How improved the sign-up ratio by 58%, optimising the website search

Investors in Property are the leading international agency specialising in the sales and marketing of ski property in the Alps. We have been selling ski chalets and apartments for over 35 years.

I started working with in January 2022 when the client asked me to make the CRO audit for them. Although I will only briefly mention how we worked, you can read more about CRO audit here. 

The process started with the meeting and Q&A session, where I raised the following questions:

  • Who is your target audience?
  • Which acquisition channels do you use to drive traffic to the website? How much money do you spend on Paid Ads, and how much does one lead cost for you?
  • How do you measure revenue?  What goals in Google Analytics are you using daily? Could you explain the main aim of each goal?
  • What is the focus of your company? Which KPIs do you track?
  • What is the current conversion rate?
  • What problem does the company solve for end customers?
  • Who are your direct (and the strongest) competitors? 

When I received the answers to these questions and analysed the existing analytics setup, I understood that there is a batch of events that we should add to Google Analytics. 

The next step was to prepare the Google Analytics implementation specification and work with Google Tag Manager to set up events. Because some events needed the installation of datalayer on the website, I worked with the engineering team for a few days. 

Google Analytics Implementation Specification

The main focus was on the website search. This is the primary tool that website visitors use on such websites. I remembered it from analysing my previous client –

After we installed all events correctly for Google Analytics and Hotjar, we started to gather new data. It took two weeks to collect the fresh new data. 

The analysis included working with Google Analytics, Google Sheets, and Hotjar and the tremendous amount of time spent with Python to work with the datasets from Google Analytics to analyse how website visitors use the search and how we could improve it. 

The main insights we received were:

  • Many website visitors used the search but didn’t select any country. The exciting aspect was that 90% of those users who didn’t specify the country also didn’t select bedroom sizes or resorts. However, they set the currency (EUR, GBP, CHF). That meant that users were trying to find the option to see the prices of the properties in their local currency instead of the currency where the property is located. Thus, I recommended that the client add the currency selector on the website. Apart from that, I also recommended automatically detecting visitors’ countries and showing the prices in the correct currency when they land on the website.
  • The most popular filter on advanced search was chalets because it was placed at the top. However, two filters were used so often that it clearly showed how important they were for website visitors. Two filters were “Available for all Foreign Buyers” and “Ski in/ Ski out”. Therefore, we recommend that the client looks at how to make these two fields more prominent.
  • Most visitors on the search results page used “Latest”. Even when users selected “Least expensive first” or “Most expensive first”, they returned to “Latest”. The conversion rate (visitors -> new registration) of the segment of users who used “latest” has a higher conversion rate. We encouraged the client to use it as the default sort option on the listing page. 
  • The main reason why people received zero-results pages is that they mistyped the resort name. We encouraged the client to change the “resort” field from “free open input” to “predefined input”. When users could select the resort from the predefined list, they won’t make a mistake. 

After these and many other insights, the client optimised the homepage and the website search.  You can see below the old and the new homepages.

Old HomepageNew Homepage

The split experiment was conducted through Google Optimise and lasted three weeks. We receive 95% statistically significant results with a 58% increase in the sign-up ratio. 

VariantWebsite visitorsSign-upsSign-up Ratio
Old Homepage64983245.0%
New Homepage64525097.9%
statistical significance of the experiment results

We deployed the winner of the experiment in a few days. Contact me through this form if you are interested in optimising your website