SPENCER’S RETAIL USES AI TO PERSONALISE THEIR

ONLINE STORE
India’s leading grocer increases conversion rates by 7% with Smartech’s
proprietary technology
“We use Smartech to power our recommendations across platforms. It
was extremely easy to integrate in Magento, Android as well as iOS. We
were a bit apprehensive about the impact of Smartech’s algorithms, as Manish Kapoor
most of their clients were in fashion, and grocery is a different beast CTO Spencer's (till Aug’18)
altogether. However, we were pleasantly surprised with the results when
we got a substantial increase in engagement and revenue in the 1st
month itself.”

Summary The Challenge

Spencer’s used Smartech to But serving right products Spencer's Retail is a chain of
Spencer’s, like almost all the
identify the most to right customers at the retail stores headquartered in
online grocery stores faces
personalised right time is a huge chal- Kolkata, West Bengal having
the ever increasing chal-
and relevant products for lenge as merchandisers can presence in over 35 cities in
lenge of converting visitors
each customer, and then only control their recom- India. Spencer's is based on the
to customers. One way of
sho w these in the “You m ay mendations overall and not 'Food First' Format. Many
increasing conversion is to
also like” widget on the on per customer basis. On outlets though sport multiple
judge what that particular
Product Detail Page and top of it, ability to predict formats for retailing food,
customer wants and guide
“You should buy” upsell customer behaviour in apparel, fashion, electronics,
him/her to the right set of
widget on the cart page advance is also not an easy lifestyle products, music and
products.
problem to solve. books.

Smartech Incremental purchase Incremental views RoI

7.34%
Personalisation
Results 7.20% 16X
 Smartech believes that every customer is different and should be treated

THE SOLUTION likewise. For 1:1 personalisation it uses clickstream data of all the
visitors along with the current context to predict next action of the
particular visitor.
 Implementation
Home Page
To s h o wca se a set of products which that
visitor is most likely to purchase, a separate
page titled 'Picked just for you' wa s created.
Traffic wa s diverted to this page via banners
on home page and links on product detail
pages. People who visited this particular
boutique converted 2 0 0 % more than others
exhibiting true power of personalisation.

Home Page N o of p r o d uc ts r e c o m me nd e d Buy-through Rate Conversion(Clickers v s B a U )

Recommendation
10 8.29% 3X
Recommendation widget
on Product List Page
A widget titled Recommended for you' wa s created for product list page and
wa s placed above the main product image. Here additional parameters like
the current product on which visitor is there wa s also taken into account
apart from giving more weightage to his recent behaviour People who
clicked on this widget converted 1 5 0 % more than BaU buying exactly the
same products which were recommended to them.

N o of p r o d ucts r e c o mm e n d e d Clickthrough Rate Conversion(Clickers v s B a U )

P r o d u c t Detail
P a g e widget
3 4.75% 2.5X
Recommendation A widget titled 'you may also like' was created for product detail page and

widget on Product
was placed just below the main product image. Here additional parameters
like the current product on which visitor is there was also taken into
account apart from giving more weightage to his recent behaviour People

DetailPage who clicked on this widget converted 150% more than BaU buying exactly
the same products which were recommended to them.

Product Display Page

N o of p r o d ucts r e c o mm e n d e d Clickthrough Rate Conversion(Clickers v s B a U )

P r o d u c t Detail
P a g e widget
10 3.19% 2.5X