PropertyGuru App Redesign
Introduction
PropertyGuru is Singapore’s largest online property group. Agents, property sellers and property buyers will probably have no problems recognising this name and what they do. As the leader in the market, most users have little choice but to use PropertyGuru as their platform of choice despite the usability issues in PropertyGuru’s *previous* mobile application. However, the rise of competitors like 99.co with aesthetically pleasing and usable mobile applications have likely encouraged PropertyGuru to introduce an updated version for their mobile application. This new application has been a 180 degree turn around for PropertyGuru, but it still has areas for us to look at and work on.
About this project
In this project, I worked with Megan and Shumei to look at the areas in which PropertyGuru’s application can improve upon. The current mobile application looks rather pleasing but at the same time there were several areas which required redesigns. We evaluated the application, interviewed users to find out about their pain points, brainstormed about solutions and eventually provided some designs as business opportunities and solutions.
User Research
In an attempt to understand the problems that appear in PropertyGuru’s applications, we set out for user research and interviewed property agents who use the PropertyGuru application as well as property buyers who might browse for properties on the PropertyGuru application as well.
From the property agents, we found that most people have little choice but to use PropertyGuru as their choice of platform because of how it captures the largest market. However, they are getting increasingly frustrated at how PropertyGuru is charging higher and higher prices. Some also misuse the application to post more of the same listings in order to get more hits — others will even leave outdated listings on the application in order to capture buyers that are looking for a certain type of property.
From property buyers, we learnt that some find it hard to trust agents as there are plenty of dishonest agents and what they really need is the ability to be able to find the best deals and be able to jump at these deals. The PropertyGuru application was also pretty usable in their opinions and they generally agreed that even though 99.co’s application slightly more usable, it was not good enough to make them move over to 99.co because 99.co’s listings are way more outdated.
Buyers also mentioned that they tend to prefer pretty pictures as long as they are realistic and wanted every listing to be able to provide floor plans of the property.
Some buyers also lamented that PropertyGuru did not have a good valuation feature that could help them to better estimate the prices of a property. They mentioned that having the last sold property price was an acceptable substitute but at the very least they would still like to see the market trends which were on the websites but not on the application.
So based on what we found from the users, we did an affinity map and here’s a summary of what we have gathered:
- I want to be informed of the best deals
- I want an honest agent
- I want accurate valuation of the property
- I want informative pictures about the listing.
- I want to see the floor plans.
- I want to see more useful information on the listings.
- *Special features
Problems found through Heuristic Evaluations
While learning about the PropertyGuru application, we found some issues in which could cause users problems. There are actually very many that could be listed but we can just showcase a few that are more important for PropertyGuru to look at.
1 — Navigation Menus
For some reason, PropertyGuru has a few pages with a different navigation menu. We feel that this is a problem because this can be confusing to many people. Additionally, the navigation is not available on most pages. This causes trouble especially when users navigate really deep into the application and have no easy way to return to the main page.
2 — Price Trend Graph
A user of the PropertyGuru website would probably want to look for this price trend graph on the application as well in order to see the market rate that a property is going at. However, this graph cannot be found on the application and to be honest, this is one of the most important things that a buyer and seller would like to look at.
3 — No information in certain tabs in the amenities page
When users enter the amenities page, they are offered a map with several tabs on it to help users to look at the amenities that are located in the area. However, there are plenty of times when these information do not appear in certain tabs and this is something that is important for PropertyGuru to look at so that the application does not look unpolished/incomplete.
4 — The map view is very cluttered
So as we can see from the above picture, there are way too many marker icons on the map. This makes it difficult for users to select an area and click onto individual markers because many markers are overlapping one another.
5 — Database on the website and the application is different
Users who have been using the website know that PropertyGuru actually allows for some types of searches. However, once they go into the mobile application, they will not be pleased to find that some of these searches will end up becoming unavailable to them now…
6 — Alerts are not truly alerts in the application
The word “alert” lets one know that they would be notified in some way or sorts. People expect that they would receive some kind of notifications— likely on new listings that meet their criteria. For PropertyGuru’s alert, this is only true to the second part of the statement whereby there might be new listings that meet their criteria, but users do not get notified. What they do is to re-use a saved set of search parameters when they click on an “alert list” that helps them to do a new search based on those previously saved search parameters. “New” listings require them to remember what has not appeared before.
7 — Price parameter in search screen has uneven jumps towards the higher price spectrum
In PropertyGuru’s search page, the range for the price parameters become much larger at the top end of the price spectrum but some users might only want to use a lower budget e.g. $40million instead of $50million. This results in them having results they may not want to see at all in the searches.
Competitive Analysis
While doing competitive analysis, we compared PropertyGuru to 99.co as well as AirBnB. Although AirBnB is not an app for buying/selling property, as a accommodation rental app, its primary focus is still very similar and its usability is also top notch — thus we decided to compare with it as well. We made sure to take references from these two other sites when improving the design and try to incorporate the most useful parts for the PropertyGuru application.
Personas and Customer Journey Maps
From all our research findings, we finalised on two personas and did customer journey maps for each of them.
Business Opportunities
We worked on a Business Model Canvas to help us better go through the flow and understand how PropertyGuru currently makes it money. This was done to help us in our search for business opportunities.
Along the way, we also discovered some useful data that can be found on PropertyGuru and Alexa.com. We know that there are 12 million website visitors in a month, of which, 16% of them are actually foreigners trying to access the site for Singapore properties. These are the potential customers, and the number of these potential foreign customers actually amount to 1.9 million people.
Through our design studio, we came up with the idea to introduce a property tour for property buyers of which we could charge an amount to bring them around from property to property. The advantage of such a tour is to provide knowledge to new property buyers on what type of properties they could get with their money and subsequently help them to gauge better if a property they would ever view in future would be worth buying. This would probably be appealing to the foreign customers who know little about the country and the types of properties. We believe that they would be likely to pay a price for it.
We further calculated that if there were just 1% of them willing to sign up for our tours (which we think can be priced at $200 per person), PropertyGuru could make up to $28 million in potential revenue! Of course, this is merely an estimate but we expect that there will definitely be demand for it.
Prototyping and Iterating
When coming up with the prototype, we had heavy focus on keeping PropertyGuru’s current user interface as we felt that it was usable enough and aesthetically acceptable. However, if a feature or area was considered unusable, we would take special notice to figure out how to change it. For this project, we did up a hi-fidelity prototype.
One of the things that users wanted was to be informed of the best deals. Additionally, an area that PropertyGuru could have improved was its alerts function. We worked on improving the alerts and make it work like a notification, but during the first iteration we forgot about putting the “red notification number”. This caused users to not know where to find their alerts so we added it in later.
However, we expect that the “red notification number” could distract users and direct their attention to the wrong place causing them to not know how to do a search. Thus, we also added icons beside the headers to help users recognise what to do more easily without having to read the headers.
When alerts get created, users had no clue which was their newly created alerts. In our iteration, we added the “new” icon and switched the latest alerts to the top.
We have also added navigation menus in every page to make it much easier to navigate than before. The previous way to get back to the first page was to click the “back button” many times until you reach the base page.
We added the “new” icon to the listings as well so that it becomes obvious which listings are the ones that the user has been alerted about.
Originally, the “take virtual tour” feature was not coloured in yellow but gray. However, because we had to move it from its original spot (where the favourite and share icons are), we found that it needed to be more visible and decided to make it yellow.
Another problem faced by users are PropertyGuru’s lack of a floor plan which we solved by making it a mandatory item in PropertyGuru’s lists.
We also added the estimation valuation calculator here as well in order to help tackle more user needs.
This estimated valuation feature is meant to be as accurate as possible in order to help users to be more informed of their buys.
We also improved on the map by switching out the marker icons with marker clusterers so that it looks less congested.
When one zooms into the map and clicks on a single marker icon, it will also bring out the information window that gives a property listing’s information. Amenities will also be shown here in order to provide more information about a place a user might be looking at.
Aside from property listings, we also have agent listings to allow users to learn more about the agents that put listings up on PropertyGuru.
In the agent profile page, we added a “reviews” button for users to leave reviews and also a “rating” info to show what users they think of these agents.
If users click into the reviews button, they would be brought to this page whereby they can actually give their agents a rating.
We also added a new chat feature to help users be able to connect with agents through the application if they are uncomfortable with calling or messaging.
Finally, we also created the new “browsing tours” features which was mainly made for potential foreign customers trying to look for property in Singapore.
Usability Tests
User Tests
We looked for 5 more people (buyers and agents) to help us out with our usability tests to see if there were more areas we needed to improve and iterate on. It appears that the application was generally passable with few mistakes. People were able to complete the tasks for the scenarios fairly well.
However, we did get some feedback on certain areas which we iterated on. These areas were talked about previously in the prototyping section.
A/B/C Testing
Aside from our user tests, we also did A/B/C testing on the price filter for the search page. We wanted to test between mobile keyboards, mobile horizontal sliders and PropertyGuru’s original vertical sliders to see which of them would allow for the fastest input.
We conducted it by getting 30 different participants to do the testing. Each person will be timed from the moment they entered the application and keyed in the price of 400000 to 800000 and the PSF of 150 to 1000 (except for on the original application it would be 300 to 1000).
From our results, we see that the keyboard ends up being the fastest, and is even faster than PropertyGuru’s original vertical slider. However, we still feel that users may find it a hassle and prefer to just use the slider instead as it does not require a different type of touch input despite keyboards being faster. Nonetheless, we see the value of having the text area and we think that the best way to go about this is to probably have the text area together with the slider.
Project Planning/Management (Agile)
If this project were to be worked on for real, this is the proposed 6 months plan for the UX team and the development team
The UX team runs a sprint ahead of the development team and user acceptance test will happen every two months. The UX team will do usability tests in every sprint while the development team will work on whatever task they were asked to accomplish. Finally, we will end off with both the UX team and the development team ending together on the third and final user acceptance test.
3 Biggest Takeaways
- In this project, I learnt that application redesign may not always have to be a full revamp. In the case of PropertyGuru, the application itself is already plenty usable for most users although is does have issues here and there. For the sake of the project, we found the most pressing issues and made sure to solve them to the best of our abilities.
- In our project discussions, we tended to have split views on what interactions can improve a certain feature. This is actually a good thing rather than a bad thing as these areas are where we can do A/B testing in order to figure out the best.
- The design studio together with service design as an additional solution made for interesting concepts that we incorporated. We actually came up with a “Hall of Fame/Shame” (chose not to implement), the “Browsing Tours” and the addition of the “Virtual Tours”. These were areas which did little to solve user needs and pain points but added a new dimension in which could attract/please users.
Presentation Slides (PDF):
https://drive.google.com/open?id=0B76twXK43Bw-eE82LWNEMlNvams
Axure Prototype:
