Taxfix, 2023 - Checkout - Payment Model
Payment value propositions
The power of behavioral design on a checkout screen, in good and bad times.

The Challenge
Uncover the reasons for low opt-in rates and optimize the screen to increase CVR to an instant payout feature named Instant Refund. Stay tuned until the end for a surprise turn of events.
My Role
UX Design, research strategy, interaction design, and behavioral design.
The Situation
What's the problem?
Taxfix introduced Instant Refund, letting users receive 50% of their tax refund within a day and the rest when the tax office paid out—cutting a 3-month wait to just one day. It seemed like a strong value proposition—or so we thought.
User problem
Instant refund is hard to understand - The Users seemed to have a hard time understanding exactly how the payout option and the transaction timing worked and opted for a standard payout in 90% of cases.
Business problem
Low opt-in rate - Instant Refund was a lot to maintain for the return on investment. We partnered with other banks to make Instant Refund a reality and needed to get more confident this was something users really wanted.
Discovery
Research
I suggested a quantitative study with Taxfix users that would uncover biases underpinning the decision for the payout model in greater detail and then address these biases with the com-b-behavioral model. We were particularly interested in users who didn't opt into Instant Refund.
Method
• Quantitative survey via Survicate
• Taxfix users
• Invites via CRM
• 400 responses
• 2 weeks
Analysis
The research outcome was analyzed by our quantitative researcher and screened for specific customer segments we could derive from using user IDs.
Findings
Research Insights
The study uncovered psychological biases underpinning why customers were hesitant to book Instant Refund.
Present bias
58% of customers didn't need their refund instantly
Loss aversion bias
27% of customers felt like it was safer to wait for the money
Status quo bias
26% of customers were always happy with the standard payout and didn't want to try anything new
Negativity bias
15% of customers were afraid of getting less money overall.
Using behavioral design and growth design principles to combat biases
Re-frame for urgency
Re-framing for potential urgently needed funds: Extra spendings, unexpected costs, Inflation, bills, christmas, holidays - anything that creates a feeling of wanting to be prepared
Re-frame for safety
Highlight trustworthyness of Instant Refund because of the IR Identity Check. Users new belief shold be: “IR seems safer than waiting for the standary payout because of the IR identity check.”"
Re-frame for efficiency
By showing users that IR is easy and significantly reduces waiting time until the refund is there, we can motivate users to give IR a try.
Show refund pairity
Clearly show that the refund the user gets fromt he tax office is exactly the same with both options.
Concept design
Optimizing the Instant Refund opt-in screen
Together with content design and product marketing, I created a new opt-in screen design concept based on behavioral design and growth design principles, combating the psychological biases that users had towards Instant Refund.
.jpg)

One day before release:
Major fraud attack
Emergency re-design
Emergency pivot - LOWER opt-ins to Instant Refund
One day before the main filing day of the year (when Taxfix makes the most revenue) leadership and security advised us to stop all Instant Refund work, as we had a major fraud attack that exploited Instant Refund. We could not switch Instant Refund off completely. Hence, we decided to use the study insights to promote standard checkout instead, limiting the feature exposure for certain refund groups most likely to be fraudulent.


Finally, the emergency re-design led to
-40% opt-ins to Instant Refund
Which was exactly what we needed to get through the fraud attack.
Final words
Learnings
This project took an unexpected turn at the last minute. This tested my ability to pivot, have resilience in a moment of crisis, stay creative under pressure, make the right decisions under pressure, keep the focus on the best possible outcome. The research insights enabled us to optimize for both user segments, which was immensely important in the end. Although I wish we could have seen the Instant Refund optimized screen perform in high season, I am happy about the outcome regardless.