Opinions expressed by Contractor the contributors are their own.
Startups are synonymous with innovation. In this fiercely competitive market, they must constantly come up with innovative ideas. However, not all ideas are good enough, so it becomes important to test the ideas beforehand to validate them. A well-known method of doing that is called A/B testing – or more accurately, experimentation.
However, experiments too often fail due to imprecise experimental design. We have all witnessed it. This leads to a bias against the experimentation process in general. We strongly advocate against this prejudice. Experimentation can be the key to growth for a startup.
However, many things can go wrong with experimental design. The one we will particularly highlight in this article is how the understanding of consumer behavioral intelligence can lead to successful experiment design. This is something that is not usually talked about, and I see many in the industry not paying attention to it.
Related: The 5 Key Elements of a Successful Experiment
Consumer behavioral intelligence, or behavioral economics, is the field in which researchers study human behavior in terms of money and value. Behavioral economics reveals that human decisions, especially in monetary contexts, can turn out to be quite irrational. Therefore, we can utilize behavioral economics in the design of our experiments.
If you’re looking to improve your revenue, take-up, trial rates, or retention, experimenting in accordance with behavioral economics may be the growth hack for you. It can be a very powerful concept for designing your experiments. Given the enormity of behavioral economics, let’s break down three concepts and discuss how they can be used in experimentation.
1. The allure of free
It seems logical that charging a negligible fee for a service should make no difference to customer growth. That is not the case.
Let’s study this concept with a much talked about case study: Hershey’s Kisses vs. Lindt, the gourmet chocolate. An experiment was conducted where Hershey’s was priced at 1 cent and Lindt was priced at 15 cents. Consumers were asked to choose between them.
A significantly higher number of people chose Lindt at a much higher price than Hershey’s, which was almost free. The consumers justified their choice by saying that Lindt was Luxury Chocolate.
Later, the experiment was repeated with the same group of people. But this time the choice was a little different. There was a 1 cent reduction in price for both cases. That means Lindt went down to 14 cents, and Hershey’s went free. To the researchers’ surprise, the results completely reversed, and a significantly higher number of people picked Hershey’s, even though Hershey’s was plain chocolate.
This experiment conclusively proves that there is a deep allure of “free” in human minds, and we make a big distinction between “free” and “almost free.” So here’s a tip: If you’re in an app business, know that in a user’s mind, “free” and “almost free” make a big difference. They are two distant options. Keep this in mind as you design your experiment as well.
Related: The basics of experimentation and why it’s key to your startup’s growth
Another experiment was conducted at an airport. People at the airport were told to either pick up yogurt or fruit from a counter. Initially, almost half of them chose yogurt, and the other half chose fruit.
For the next part, someone on the way to the counter was talking to people in the queue. What they found is that when this person talked to them about yogurt, people picked more yogurt. And when this person spoke to them about the fruit, they picked more of the fruit. This is a good example of priming.
What we learn here is: The key is to draw the consumer’s attention to a product/service you want to sell. Their decision is likely to be influenced automatically. The way to attract attention doesn’t even have to be direct.
The person talking to the people in the queue didn’t necessarily have to point out the product. What they say may simply be about the product. To sell the fruit we don’t need to ask them to buy the fruit, we can just ask them which fruit they like and that does the magic.
3. The lure effect
Let’s take the example of a dating app. The app provides three matching profiles in the free account and then has an upgrade option. The upgrade has two options.: There is a basic upgrade and a premium upgrade. Let’s say there are 5% of users who upgrade, and the distribution is 4% for basic and 1% for premium. Can we introduce a decoy in these choices to switch the relationship around? Well, it is possible!
Suppose the app introduces another upgrade option. It is in line with the premium package and is called standard pricing, but it is inferior to premium pricing. Let’s say the basic upgrade has five features, the premium upgrade has five plus seven amazing features, priced at $999, and the standard upgrade has five plus another feature, priced at $899.
You will be surprised to know that just by introducing this bad option, the app will be able to switch the ratio from 4%:1% to 1%:4%. The reason for this change is that users were previously unable to directly compare basic with premium. Now, however, they have found a comparison between the premium and standard packages. Five plus one is available for $899, instead of 5 plus 7 for $999. It’s a simple comparison and the app might sway people towards the prize. This is another very useful concept that you can use to design your experiments.
Related: 4 Ways to Get the Most Out of A/B Testing Right Away
Many studies in behavioral economics give us a good idea of consumer behavior in terms of money and value. Startups can leverage these studies to design and execute successful experiments. We discussed three simple but very powerful concepts of behavioral economics above, and implementing them can help you conduct successful experiments yourself.