Aaron Cutter and Luke Cassar share their impressions of the 2018 ADMA Data Day, including how marketing is getting more personal through better use of data and machine learning, and opportunities for actuaries to lead the way in developing smarter analytics.
The Association for Data-Driven Marketing & Advertising (ADMA) is the largest marketing and advertising association in Australia. On 23 February 2018 in Melbourne and 26 February 2018 in Sydney, ADMA hosted Data Day for the 14th year. ADMA advertises Data Day as the “largest, most comprehensive data-driven conference in Australia” and seeks to aid its members to connect data, technology and marketing. This year’s conference explored topics such as:
- customer experience;
- converging technology; and
- data and market strategy.
ADMA Data Day 2018 had over 1,000 attendees across the two days in Melbourne and Sydney.
The word “data” has appeared six times in this article’s opening paragraph, so it’s perhaps no surprise that such an event was a golden opportunity for actuaries to enrich their understanding of how their data-based skills can be used in a world beyond traditional actuarial roles in the financial services industry.
This article will take you through the key messages of the day, which featured talks from the likes of online retailer The Iconic, ANZ, Woolworths Food Group, Ooh! Media and more.
Themes of Data Day
ADMA’s CEO Jodie Sangster gave the opening address and suggested that the key themes of this year’s conference would be:
Businesses are now looking at offering an end-to-end customer experience
- Artificial intelligence
Including the application of machine learning and smarter data analytics to drive marketing strategy and improve customer experience
- Peer-to-peer learning
There may be a significant amount of up-skilling required for some organisations
Firstly, in the sense that customer privacy is always an issue and that there is a risk of crossing “the creepy line” of knowing too much about the customer and secondly, ensuring that offering value to the customer is always at the centre.
ADMA CEO Jodie Sangster © Photographic Memory Liz Moore of Telstra © Photographic Memory
Using machine learning to drive marketing outcomes
The speakers at Data Day were from a wide range of industries. Steve Lok from The Economist shared how data-driven marketing and automated customisation of The Economist’s web portal based on an individual subscriber’s demographics and topic preferences resulted in a flip of their main source of revenue from advertising to subscriptions.
We also heard talks on a range of different aspects of data analytics use in marketing. Liz Moore from Telstra convinced us of the power of Bayesian belief network models to determine causal relationships between marketing activities and business performance. This helps Telstra understand which marketing channels worked best where and for what kind of customer.
From the sessions touching on machine learning, two key messages emerged:
- The more data, the better – one presenter talked about testing as many as 5,000 explanatory variables in their models
- Timeliness is paramount – in some cases, analysis was automated and turned around almost instantly
Revealing the Truth
Willem Paling from IAG talked about how marketers can effectively engage with artificial intelligence (AI) and machine learning. One interesting insight is the confirmation of something many internet users have thought for a while – targeted digital advertising based on some previous interaction with a product (perhaps getting half way through an insurance quote) is more likely to turn away a customer than lure them back in. This was another example of how assumptions which may seem like common sense (that people who were looking at a product online are more likely to purchase it) can be overturned by data and analytics.
Managing data-driven thinking
Patrick McQuaid from National Australia Bank talked about some of the challenges in leading a team in a large organisation to think away from a product-focus to a customer-focus.
Everard Hunder from Monash IVF talked about the importance of testing marketing strategies. Testing, in the sense of marketing, is probably not something us actuaries and data scientists working outside of marketing departments would be familiar with. For Monash IVF it involved running A/B tests on small marketing campaigns. For example, sending two different emails to two groups of test subjects and determining which style of email was more successful. Some example results were very unintuitive, highlighting the importance of A/B testing. Monash IVF allocate individuals budgets each year for staff to conduct such testing. Such an approach is an innovative way for staff to compete against each other (in a collegiate way) in order to find the most effective marketing strategies.
Considerations for actuaries
There are some clear parallels in data-driven marketing and advertising and the work that us actuaries have expertise in. Finding new customers for any organisation is a lot like finding good risks for an insurer. Our experience in building pricing models, incorporating high-level statistical modelling, sets us up perfectly for applying these skills to marketing and advertising. Actuaries working in or with insurers also require the ability to see the whole picture and understand how underwriting, claims management, reserves, profits and capital are all intertwined. This breadth is useful in a marketing context too. Having an ability to zoom out gives us an advantage in being able to understand how analytics can fit into improving the customer journey, from initial contact through to user experience.
"Our [actuaries] experience in building pricing models, incorporating high-level statistical modelling, sets us up perfectly for applying these skills to marketing and advertising."
We saw talks featuring a sophisticated use of machine learning techniques and AI, but at the same time talked to a lot of people who were there to see the forefront of data-driven marketing to inspire them to bring focus to data science and analytics in their own organisations. Clearly, there are opportunities for actuaries to lead the way in developing analytics in organisations outside of financial services.
So we’ll see you there next year!
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivatives CC BY-NC-ND Version 3.0 (CC Australia ported licence).
CPD Actuaries Institute Members can claim two CPD points for every hour of reading articles on Actuaries Digital.