A need for insights with speed
Domain had an in-house analytics system built to measure conversion and engagement throughout different parts of the website. Its marketing team was also using a number of well-known data visualisation tools for their dashboarding and reporting needs.
When product managers tried to use these, they faced a few challenges. Given the amount of events in these platforms, it would take five to ten minutes to generate one report, especially when segments and funnels were added. They also found it time-consuming and sometimes challenging to self-serve with these tools.
They were simply not able to explore their data freely and dive deeper with breakdowns with the flexibility and speed that the business required.
The pitfalls of not having quick access to insights
“We’re in a very competitive industry. We consider ourselves to be bold and innovative — a company that gets features to market before our competitor does. We don’t want to wait weeks to see whether an experiment has moved the needle or wait for reports to be built.”, said Mike Salway, Principal Product Manager at Domain.
Mike Salway and his team have a firm belief that being able to quickly and easily self-serve on key product metrics is essential to making fast product decisions.
“Too often in the past, we made design or UX decisions based on intuition.”, he continued. The team wanted to be able to understand if new feature/experiment launches made a difference to their key metrics so that they could decide to continue, or try something else.
Defining the metrics that matter
Mike Salway looks after the Seekers (anyone looking to buy or rent property) experience. This includes the main homepage, the search results page, the listings page, and anything around that core funnel. One of the main metrics that matters is the number of Weekly Active Property Users (i.e. people who view at least two listings in a week).
“This lets us capture users who want to do more than just look at a single property because looking at two or more shows their higher intent. The more people that we can get to view more listings, that converts to higher enquiries, which leads to higher retention. And if we can drive retention, they’re going to come back and look at more properties.”, he explained.
These users fall into two categories:
- Casual browsers – people who might not purchase for years, but are getting ideas and figuring out what they can afford
- High intent users – people who’re interested to transact and show this intent through various actions like adding an inspection to their calendar or shortlisting a property
Focusing on converting high intent users to enquiries enables Domain to deliver value to their paying agent customers. It is also important for Domain to provide great features and experiences to casual browsers. When they’re ready to transact in the future, Domain should be top of mind.
The team wanted a solution that allows them to:
- Figure out baseline metrics
- Use data to drive new feature development
- Freely explore user engagement data to spark new feature ideas
- Track user retention
- Understand the user journey
Putting Mixpanel to Work
Capturing more long-term users with school search features
The Domain SEO team identified that there was significant inbound traffic from school related searches. In Australia, parents can send their children to public schools if they live within the school catchment for those schools. To be able to enroll their children in sought after schools, it’s important to check if their potential home would be within the catchment area for that school.
Mike Salway had experience with Mixpanel at a previous company and when he saw that Domain was already trying the solution out, he decided to spearhead the effort. The team set up instrumentation in Mixpanel to measure traffic coming into the school catchment pages on Domain that showed the school catchment for public schools.
With this, the team found that “property-hunting parents”, or users who want to buy or rent a property in a particular school catchment, retained more because they were inquiring more than the average user. This was identified as a cohort of higher intent users.
“Our hypothesis was that we should build out our school-related features in a better way, i.e. more info for the user, better UX on the pages, newer and more up to date data because they change with time. So we created a new set of features for this cohort, got new data sources loaded, created new experiences on the school catchment page, and then better optimized the page for SEO.”, shared Mike Salway.
After launching the new features, the team tracked the metrics in Mixpanel and were able to see that users of those features retained at more than two times the average, and they also made two times more enquiries than average.
They also created cohorts in Mixpanel to identify users who came in via SEO searches related to schools, and a separate cohort for those who used school-related features on Domain but did not come via SEO. This allowed the team to compare these two groups against their usage of the features and their funnel conversion rates.
Powering product decisions with solid insights
The insights did not stop with the school search features.
Domain users can filter a property by type, bedrooms, parking and more. The bed, bathroom, and car filters on the website are an awkward slider experience that confused users. They can’t figure out how to set the bedroom and car filters properly.
By using Mixpanel reports, the product team dug into what the most commonly used filters were, and whether people were using a range of values or specific values when filtering.
With their findings, they can now redesign the experience knowing that these decisions are made based on actual user behaviour insights, and not someone’s subjective opinion.
Data-backed decisions came into play too when a designer wanted to remove breadcrumbs from a particular screen on the platform.
“I pulled up Mixpanel to show what percentage of users use the breadcrumbs on desktop, mobile, and tablet. The designer had expected it to be less than 1% of the users but we saw that it was actually 5-8% depending on device type. It was a quick and easy decision to keep the breadcrumbs.”, said Mike Salway.
Why Domain Loves Mixpanel
Full of features that drive value
Mixpanel contains many features that the Domain team has come to rely on and love.
- The conversion funnels help Domain to keep on top of driving Seekers to make enquiries, which allows Domain to demonstrate value to their paying agent users
- Retention reports help the team to track retention over time, break them down by different cohorts, and compare the different cohorts to see how they differ from the average user or a high intent user
- Domain monitors the usage of various features and the average number of listing views per user with the Insights report. They break this down by cohort and make comparisons with other cohorts to better understand their users
Professional Services that are partners for success
The Mixpanel Professional Services team also shines by partnering Domain to help ensure their success.
“They’re just fantastic to have as an on-hand resource to quickly ask questions about the best way to instrument specific events based on what questions I’m trying to answer with my data, or how I might want to break down the data later. It pays for itself in the saved time or the additional value you can get out of the product because you’re using it the right way,” said Mike Salway.
The Domain team also looks to Mixpanel Professional Services for advice, tips, and help to create advanced reports.
Delivering Value Across the Company
The insights in Mixpanel have proven valuable not just for the product and design teams responsible for Seekers.
Domain’s content audience team builds the content that brings traffic to Domain and Mixpanel now allows them to effectively measure the users that are acquired with this content and look at how they can be encouraged to look at the property listings and become members.
Mixpanel continues to be rolled out to other product teams within Domain, including the Domain Loan Finder and Domain for Owners products.