.svg)
Crafting intuitive interfaces within real estate CRM tool, we used machine learning to simplify processes, ensuring effortless navigation for broker agents to find matching buyers to their property.
A real estate company using technology to enable broker handling more than 25 properties per month. (usually it would be 5)
Real estate brokers, aiming to assist them in efficiently matching available properties with potential buyers.
What are the specific pain points experienced by brokers in matching properties with potential buyers?
How can the user experience be improved to streamline the process of matching properties with potential buyers?
What iterative design methods can be employed to continually refine the solution based on user feedback and evolving needs?
What was deliver and how
Some of the details in this case
study may be vague to protect
the client's intellectual property.
User research and analysis
Collabotatiojn with tech department
Prototyping User testing
MVP delivery
User stories,
UX flows
Prototypes Design deconstruction per ticket for smooth engineering integration.
Key performance indicators (KPIs)
Enhanced broker efficiency by facilitating quicker property-to-buyer matches and significantly improved user experience, elevating broker productivity. Moreover, the user-friendly matching system successfully increased buyer engagement, while automated email notifications contributed to a remarkable 40% surge in buyer conversions.
In Germany's real estate market today, increasing interest rates have made it harder for buyers to join in, resulting in a challenging market environment for brokers.
This led us to focus on making it simpler for brokers to find the right buyer, especially considering the decrease in buyers in the market.
This situation pushed us to uncover a solution that can be build by the internal tool team.
But before building we must understand. Therefore we conducted 6 contextual inquires interview with 5 brokers and 1 directors to understand how they palliate this buyer drought.
This situation pushed us to uncover a solution that can be build by the internal tool team.
But before building we must understand. Therefore we conducted 6 contextual inquires interview with 5 brokers and 1 directors to understand how they palliate this buyer drought.
The interview was composed of different questions such as:
Left side
Right side
These findings have paved the way for a clear action plan:
Acquisition broker will use the matching tool to show owner they have potentials buyer for the property and even sell it before it hit market.
But how would this tool be integrated in their workflow?
Our research found that showing proof is important for owners. In our new process (illustrated below), brokers can access a list of interested buyers. They can print it and share it with the property valuation. To keep it secure, only first names and hidden contact info will be shared. This helps owners see potential interest without contacting buyers directly, making it safer for everyone involved
Selling brokers are currently missing applicants for their opportunities. They will first look for people applying to their specific opportunity in our internal tool. If no luck their they can look in 3 data sources, their past application, past application for other brokers in the same areas and people who created search profiles to be alerted if Mcmakler has a matching property for them.
.png)
To understand how well our solution is working, we use two kinds of tools: Key Performance Indicators (KPIs) and Metrics. KPIs are like big goals we want to achieve, while Metrics are specific details helping us track progress.
These are important goals we're aiming for, like:
These are specific numbers we keep track of to understand how things
are going, like:
Both KPIs and Metrics help us see if our solution is doing well, giving us
the big picture and the small details
The market situation urgent us to focus on helping selling broker finding a buyer, therefore we are focusing on this journey. Nevertheless the created solution will also be usefull for aquisitiion broker who will have access to the list as well as early as possible.
Based on my knowledge learned during research and the UX flow agreed on , I was able to put myself in a selling broker shoes and imagine how and when they would want to interact with a potential buyer list.
Because we have components ready and to make it more realistic we builded a prototype allowing a define set of actions for brokers to test and give feedback on.
The scenario is the following:
.png)
.png)
These prototype was shared with engineering team to assess technical feasibility while simultaneously guiding backend engineers in building a more query-friendly database.
6 brokers spanning various experience levels and age demographics were engaged to test the prototype, allowing for comprehensive feedback. Iterative improvements were then made based on this feedback, with each iteration in alignment with the engineering team.
We are prioritizing the mobile version for testing as BrokerForce was specifically designed to empower brokers who are frequently on the move, with approximately 80% of their time spent traveling to properties. Ensuring optimal performance on mobile devices aligns with the practical needs and usage patterns of our target users.
As stakeholders' requirements evolved during the project, in order to be agile, we integrated those requirements into the testing process while implementing iterations from the user
During testing we realised that user took more than 1 min to notice they have a new property in their list. So we decided to add a toaster to the screen so they understand the information faster.
After two rounds of testing involving 12 brokers, we observed that less tech-savvy brokers faced challenges distinguishing applications from potential buyers within the same list. To enhance user comprehension and ease of comparison, we opted to present them in separate lists
After two rounds of testing involving 12 brokers, we observed that less tech-savvy brokers faced challenges distinguishing applications from potential buyers within the same list. To enhance user comprehension and ease of comparison, we opted to present them in separate lists
.png)
During testing, we noticed differing preferences in the Potential Buyer tab. To confirm these insights, we sent a survey to all brokers, asking them to prioritize the data. This validated and guided our optimization efforts.
To evoid data scrapping, we decided to hide last name from broker in the table. The complete name appears ones they open the card.
The survey showed a big imprtance to btoker if the Potential Buyer was answering the phone and if yes in which time frame. Therefore I created a chip system based on the report to say if they answered or not and in which time frame.
During the testing phase, users expressed a desire for more information about applicants and an intuitive navigation system. To fulfill this need, I created an applicant card. To ensure consistency, we applied the same logic to Potential Buyers.
.png)
key elements of this card:
Testing revealed brokers' need for enhanced reporting and a quick, efficient way to recall details about Potential Buyers. To address this, we introduced tags based on the broker's past reports for specific opportunities.
We drew insights from 'The Best Interface is No Interface' by Golden Krishna, recognizing the negative impact of dropdowns on users. Inspired by the filter design of platforms like Airbnb, we opted for a chip-based approach to display options in the filter section
Throughout the development of this feature, I've gained valuable insights into the iterative design process and the critical role of user feedback in shaping product features. By closely collaborating with both users and the tech team, we ensured that our solutions were not only feasible from a technical standpoint but also highly usable for our target audience.
The positive impact of our feature implementation became evident through the metrics we observed during testing. Within just three weeks of its introduction to 20 brokers, we witnessed promising results: each broker received at least three applicants through the new feature within the two-week testing period.
This success not only underscores the effectiveness of our solution but also highlights its significance in enhancing user experience and driving tangible outcomes.
