The 4 Pillars of Change in Gamification, Optimisation and Behaviour Change

In my new world, I focus on optimisation a lot in the realms of e-commerce performance. Testing, personalisation etc. It has taken a very long time for me to realise that my new world is almost identical to my old world of gamification. At their most basic, they both rely on 4 key pillars to succeed as they are both focused on behavioural change.

I wanted to just explore those briefly here with you, as much as a way of consolidating the thought process as much as anything else!!

The four pillars are.

What do you want to change?
Why do you want to change it?
How are you going to change it?
How are you going to measure the success of the change?

What do you want to change?

In gamification as in optimisation, this is always the starting point. What are you going to change? When considering using gamification, the answer might be “I want to get fit”. In optimisation, it may be “I want to make it easier to add products to cart”.

Whatever the change is, you need to be able to define it as simply as possible and as concisely.

Why do you want to change it?

This is more important than anything else. If you don’t understand why you want to change something, you can never hope to actually change it. Usually, you would want some sort of data to support the What.

In the case of fitness, it could be “I am overweight for my age and height and feel unhealthy.” In optimisation, it might be “UX research has shown that people get frustrated needing to go to the product detail page to add common products to their cart.”

How are you going to change it?

This is where you get to look at the problem from as close as possible. Just saying “I want to get fit” is a pretty big challenge. You want to set some manageable goals and have multiple how steps in reality. In this instance, it might be

Week 1: Reduce sugar intake and do 10 press-ups and 10 sit-ups a day
Week 2: Reduce fat intake, do 10 press-ups, 10 sit-ups and walk for half an hour a day
Week 3: Reduce alcohol intake, do 20 press-ups, 20 sit-ups and walk for an hour a day

And so on. You would then include some kind of gamification to make some of this more enjoyable and manageable such as Zombies, Run!

With our e-commerce challenge, we may look at it and say “Adding an add-to-cart button on each product on a product listing page will reduce the need to go to the product details page thus reducing customer frustration”.

In both instances, you are choosing a change or changes that are easy to achieve and help chip away at the initial challenge you have discovered in the Why part of this process.

How are you going to measure the success of the change?

None of this is of any benefit if you are unable to define and measure the success of the change you are making, as well as consider what other effects this change might have.

With the fitness challenge, you could look to measure your weight at the end of each week or month. If your weight is decreasing, then your changes might be working. However, you may also want to check “guard rail” metrics. Check your blood sugar for instance. If your weight is dropping, but your blood sugar is dangerously low – then you need to rethink the changes you are making.

In the e-commerce space, you could measure the number of items being added to the bag, to prove that making it easier to add content to the bag increases something. However, if not letting people get to the product details pages means they are buying lower-value items or even returning more items because they don’t see important info – then the change is actually doing more harm than good.

A great example a friend of mine uses is “I can make your customers buy tonnes more products by dropping the price of everything to £1, but will that change really be of benefit to you and how would you know if you measure nothing other than the purchase rates?”

So always ensure you can define and measure success with as many metrics as you need to mitigate unintended consequences wherever possible!!

Putting it together

In our gamification example, we would have a hypothesis that looks a bit like this.

I want to get fit because I am overweight.
I am going to do this by changing a number of existing dietary habits and introducing more exercise with the aid of Zombies, Run! I will know I have succeeded by checking my weight and also ensuring that my general health is maintained or improved.

In the optimisation version, we would say.

I want to make it easier to add products to basket because customers get frustrated having to go to the product details page every time.
I will try doing this by putting an add-to-basket button on each product on the product listing page. I will know this is working because average items in the basket will increase, frustration metrics will decrease and there will be no adverse effects on average order value or overall revenue.

Before you go!!

It is really, really, really important to not view this as linear! You will need to iterate on the hypothesis if it starts to show signs of failure! If it is succeeding, you may also want to iterate – add new ideas and get even bolder with the “What you want to change” part of the hypothesis as more data becomes available.