• Personalisation

Automated Personalisation: Pomegranate in Action

Last month I announced the launch of our new product – Pomegranate. Now it’s time for a more in-depth look at how the platform can be used in practice.

At DataSine, our goal is to personalise every interaction between a business and customer. While there is some great research being done in the field of personalisation, intrepid marketers may find it difficult to capitalise. After all, how can you effectively cater for personality without hiring a team of Psychology PhDs into your marketing team?

With Pomegranate, you can now access this level of expertise through an AI-powered online platform. Our machine learning models understand how different content – in multiple languages – appeals to different personalities. Based on these models, our interactive tool provides in-line suggestions to maximise appeal, without demanding years of studying individual differences, and without needing to test every possible combination of content to find the most effective. With Pomegranate, the entire customisation process is automated.

Pomegranate in action

Pomegranate in action, suggesting words with a proven appeal increase for Extraverts

A simple example

Let us suppose that you work in the marketing department of a small chain of boutique confectioners. You and your team have a good understanding of the main customer segments that are likely to come into the store:

  1. People with refined taste in chocolate looking for novel experiences;
  2. People who are looking for a last minute gift for a partner;
  3. Those looking for treats to share with friends over a cup of tea.

Using Pomegranate, you fill in a short survey for each of your segments. What publications they read, what their hobbies might be, how tech-savvy they are and so on. DataSine identifies three corresponding personality segments:

  1. High openness – people who are open-minded
  2. Low conscientiousness – people who are impulsive
  3. High agreeableness – people who are agreeable

Next time you plan to send an email out to your customers, you first upload it to Pomegranate. It creates three copies of your email, each with different suggestions for improvements which will maximise the appeal to each segment, across a variety of content types:

  • Words and phrases to replace, to match the language that each personality uses;
  • Themes and concept appeal for each sentence, making it easier to emphasise the right things;
  • Colours, fonts and even layout to appeal to individual design tastes;
  • Images, analysing your uploaded images and suggesting royalty-free images from Unsplash with stronger appeal to each personality.

And because machine learning powers every suggestion, it learns over time.

Pomegranate in action

Gorgeous images suggestions from Unsplash, that will increase appeal to Introverts

By using these tailored emails, matched to your customers personality, you can drive engagement rates to much higher levels than a one-size-fits-all approach, and see a genuine change in customer behaviour and loyalty.

The future

This month, we are launching a small feature set as a closed beta, limited to English and selected personality segments. But our roadmap is packed with exciting new features that we will be rolling out over the coming months. The best way to find out is to sign up for our newsletter here.