Debenhams delivers an additional 1.3% uplift in online conversion on the 3.29% rise already witnessed in the proof of concept with social proof messaging

Debenhams ran an initial proof of concept trial of social proof messaging from Taggstar in the UK for ten days before it hit statistical significance, meaning that it had proved that social proof messaging had a positive impact on the business.

Debenhams then quickly moved to a machine learning-supported social proof messaging trial, which ran for 23 days. Taggstar’s machine learning solution uses an algorithm that selects social proof messages based on shoppers’ behaviors, continuously testing all message combinations and types for feedback to accommodate changes in message availability and conversion rate performance over time. This ensures the right message or combination of messages is delivered at the right time in the customer journey to optimize results.

Both the proof of concept and machine learning trials operated across Debenhams’ PDP and basket pages.

In the proof of concept, 12 message types were used, including ‘Trending Now! sold X times in the last X’’ and ‘Popular! X others are looking at this right now’.

In the machine learning experiment, this was extended to 13 message types, with the addition of a category bestsellers message ‘Number X best seller in {category}’.

The results were impressive from the start. In the proof-of-concept evaluation, Debenhams achieved a 3.29% conversion rate uplift and an ROI of 36.4X.

With the addition of machine learning, a further 1.3% uplift in conversion rate was achieved, delivering a 23.5X ROI.

Following both successes, Debenhams is looking to test social proof messaging further, including review messaging, which will allow it to share social proof based on how other customers have rated products they are looking at. Social proof messaging will also be extended across the full funnel with its incorporation on the PLP in addition to the PDP and Basket pages.