HLEDÁ SE NEMO

2015 - 2016

A real estate finder and alert bot concept that aims to help buyers take easy and safe steps in the property buying process on their own within the Czech market. The main aim is to save the buyer significant time, looking for new offers faster than the user could by themselves.

In this story, I would like to demonstrate my design process of this real estate concept.

My steps involved defining a research strategy to learn about the audience. I learned about customer journeys and performed experience mapping to uncover the truth. This helped me to explore the big picture and see the users’ problems to identify the design and business opportunities. This knowledge became my foundation for shaping the ideal experiences to deliver a viable concept that solves users’ struggles while being more simple, useful and fun.

My Role

As this was my side project, I was establishing the high-level objectives for the product and defining the experience strategy to shape the concept. I partnered with field experts to explore the buying process.

Discovery

I conducted the research to explore who the buyers are, how they purchase properties, and their goals and motivations.

Additionally, I also analyzed the competition landscape to learn the pros and cons of the key players on the market.

Experience mapping

I did experience mapping to gain broad knowledge of the property buying process and ultimately to identify the pain points of the buying process.

Design execution

After identifying the opportunities, I created mock storyboards to narrate and structure the ideal experiences.

Ultimately, I delivered early designs and prototypes to communicate a new concept based on discovery findings.

Design validation

I collaborated with field experts to validate my assumptions and to see if I am on track.

During one experiment, I asked buyers for feedback on my concept.

Challenge

The real estate (online) market was already mature and competitive enough, so it was hard to imagine bringing up something innovative and meaningful that was differentiated. On the other hand, I also knew that it was not perfect and this motivated me to explore and experiment.

At the beginning, I did not know who the right people to interview were. I needed to experiment with the unknown - facing the fact I would end up in a blind alley as well as being opened up to the new things I may discover on the go.

Audience

  • Real estate buyers/sellers
  • Real estate mediators

The Approach

Design Process

From the beginning, I wanted to learn more about the experience journeys of real estate buyers. This allowed me to uncover details about the customers’ goals and expectations, identify the common touch-points they interacted with and their satisfaction with each stage of the property buying process. I needed to go deeper, below the surface to understand the entire context.

It was not a redesign of something existing, but rather an experiment in pursuit of finding new opportunities.

Finding the pains of buyers on their journeys reveals design problems to solve and brings business opportunities.

Discovery

Research

The discovery phase helped to explore who the buyers are, how they purchase properties and what their goals and motivations are. Learning about their experience journeys gave me an insight into the typical interactions and touch-points.

Additionally I also needed to understand and map the competition landscape to learn the pros and cons of the key players on the market.

Research strategy

Prior to experience mapping, I needed to have an input of information to work with. This I obtained doing research, interviewing various people that were involved in the real estate buying process. A qualitative research was my choice to start and explore.

Primary insights

To learn stories about the purchase journeys, pains, fears and expectations, I decided to conduct interviews. Since my primary objective was to explore the buying process, I recruited people that bought or were buying a property recently.

Other side of the coin

Later, another discussion led me to an idea of interviewing the other side of the coin - sellers and mediators (real estate agents or even mortgage brokers). Following this path was very beneficial to gain new views, especially to clarify the big picture. I learned a domain knowledge from field experts who buy and sell every day.

The challenge of research is facing the unknown.

Knowing the audience

Personas

The initial discovery delivered more information than just about property buyers - my primary objective. Exploring the entire process, I learned about more people and touch-points involved within the journey.

I started forming the future personas, synthesizing all the typical user characteristics. My goal was to identify several personas describing the basic needs, goals and behavior. These were distinct enough to represent different group of target users.

Having distilled more personas I had the possibility to prioritize where to focus in the first stages of the concept development. I grouped them as primary and secondary personas.

Personas - the audience
Capturing Requirements

Synthesizing User Goals

For each persona, I started to synthesize the user goals. This gave me a clearer idea of what their key objectives were. The user scenarios were my tool, helping me to stay focused on the user needs.

John (first-time buyer)

  • Saving time
  • Be first who see new offer
  • Affordable price
  • Safe transaction
  • Good location

John's Motivations
• I want to spend less time looking for a new offer. I don’t have time to refreshing real estate sites multiple times per day. • I want to be among the first ones who contact the seller, otherwise the best offers are gone quickly.
John's Goals
• I want to buy for a good price, relevant to the current market. • I want to make a safe purchase having all under control, avoiding malicious brokers or sellers. • I want to find a property in a specific location, so my kid can go to school nearby and I can jog in a park.
John's Scenario
• John knows the type of property he wants to buy, location and his budget (e.g. apartment in Prague, 2 rooms, 1 bathroom, ~ 80 sq m with balcony, park and school nearby - price below 4,500,000 CZK). • John looks for a new property online usually on his workplace’s computer or using his phone when commuting. John and his wife also enjoy searching together, usually after dinner or on weekends using a tablet. • John goes to several property search engines, to be sure he does not miss an offer. John looks for new offers several times per day. • John and his wife frequently email each other links to an offer they just found. • John and his wife store links of their favorite offers that they would like to see later. • John finds an interesting offer and calls the seller (or agent). John schedules a visit of the property, according to his and his wife's availability, so they can go together. • John visits a property, asks questions and makes notes. • John hires an external real estate expert to do a property technical check and overall review. • John and his wife decide to purchase.

John - first-time buyer

The user scenarios were my tool, helping me to stay focused on the user needs.

Clarifying the Big Picture

Experience Mapping

Buying a property involves complex customer interactions, stages and touch-points where users interact. As my primary objective was to discover and identify the pains of the buying process, I needed to capture the emotions to understand the problematic parts. I decided to apply experience mapping, as it seemed to be a good strategic method to capture customer behavior and interactions across channels and touch-points.

Research Insights

Uncovering Opportunities

Since I was trying to keep the user scenarios at a high level, it gave me the freedom to play with various concepts more easily. Seeing how buyers work revealed many opportunities.

Shared interests

Thanks to broader research I discovered that even mortgage brokers can be interested in finding a property to buy. A mortgage broker shares the same interests as the buyer. In other words, mortgage brokers are helping buyers to find a property for no added cost - just for commission from a given loan that the buyer would need anyway.

Customer fears

I also learned that buyers like to be very independent in the property search and buying process. The research uncovered a hidden fear of being a victim of cunning real estate agents or being charged a big commission for no value added. This turned out to be a very important finding, especially for an online service that could be helping these people take steps on their own.

Buyers with different needs

Learning more about buyers, I found out that they can have different motivations. For example, John (a first-time buyer) is looking for a new home for his family and has different requirements and needs to Robert (a real estate investor). In general, they both want to purchase a property, while having different goals, priorities and timing. This information is crucial for shaping the product based on the needs of each persona as well as for shaping the business strategy.

Problem-Solution Fit

Vision

I did not want to build another real estate search engine clone. Knowing the buyer pains I rather wanted to transform the complex old-school buying process into a service, that would be simple, useful and fun. I was aware that I cannot solve all users’ problems, therefore I needed to prioritize…

Saving time

My primary objective was to save buyers time. Users used to refresh various sites repetitively to ensure they didn’t miss their dream home offer. To solve this problem, I would simply alleviate users from doing the repetitive job.

Single-point search experience

Regarding the searching experiences, users frequently reported how difficult it was to use various, often cluttered property sites. The problem was with remembering what they already checked out or getting lost with different content presentations on various sites. Another issue was duplicates, since agencies regularly posted the same offer to multiple sites.

Here I came up with a solution to synchronize all unique offers on the market into one focus point.

Giving a hand to novice buyers

Let’s face it, how often do we buy a house or apartment in our lifetime? There are many novice buyers with zero experience in purchasing a property, without knowing the typical prices to be able to negotiate with the seller accordingly.

Being aware I could not guarantee John would buy for a good price, I came up with an idea to support him by providing smart insights into the average market price. This would give a hand especially to novice buyers, that are still learning what is cheap and what is expensive. Eventually it would be a good aid to prevent quick emotional decisions, buying overpriced property and later regretting a bad investment.

Building trust

Asking users to pass all searching responsibility to a bot (or service) is not an easy task. I realized that I would need to build users' trust as an important part of the experience strategy.

To give an example, imagine that John did not receive any message during past days because his criteria is very narrow. John starts doubting: "Does it really work? Am I missing good deals?"

To build the trust of your users you need to think a step ahead, predict and let them know that you have done so. In our story, I would assure John proactively, communicating that within the last 24 hours there were no new offers, to give a feeling that he was not forgotten or ignored. I would eventually let him know that this type of property appears less frequently in the region to encourage him to wait more or change his criteria.

To build the trust of your users you need to think a step ahead, predict and let them know that you have done so.

Experience Design

Hierarchical Task Analysis

My next task was to sketch the ideal user flows. Knowing user goals and pains I uncovered in previous steps, it was easier to see design opportunities and to lay out possible solutions. The only thing was to stay focused and ideate. Doing this I explored several concepts, before they would be narrowed by constraints (budget, technical complexity…).

To capture my ideas, I layed out several main user flows and interactions. Using the Hierarchical Task Analysis, I could understand what tasks users need to perform to achieve certain goals. I began exploring design opportunities to solve user problems for my primary personas: John (first-time buyer) and Robert (real estate investor).

John - User Journey #1
Design Execution

Visualizing Ideas

I would like to demonstrate several key interactions that I believe would solve the user problems previously described.

Speaking user's language

During the ideation phase, I came up with an idea to offer users a different way of searching. I wanted to experiment with a conversational UI using natural language and artificial intelligence (AI).

Users would be able to ask a bot through a messenger services (Facebook, Viber, Skype, etc.) in the same way they would text to a friend.

Instead of forcing them to have an individual app to search for property, this would make the service very accessible, without having the need to install a new application. Secondly, users would not have to learn a new interface to work with, since they could talk naturally in their own language.

Check out the following prototype, informed by the John's user journey #1; thus is how it could eventually look if John would talk to a bot using a messenger service.

Conversation with a bot

Hi there!

Hello, I am NemoBot. I will be assisting you in finding your dream home!

What is your name?

I am John

Hey John, nice to meet you! How can I help?

I want to buy an apartment in Prague.

Great choice! How many bedrooms would you need?

Sure, 3 bedroom apartment sounds easy to get in Prague. What is your budget John?

Do you have some other requirements?

Yes, I also need a balcony, a nearby park and a school in walking distance.

Awesome, here is what I've got…

Thank you! However, I don’t like any of those…

I understand. Stay tuned, your dream home will come soon for sure!

Would you like me to send you an alert when something comes across?

Yes please! Awesome!

Perfect, you are all set. I will be looking 24/7 and once there is something new, I will send you a message!

Thanks!

Goodbye John and have a great day. I will be in touch!

Now let’s see what the bot could already understand from the conversation with John. The first important step is to recognize the intent, which is mapping a phrase to a specific action that we can really provide.

Let's first focus on the intent, as it is the core concept in building the conversational UI, which drives the flow. In the sample conversation, John says:

John: "I want to buy an apartment in Prague." 
In this case, the intent is "to buy something", thus mapping a phrase to a specific action. Therefore, it allows me to take different actions based on the intent. If the intent would be "rent something", the bot would reply to John with different results, etc.

Though the intent is important, it is not enough, since the bot needs to know what John wants to buy. Let's extract all necessary parameters (also known as entities) from the following phrase as an example:

John: "I want to buy an apartment in Prague." 
In this example, words "apartment" and "Prague" are the entities.

Bots also collect other entities like "balcony", "a park located nearby", "a school in walking distance".

Bot: "Do you have some other requirements?"
John: "Yes, I also need a balcony, a park located nearby and a school in walking distance."

Note:: While collecting parameters, it is crucial to stay in the context of the intent "buying something". Leveraging the context will reduce the friction of the user experience and will help the bot know what to expect.

The following information is what the bot could understand from the entire conversation shown above. I distinguished the entities by their type to illustrate the richness and nature of the information collected.

      - Intents: Buy
      - Entities:
          - Name of user: John
          - Property type: Apartment
          - Location: Prague
              - park nearby
              - school in walking distance
          - Bedrooms: 2 bedrooms
          - Size: >= 80 sq m
          - Features: balcony
          - Price: <= 4.500.000 CZK
Creating alerts

This example intends to demonstrate how users could easily turn their current search into a new alert using a bot or on the web. Setting up an alert would update the user instantly any time a new offer appears on the market.

Get alerts for new real estate offers.
Get alerts for new real estate offers.
Recieving alerts

Alert notifications on mobile demonstrates how users could be informed more naturally, without having an email lost in their mailbox and with the ability to act immediately.

Mobile notifications as an alert for a new property.
Staying on Track

Validation & Feedback

As the project was in a very early stage - exploring and defining a new concept - I needed feedback. In the initial stages I partnered with field experts, mostly real estate agents, that were my bridge to buyers. Field experts and buyers helped me to validate if I was on track, to learn more about the buying process.

Reviewing user journeys

That happened as form of several discussions, where we reviewed the user journeys as well as finding out the business part of it - a value proposition. We were shaping the value for the user, to deliver a meaningful service that people are willing to pay for.

The Experiment

In parallel, I needed to evaluate if the idea had the potential to stay viable. However, the product did not exist. Since I treated it as an experiment, I prepared a very simple automated mechanism, that was matching buyers’ criteria. For this evaluation, I tried to stay minimalistic to evaluate fast. I decided to use an email to deliver the information to the users.

The recipients in the experiment were various recruited people from interview, that were in the buying process. I also asked field experts, that gave me input on it on a regular basis, since they were constantly looking for properties for their clients - buyers.

…shaping a value for the user, to deliver a meaningful service that people are willing to pay for.

Final Outcomes

Results

I went through an initial research, mapping the customer experience journeys to understand all pains that buyers were facing. I did a comprehensive analysis to synthesize all my findings to define the personas and their scenarios.

After identifying the opportunities, I mocked up storyboards to narrate and structure the ideal experiences. Ultimately, I delivered early designs and prototypes to communicate a new concept based on discovery findings.

During the design process, I needed to stay aligned with business as well. My goal was to define a meaningful product that was adjusted to the user and market needs.

As a result of my pursuit, I defined the Unique Value Proposition using the user-centered design canvas  tool.

User Centered Design Canvas