While the guide, One Not Everyone, highlights how personalisation is part of brand evolution it warns about ignoring the limitations of data and the potential downside of delivering personalisation based on data-driven algorithms without a coherent strategy for individuals’ privacy.

It says organisations must strike a balance between personalising offerings to make them more relevant to consumers, and maintaining brand reach.

For example, the report points out that while algorithms can predict consumers’ preferences based on their past behaviour or that of similar customers, they can also reinforce people’s existing behaviours and limit their ability to discover unexpected information.

Report author Celina Burnett (pictured), marketing analytics lead, ASOS and member of the #IPASocialWorks Group said: “The insights we draw from data – and the uses we can put this insight to – rely on the availability, integrity, and completeness of the underlying dataset, as well as the way in which we treat and analyse it. Marketers should acknowledge any limitations to ensure the data informs their decisions, but does not dictate them.”

The #IPASocialWorks report was written by Burnett and Colin Strong, head of behavioural science at Ipsos. It showcases best practice principles for using social media and social media data to create more personalised consumer experiences and includes case studies from Coca-Cola, Spotify, BBC and O2.