BUSINESS CASE

An AI product in 2 weeks

Challenge

Buying the vintage lamp that will perfectly suit your newly-decorated living room is hard. You know that you want a 3-legged lamp, but you can't find the ideal item. For ManoMano customers, we built a ‘same style’ recommendation engine. Based on your search, the engine presents your tailored selection of ‘same style’ vintage-looking lamps.

ManoMano, E-commerce, Recommendation Engine, Artificial Intelligence, DIY, Sicara
ManoMano, e-commerce, construction, prediction software, AI, machine learning, DIY

ManoMano is the online reference for DIY and gardening in 5 European countries. It is among the 50 most visited e-commerce websites in France. With a revenue increase of 70% in 2018, ManoMano aims to become the French Amazon of DIY and the next French unicorn by pushing continuous web innovations for their customers.

424M

Revenues in €

101

Product Cat.

2013

Launch

Our impact

+95M€
+5 % added business value in 6 months
+1,7%
Conversion rate in 2 weeks
Background
Background
Quotes

Sicara’s support stands out for 3 reasons. First, the Sicara teams have seamlessly integrated into the ManoMano teams. Second, the Sicara teams adapt to all business matters and are proactive in offering innovative solutions. Third, the ManoMano teams took advantage of the agile dynamism brought forward by Sicara.

ManoMano, Hosanski

Emmanuel Hosanski. Lead Product Manager @ManoMano

What we did

We implemented a ‘same style' product recommendation engine based on visual similarity.

ManoMano, DIY, lamps, recommendation engine

What we did

We implemented a ‘same style' product recommendation engine based on visual similarity.

We launched this ROI-driven recommendation engine to answer customers’ needs identified in pre-production extensive user research. ManoMano visitors browse similar products based on their style. Offering a tailored selection of products based on customers’ style preferences increases the conversion rate and enhances customers’ experience.

How we did it

A pragmatic solution

ManoMano, lamps, Sicara

How we did it

A pragmatic solution

(Re)Qualifying existing products based on their style is time-consuming and very subjective. Thanks to our library of image recognition, we knew that an algorithm could recognize visual similarity (or same style) and push forward adapted product recommendations based on the visitor’s browsed article. To put in production the recommendation engine for 10 categories of product in minimal time, we put forward a pragmatic solution: an open-source and pre-trained convolutional neural network. Meet VGG16 deep convolutional neural network!

logo, learn, manomano
manomano, opencv, logo
logo, python, manomano, sicara
keras, logo, manomano, sicara

Our team set up

We worked in full integration with the ManoMano data scientist and feature teams.

startup, sicara, team, teamwork

Our team set up

We worked in full integration with the ManoMano data scientist and feature teams.

Our cross-functional teams at Sicara were designed so as to integrate seamlessly into the ManoMano in-house teams.

Olivier

Centrale Paris, Polytechnique

Tanguy

Polytechnique, PhD

Elisa

emlyon


Related articles written by Sicara data scientists

Edge detection, tutorial, knowledge

Edge Detection in Opencv 4.0, A 15 Minutes Tutorial

This tutorial will teach you, with examples, two OpenCV techniques in python to deal with edge detection.

TensorFlow, AI, Docker, GPU

Set up TensorFlow with Docker + GPU in Minutes

Why Docker is the best platform to use Tensorflow with a GPU.

blurry street

GAN with Keras: Application to Image Deblurring

A Generative Adversarial Networks tutorial applied to Image Deblurring with the Keras library.