through a strong online presence. It would also fund local online advertising.
If all our Prisoners – the takeaways – agreed to stick to the rules, all of them would benefit.
But because deep down they all want to make their own deals to benefit themselves, Game Theory – and human nature – suggests that in going it alone against disruptive tech like Just Eat, our local takeaways will just go from the frying pan to the fryer.