consumers are hungry for personalized experiences — and brands are eager to
deliver. According to Accenture, 91% of consumers are more likely to shop with
companies that provide relevant recommendations, while 89% of digital
businesses are actively investing in personalization, according to Forrester.
clear that personalization is a top priority for digital professionals. But are
you tackling this goal in the most effective way? Personalization isn’t what it
used to be. Now it’s about using your first-party behavioral data to build your
own personalized recommendations.
recently hosted a webinar to share how leading consumer electronics and mobile
electronics retailer Dixons Carphone used Syntasa to implement data-driven
personalized recommendations — increasing conversions 3x in 3 months.
Watch the full webinar here, or read on for the highlights.
two primary goals were to a) improve attach rates for online purchases and b)
increase add-to-basket rates using data-driven product recommendations. Previously,
Dixons had used manual product recommendations, based on suggested product
bundles created by members of their online merchandising team.
using Syntasa, Dixons could now provide algorithmically personalized
recommendations that took into account recent related purchases for all
consumers, as well as individual behavioral data (e.g., context, brand, and
price affinity). The results included:
- 3x increase in add-to-basket rates
- +50% growth in bundle coverage (the share of
product views on the website where a recommended bundle was displayed)
- +5% units per order when any personalized bundle type
added to basket
How Syntasa Works
are four key capabilities any company needs to deliver algorithmic
personalization to customers at scale:
- Data: The ability to unify your
first-party behavioral data and enterprise data at the individual level is
crucial. This can be difficult since data typically lives in different
silos and can be massive in volume.
Building custom AI/ML models will help you achieve your specific
personalization goals. This is where combining your data really pays off
you — training your algorithms with your company’s data means they’ll
perform better than off-the-shelf black box models from personalization
need the ability to activate these recommendations across all your
data (which is key to personalization) is difficult to process with legacy
enterprise technologies. The massive volumes and complexity of this type
of data requires the use of cloud services. And the more customer data you
include, the more important it becomes to keep that data private and
secure – your best bet is to use your own virtual private cloud.
these capabilities have proved prohibitively costly and time-consuming for most
companies. Syntasa’s self-service web application enables digital teams to
implement cost-effective (and just plain effective) personalization strategies
in less time.
just three months, Dixons used Syntasa to build a data pipeline to ingest
behavioral data from Adobe Analytics, combine with other customer data, and
stitch together this data using our ID graph. Dixons used Syntasa to create
AI/ML recommendations and activated these across their channels to provide
customers with algorithmically personalized product recommendations.
For more on how Syntasa can help your team, watch the full webinar here.