Understanding Automatic Behaviour:
To encourage visitors to undertake a new behaviour on your website, such as making a first purchase or playing a game, you might want to offer something to reward that activity. But when should you provide a reward and how do you encourage repetitive behaviour?
These were the same kind of question that the American psychologist B. F. Skinner wanted to answer in the 1950s. He wanted to go beyond the work of Pavlov on ‘classical conditioning’ and investigate how rewards influence behaviour.
Pavlov discovered that animals (including humans) learn to respond automatically to certain stimuli (e.g. food) and that if you add further stimuli (e.g. a bell before the food arrives), you can also get them to automatically respond to that (e.g. salivate) in the same way. If you then remove the original stimulus (e.g. food) animals will continue to respond to the new stimuli because the behaviour has become automatic.
Encouraging Automatic Behaviour:
Classical conditioning is important to conversion because we are more likely to be successful if we can make people do things automatically. This is one reason to ensure your navigation and site layout conform to web conventions (e.g. navigation is either at the top or left hand-side of the page) as people will automatically know where to look for them.
Skinner developed powerful new theories of behavioural analysis (often referred to as operant conditioning), based upon continuously testing different scenarios to identify how the frequency and predictability of rewards influences behaviour. He called these differences “schedules”. The five schedules he examined are:
- Continuous reinforcement – You reward the behaviour every
time it is completed.
- Fixed interval – You reinforce the behaviour with a reward
at fixed time intervals after the first time the behaviour is undertaken.
- Variable interval – You provide a reward after an interval
of time, but you randomise the amount of time to make it unpredictable.
- Fixed ratio – Rather than basing the reward on time, it is
determined by the number of times the behaviour is completed and this number remains constant.
- Variable ratio – Here the reward is again based upon the
number of times the task is undertaken, but the number is randomised to make it unpredictable.
So, what did Skinner discover about these different
schedules and how can we apply it to the real world?
Establishing a new behaviour:
Continuous reinforcement is the most effective schedule when we are trying to establish a new behaviour. Providing a reward every time helps to reinforce the behaviour and there is no uncertainty about getting the reward. However, once the behaviour is established switch to a different schedule as otherwise the behaviour will become intermittent and stop if the reward ceases.
Many ecommerce websites use money off discounts or coupons
to reward people for making their first online order. It is a clear and unambiguous way of incentivising people to add items to their basket and complete the order process. However, many customers fail to return regularly and so what should companies do next?
Making Behaviour Stick:
To encourage repeat behaviour a variable ratio schedules is most effective because it is linked to how often the behaviour is undertaken. As it is unpredictable people complete the behaviour over and over again as they are unsure when the next reward will be received. Such schedules are resistant to extinction by their very nature and help to make some behaviour addictive.
Gaming websites understand this aspect of psychology very well. They ensure rewards are directly linked to how many times a game is played, but use the unpredictability of randomising the frequency of rewards to maintain a level of suspense. Loyalty schemes often fail because although they may be based upon the frequency of purchasing behaviour they are usually too predictable and lack that random reward.
Regular and Stable Behaviour:
When you don’t want to encourage a lot of the behaviour at once, but would prefer it to be steady and regular, a variable interval
schedule is usually best. This is not very effective at establishing new behaviours as people are unsure if they will get rewarded when they complete the task. However, once the behaviour is established
this type of schedule provides a clear incentive to continue with an activity provided the maximum time period between rewards is not too great.
Ecommerce sites use this to their advantage by varying the frequency of discount offers and sales to encourage visitors to keep returning to their website. If the timing of discounts is too predictable visitors soon learn only to return to the site when they know an offer is available. Making sure that the timing of discounts is unpredictable creates a reason to return on a regular basis.
For a Burst of Behaviour, But it Won’t Last:
To get a sudden increase in activity a fixed ratio schedule will result in such behaviour. But beware as once the goal is achieved you will see decline in the behaviour. The classic example is the buy 10 cups of coffee and get the next one free.
Show How Close Customers are to The Reward:
Research into fixed ratio schedules has shown that you should show how close people are to achieving the reward as people accelerate their behaviour as they progress towards the goal. The insight here is that people are more motivated by how much is left to reach their target rather than how far they have come.
For a Sudden Spike in Behaviour:
The problem with fixed interval schedules is that people have a tendency not to take much notice of them until just before the time the reward is due. They are too predictable and so continuous reinforcement would be more effective at encouraging a new
behaviour. Seasonal sales suffer from this as visitors surge during the sale but fall off dramatically once it is over. Flash sales can help counter this predictability and give visitors more reason to return to your site on a regular basis.
Encourage Baby Steps:
What if you want to encourage a new behaviour, but your visitors can’t or won’t adopt the behaviour until they have undertaken other tasks (e.g. registration or adding items to the basket). Shaping is psychological process that involves rewarding an earlier activity that will lead to the desired behaviour.
This works best with small behaviours and that you identify every
instance of the activity that you want to encourage. It is then important that you reward it every time with something the visitor sees as a positive reinforcement (e.g. praise or loyalty points etc.) and give it immediately after the behaviour. Shaping is a great way of nudging your visitors towards your ultimate goal by positively reinforcing those small baby steps that lead to the target behaviour.
Choice of Reward:
When selecting rewards it’s important to give something a person really wants as otherwise it won’t motivate them. Try A/B testing different rewards to see what engages your visitors most. Many organisations copy what their competitors are offering, but try testing something novel as people are instinctively drawn to new or different rewards.
Remove Something Your Customers Don’t Like:
Negative reinforcement, where you take something away that
people don’t like, can also work to encourage a desired behaviour. Many e-commerce sites now offer to remove shipping costs completely if your order value is above a certain amount. People
dislike paying for shipping costs and so by removing them you reward customers whose basket value is over the specified amount.
Punishment is Less Effective:
Punishing people is different from reinforcement as you are
applying a behaviour that the person doesn’t want to reduce an activity you want to stop. However, punishment is generally less effective than reinforcement because it is harder to get people to stop doing something than finding an alternative for them to do instead. Indeed, punishments can sometimes back-fire as people change their behaviour in unexpected ways to avoid the pain.
In one case a day care nursery applied a late pick-up fee for parents who collected their children after 6pm with the intention that it would reduce the behaviour. The opposite happened as the nature of the relationship changed. Before the fee was applied parents might feel some guilt that they were preventing the day care staff from getting home. But once the fee was applied it became a pure monetary transaction that the parents felt they were compensating the nursery for and so did not feel so guilty about.
Using the appropriate schedule for rewarding visitors on your website can be a powerful means of engaging and encouraging desired behaviours. Ensure you take time to test different rewards though so that you find what motivates your customers. Try removing things that might cause your customers pain (e.g. fees or charges) especially if you can link it to targeted behaviours.
Thank you for reading this post. If you found this post useful please share using the social media icons below.
You can view my full Digital Marketing and Optimization Toolbox here.
To browse links to all my posts on one page please click here.
Recommended reading: How to Get People to Do Stuff: Master the Art and Science of Persuasion and Motivation by Susan M. Weinschenk, Ph.D.
- About the author: Neal provides digital optimisation consultancy services and has worked for brands such as Deezer.com, Foxybingo.com, Very.co.uk and partypoker.com. He identifies areas for improvement using a combination of approaches including web analytics, heuristic analysis, customer journey mapping, usability testing, and Voice of Customer feedback. By aligning each stage of the customer journey with the organisation’s business goals this helps to improve conversion rates and revenues significantly as almost all websites benefit from a review of customer touch points and user journeys.
- Neal has had articles published on website optimisation on Usabilla.com and as an ex-research and insight manager on the GreenBook Blog research website. If you wish to contact Neal please send an email to firstname.lastname@example.org. You can follow Neal on Twitter @northresearch and view his LinkedIn profile.