Over the past six months a taskforce of data scientists and strategists from Zenith has been developing sophisticated automation of digital planning using the network’s machine-learning technology and bespoke algorithms.
This radical automation of digital planning is being done using cloud-based technology, with the client retaining full ownership of their first-party data throughout the process.
Vittorio Bonori, Global Brand President at Zenith, said: “Zenith is leading the way in changing the business model for digital. This important programme is part of our strategy to leverage the power of data and technology to drive profitable growth for our clients.”
First live test with Aviva: 6% to 10 % CPQ savings
Using live Aviva campaigns, the taskforce collected advertising cookie data from the technology stack of a leading demand-side platform (DSP) and matched it with corresponding first party sales data. Applying Zenith’s machine learning algorithm, the taskforce was able to precisely attribute sales conversions to specific digital interactions.
Zenith was able to automatically optimise Aviva’s digital planning by pushing the algorithm output back into the DSP’s stack. This dramatic move closed the automation loop – data collection, attribution and a full set of planning changes across multiple digital touchpoints all done automatically.
This application of machine learning saw Aviva benefit from a 6% cost-per-quote (CPQ) improvement on car search through implementation of the automation programme. For display, Aviva saw a 10% improvement in CPQ through automation
James Turner, Head of Marketing (Trading) at Aviva, said: “We’re delighted that as part of our commitment to digital and media transformation at Aviva we are breaking new ground with this pilot automation of our search and display. The benefits of attribution modelling will be realised in terms of ROI improvement as well as through operational efficiency.”
Further enhancements to be introduced
But Zenith is not stopping there, the network is adding first party drivers-of-demand data into the algorithm in order to enhance the effectiveness of the automated planning changes. In this way, data – such as how price affects sales or the success of creative assets – will be fed into the automated optimisation.
