Challenge

Luxury beauty brand Coty wanted us to come up with a recommendation engine for fragrances to target customers with perfumes that they were more likely to buy.

Solution

Our data science team used machine learning to formulate 18 simple, effective questions for Coty’s customers to build their individual preferences.

The data collected gave us the raw material to:

  • Cluster the population into 12 personas
  • Define six rating rules to quantify how much a customer likes a perfume
  • Design five recommendation engines working in symbiosis

We used machine learning to formulate 18 simple, effective questions for Coty’s customers.
Using an iPad app, the questionnaire was put to the test in Luton airport and Oxford Street for three days with real Coty customers.

Impact

82.3% of respondents indicated that their recommendations were relevant,
giving high scores of 4/5

The first trials generated a buzz.
Customer feedback was exceptionally positive:

I love it! | Flamboyant! | Reminds me of university, positive memory, my sister wears it! | Just perfect! Very smooth and pleasant. | Own it, love it!

More client stories

Connect with us

Fire us a message and we’ll be in touch

Contact us

‘Knowledge’ –
directly to your inbox

Want to make life easier by staying on top of market trends?

Fill your mind with insights.