Invocare is a leading provider of funeral services in Australia, New Zealand and Singapore. In Australia their portfolio of over 250 locations is split among 3 national brands, 5 state brands and almost 40 local homes. The 40 local homes each have their own identity, and website. Among the local homes there are also a number that are focussed on funeral services for specific cultural backgrounds or beliefs.

Primary Objectives

Deeper and Smarter understanding of customers’ needs through thorough data.

Improve conversion rate.

Improve cost per conversion.

The challenge

Invocare welcomed a new Managing Director & CEO in 2021. He, and a refreshed Executive Leadership Team, have reset the company’s strategic vision for the next 5 years through to 2025.

Management has turned to digital as a means of helping achieve this vision, with specific focus on digital as a way to:

• Innovate and diversify to expand addressable market and meet future customer needs
• Excel in servicing customer needs and grow share of market value
• Optimise foundations for sustainable returns

The complexity

Delivering efficiency across multiple brands in multiple locations that address multiple customer needs. That means either not competing unnecessarily, or competing efficiently, for the same advertising placements as user location and Invocare service areas dictate. For example, at any given time there may be a couple of Invocare National Brands, an Invocare State Brand, and another one or two Invocare Local Homes all able to service the same customer. With over 45 domains each with their own AdWords Account this was a challenging request.

The strategy

FIRST took the approach that each Location needed to be treated individually to achieve Invocare’s desired results.

Within each group – National Brands, State Brands and Local Homes – each funeral home was allocated a dedicated territory for advertising purposes, as well as for market share analysis. Thus for one National Brand, no two branches would be targeted for advertising in the same locations. However, a different National or State Brand, or Local Home may share a target area.

This approach meant that there was scope for competition between each National Brand, State Brand and Local Home, as Invocare still had the opportunity for two or more of their Brands to appear in the Search Results. However, having each of them allocated specific geographic areas reduced wasted spend and unneeded competition.

Google Ads Accounts were modified to match this structure, with National Brands and State Brands left in their own Account, while Local Homes were rolled into a single Account, with each home having a dedicated group of campaigns addressing various aspects of the market relevant to that home.

In addition to the restructuring, FIRST took the opportunity to make many optimisation improvements including widespread improvement in conversion tracking driving the use of machine learning bidding strategies.

This strategy is not only a strong foundation for Search Advertising, but can be a foundation for other advertising and marketing initiatives including GDN, YouTube, and Discovery.

The results

First took 8 months of AdWords data from Invocare’s Accounts from prior to when we took this approach to their Campaigns and compared it to performance after the changes were implemented.

During this time there was a negligible change in Media Spend (-1.15%) so any results achieved are purely as a result of a change in Strategy and use of Google tools.

Across all of their Accounts Invocare achieved the following results –

• Cost per Click; -3%
• Click Through Rate; +9%
• Cost per Lead; -13%
• Conversion Rate; +26%
• Total Leads; +29.75%

The conclusion

First’s approach to this Account drove an improvement in Cost per Click (meaning Invocare are getting more visitors for the same budget), an increase in the Click Through Rate (showing that the new approach to ad targeting and bid strategy is presenting relevant ads to searchers), and a reduction in the total Cost Per Lead (proving that not only are the searchers getting a relevant ad but they are also being presented with the right content and landing pages).

The tools used

First used many Google tools to build out and manage these Accounts right down to the Keywords level.

Google Search Console – First used Search Query Reports to identify high volume, low performing terms that could benefit from having their own dedicated Paid Search support. First also used this data to find Keywords that performed well Organically and that we could then remove from the Paid Search Accounts (under strict monitoring to ensure there was no loss of Traffic or Conversions) and reallocate the budget from those Keywords to other underperforming terms.

Google My Business – First used the Google My Business data to help allocate Australian Bureau of Statistics (ABS) SA2 areas and postcodes to each Location and to identify new “Near Me” Search Terms that we then used to create new Ad Groups.

BigQuery – Big Query GIS functions were used to allocate (ABS) SA2 and postcodes to branches for market analysis and also location targeting in Google Ads. Postcode data was joined with suburb name and keywords to generate suburb campaigns, RSA ads, keyword groups, and ad groups by brand and branch for import of campaigns. This allowed us to both create the keywords and ads at scale (every Suburb in that Postcode multiplied by every Keyword combination) and to keep them updated as we added new terms over time.

We used BigQuery to ingest 20 years of ABS data on mortality rate in Australia at an SA level, and then map these numbers back to each branch as an average per month. We then took Invocare’s Case Data history by branch and did the same. This gave us a proxy for their Market Share at a National, State, and Branch level, and a forecast of the expected number of Cases for each of them in the next month. This was used to inform where budget needed to be allocated at the Brand and Home level to support areas of underperformance.

We also incorporated:

Responsive Search Ads – Generated at scale out of Big Query in combination with campaign keywords to deliver great ad scores and relevancy.
Machine learning bidding – A combination of Max Conversions to drive the most leads possible and Target CPA bidding has been used to drive efficiency

Google DataStudio – Data from Big Query (GIS, SearchConsole and more), as well as data from Google Analytics and Google Ads was used in DataStudio for comprehensive analysis and advanced automated reporting. Reports created included senior management level reporting at weekly and monthly level, mapping of branches by grouping, search console performance and error reporting, and more.