Fairfax successfully uses Big Data to direct content recommendations, ad targeting
Ideas Blog | 03 February 2015
Editor’s note: This is one of 17 case studies featured in INMA’s strategic report “Making Big Data Smarter For Media Companies,” released in December.
Australia’s Fairfax Media is using data to guide an array of interactions with users, says Andrew Lam-Po-Tang, Fairfax Media’ s chief information officer, chief technology officer, and director of group services.
Specifically, Fairfax is building propensity models to improve its content recommendations, with the goals of:
- Increasing reader engagement and satisfaction.
- Making subscriptions more appealing.
- Increasing its advertising inventory.
Fairfax has also experimented with different means of surfacing those recommendations to readers.
As potential subscribers move through its acquisition funnel, Fairfax has analysed their behaviour to optimise conversion. The company also uses both multivariate and A/B testing for optimisation and is developing churn analysis and predictive modeling for existing print and digital subscribers to help reduce churn.
Data is also part of the Fairfax strategy to improve ad targeting and provide advertisers with an array of targeting options. The company is performing a yield analysis of its online inventory to optimise sales and pricing. It has launched a data services business providing an array of data and analytical services to advertisers.
Fairfax also has launched new businesses in events and content marketing. It plans to expand its data analytics by using its propensity models to drive growth in those revenue streams.
Lam-Po-Tang describes two key lessons Fairfax has learned:
1. Have a defined mission from the start.
After partnering with a variety of data analysis and ad-serving vendors in these programmes, “the caveat emptor is that, without clear business objectives to focus efforts and measure results, it can be difficult to identify and optimise bang for buck.”
2. Have patience as you embrace new digital opportunities.
Five months after launching its main digital subscription packages and bundles, Lam-Po-Tang says, Fairfax began a pilot to analyse digital subscriber churn.
“It soon became clear that it was too soon to complete an effective analysis because launch promotions were dominating subscriber behaviour, and there was insufficient data to provide insight into what might drive ‘steady state’ churn,” he says.
While using data in so many different ways, Fairfax is careful to ensure customer privacy, using its information security policy and security actions to safeguard customer data throughout its network.
The company is transparent with users about what data it will collect and works to secure their informed consent. When customers make inquiries about their data, Fairfax responds in compliance with local laws and regulations.
Lam-Po-Tang says some of Fairfax’s more modest data strategies have been active for a considerable period, but its more ambitious efforts are relatively new. Still, most of these pilot programmes have succeeded in generating a positive return on investment within just a few months.