Adding biases in UX design
During the concept stage, in the CEJ, I added another layer where potential behavioural economic biases that can be added. For more detail decision on the UX design you can read the case study for lending redesign.
I had some basic training in behavioural economics in my previous job with a book club reading through the nudge book and was recommended this resource http://www.brainsbehavioranddesign.com/ and their strategy cards.
Loss aversion — in the content
I tried to increase the desirability of the card on the app by testing different content with some loss aversion strategy.
I separated in the losses and gains into present and future for the customer and tried to incorporate the current gains into the content.
Examples would be getting the loan funds upfront, no jumping through hoops and able to accepting the loan and save time on the app.
Unfortunately, it didn’t work as user’s didn’t believe it and we had other factors in play.
Reducing Anchoring, confirmation bias and cognitive dissonance
“Anchoring or focalism is a cognitive bias that describes the common human tendency to rely too heavily on the first piece of information offered (the “anchor”) when making decisions. During decision making, anchoring occurs when individuals use an initial piece of information to make subsequent judgments.” — wikipedia
No matter where we presented the fee information, users were anchoring on it and because we don’t show the interest rate they are satisfied they have found the ‘catch’ that we are hiding it.
The skepticism with the ‘fee’ information has satisfied their confirmation bias from the beginning — the loan was “too good to be true”. Our loan model is different to how a traditional loan works (no interest rates/fixed term etc) and users are experiencing cognitive dissonance when they are presented with new a new model of our loan and trying to marrying it together to how they think a loan works.
Confirmation bias, also called confirmatory bias or myside bias,[Note 1] is the tendency to search for, interpret, favor, and recall information in a way that confirms one’s preexisting beliefs or hypotheses.[1] — wikipedia
How we tried to counter this is to provide more clarity and information for the fee to lessen the impact of anchoring. Also if we created a more ‘personal’ experience it can help combat the confirmation bias and cognitive dissonance popping up.
Also on the button for the next screen, we are priming users with the Ikea effect — “Build your loan”
Information avoidance
“Information avoidance- when too much information may overwhelm people and cause them to exhibit the ostrich effect — stick their head in the sand and ignore the information altogether.” — From the strategy cards from brainbehaviordesign
We found a high percentage that 94% of users continue onto the next stage from the metrics. But after some qualitative testing, we found out that people didn’t engage with this page and skipped unless we took them back to this page and asked the questions.
I did some internal testing on what was the optimal layout for engagement what information users digest before they zone out with the ostrich in the sand effect. I did 7 tests, incrementally and here were the results:
Unintended goldilocks effect
There was a bias in one of the concept designs in the table format. The assumption was if everything was laid out, there will be higher in comprehension of the different fees, hence would lead to higher conversion.
What we found in testing- because of the goldilocks effect:
“Goldilocks pricing is a marketing strategy that, although not directly related to the Goldilocks principle, uses product differentiation to offer three versions of a product to corner different parts of the market: a high-end version, a middle version and a low-end version.” wikipedia
We found out that users foremost based their decision on the % of eftpos, then secondly the fee as the main deciding factors of taking a loan.
Duration was not a big factor, but they had a fixation of about it because it was in the centre. In the next concept designs we made sure that the % of eftpos and fee was more important and less emphasis on duration.
Ikea effect and loss aversion (creating ownership)
“The IKEA effect is a cognitive bias in which consumers place a disproportionately high value on products they partially created. The name derives from the name of Swedish manufacturer and furniture retailer IKEA, which sells many furniture products that require assembly.” — wikipedia
If we made the page feel like they are building the loan, then users will think more highly of it and create a sense of ownership — loss aversion.
From a UX point of view and I learnt from my stint at ING, a slider is generally not a good ux practice because it becomes inaccurate for users to choose a specific amount.
In this case, the slider created more engagement and play compared to other designs that just had a text field. We learnt from our previous concepts that users learn more by engaging and playing on this screen compared to layouts where everything was laid out but engagement was minimal. But I still added the edit functionality so users can add a specific amount which tested well in usability.
As we are priming users to ‘build’ their loan, the bottom bar becomes like the built product- the sum of their chosen parts.
We also found that the large buttons of the graph with custom options encouraged learning through playing and had a much higher engagement that the format where all the options were laid out in a table.
What we didn’t add
We can manipulate the design such that the first fee bar is lower than the duration in order to give the impression the Fee is lower/more reasonable or affordable with the contrast effect.
We can also also have a default option set on the highest fee for the status quo effect . I call this the lazy effect where users stick the selected default with the assumption from users that default options come with an implicit endorsement from the default setter.
And if users don’t default select this then all the other options will be cheaper and seen more positively.
But we wanted to encourage responsible lending so we didn’t implement this.
We also know that 62% of users drop off from the loan summary page, but we didn’t want to add any biases on this page such as loss aversion etc to promote responsible lending.
But we did add the recency and primacy effect to aid in understanding that the one flat fee is really just only one fee and it is the total of the loan amount and the one flat fee. Surprisingly this tested well.
The recency and primacy effect is when a user takes in the first information, skips the middle section and takes in the last information.
