In 2023, the German real estate market was in a complex and delicate dynamic. While the demand for properties remained high, increased interest rates made it difficult for most of the German population to afford real estate. The chart below clearly shows the correlation between the increasing interest rate and the decreasing transaction volume.
Problem complexity
▪️ Property developers need to plan years ahead before selling. Therefore, they couldn't foresee such a market change and can't decrease their prices as they need to cover construction costs.
▪️ The majority of the German population can't afford to buy for self-use unless prices or interest rates drop significantly.
▪️ Buying for investment is still affordable. However, many investors have anchored themselves in a timeframe of low interest rates. Their calculations on whether a property is worth buying are based on a low-interest-rate market.
The solution I will present in this case study is the result of nearly a year's worth of insight gained from this challenging market situation. The following were the primary sources of learning that contributed to the solution:
▪️ User Research: We conducted two rounds of user interviews. The first aimed to understand current user challenges and their interaction with our unit page, which displays specific property information. The second focused on understanding the motivations for buying, the decision-making process, and the search experience.
▪️ Lean Design: Throughout 2023, the team conducted small experiments, by delivery new or improved features, to validate hypotheses about which ideas would be most relevant in the current market.
▪️ Sales experiments: The sales team also experimented with different approaches to determine which one would be more effective.
Throughout the year, I monitored what our competitors and international benchmark markets were doing.
▪️ Behavioral economics studies: in specific, two scientific papers, the first from Amos Tversky and Daniel Kahneman (1981) and the second from Maya Shanton (2017)
In a notable 1981 article in Science Magazine, Amos Tversky and Daniel Kahneman demonstrated how the presentation of information can influence conclusions. Their arguments were based on a series of lab experiments, where they observed how different framings of economic and moral problems led participants to opposing conclusions. This effect, known as "framing," is a recognized bias in the field of product design.
Based on our user research, for an investor, the balance between mortgage and rent is more important than the actual property price. Furthermore, in the German market, an investor can sell the property after 10 years without having to pay capital gains tax. Therefore, it's common for someone to buy a property for investment purposes and sell it in 10 years, potentially before the loan is fully paid. This means that the total price is not the most relevant key performance indicator (KPI).
It was clear how the way we (and the majority of the market) were framing the problem was not helping the user in the current market.
Research on framing became more intriguing when scientists transitioned from lab experiments to real-life case studies where decisions had tangible effects. In 2017, Maya Shanton published a paper on her study examining the effects of a new regulation on retirement funds in Israel. This regulation prohibited the display of fund returns for periods less than one year. This change influenced people's investment behaviors, leading to increased investments in higher-risk funds, which generally yield better returns over the long term— an ideal strategy for retirement investment.
Moving in the same direction, our sales team had relative success when presenting a business case for a particular project. By combining unique financing options and tax benefits from investing in real estate, then selling after 10 years, the business case demonstrated how a property could yield impressive profit.
From a feature we launched in early 2023 that was not successful, we learned that users are skeptical of numbers that contradict their market beliefs without sufficient clarification. Therefore, I designed a calculator for the unit page to better illustrate how this profit can be realized.
To test this calculator, I created an Excel file that replicates all fields and the organization of information. This file was shared with experienced and non-experienced real estate investors. The main insights from this testing were:
▪️ The presentation of data captured both users' type attention.
▪️ For non-experienced users, because they are not accustomed to thinking from a 10-year perspective, it is challenging for them to determine whether or not the profit is good.
Iteration: Add a second line to the chart to show how an investment in ETF would perform.
Despite positive feedback from user research and internally, the company couldn't risk developing a full feature without concrete market feedback. Therefore, I proposed a Minimum Viable Product (MVP) plan to test our hypothesis with minimal engineering effort.
The MVP plan involved creating a landing page where users could download the Excel calculator used in user testing. We would then monitor the conversion rate of this page, compare its performance with other pages, analyze if those who downloaded the Excel converted better, and interview users who interacted with the feature.
