Skip to Content

Respectful personalization turns engagement into delight

Padmashree Shagrithaya
27 July 2022

AI enables enterprises to interact with customers one-on-one, at scale

Imagine discovering a boutique clothing store that completely understands you. The staff knows what you have in your closet, what you are missing, and what you’re willing to spend. They understand your style – the colors, patterns, and cuts you like but also what looks good on you and what you’re comfortable wearing. They go beyond selling you a shirt or a jacket to curating your wardrobe, making recommendations from socks to suits that they know complement the clothing you already have and that you will find delightful. And they understand how to communicate with you about those recommendations on your preferred channel– in a way that ensures you are actually excited to receive that email, text message, phone call, or invitation to a trunk sale.

Now imagine there was a way to deliver this level of service and sales recommendations at scale – enabling a multinational brand to provide that boutique level of personalized experience to its customers.

It leverages data and AI, ML, and other advanced data-analytics technologies to engage with each customer 1:1 – with the right content, at the right time, on the right channel, and at the right frequency – in order to build satisfaction and loyalty. It does this while complying with all data-privacy laws and other regulatory requirements, and in a manner that doesn’t feel invasive, unsettling, or untrustworthy to the customer.

That’s what respectful personalization is all about.
In helping Capgemini’s clients leverage data and AI (advanced analytics) to improve how they manage & engage with their customers; I’ve identified some common attributes that characterize the most successful deployments.

While AI poses immense opportunities, what is important is to ensure that our focus is not just on the “AI Algorithm” but on the entire implementation ecosystem, such as the architecture, technology interfaces, change management and the like. Policy implementations for key areas like privacy and security at all levels – Technology, Data & Algorithms, must be well established.  

Data sourcing strategy is key to “Respectful Personalization”. The following elements need to be carefully considered while coming up with a strategy – Which ecosystems are tapped into? How were they sourced? Does the data have the explicit approval of the customers regarding the purpose for which it is being used? What is the law of the land? What are the desired outcomes?
Methodologies adopted for algorithm building may need special attention. Many successful deployments occur when people define goals such as – increasing loyalty in a certain environmentally friendly segment, or boost viewership or minimizing inventory. Once this is defined, allow AI to optimize accordingly. This comes more naturally to organizations that have fostered a culture of experimentation – one in which the enterprise tests engagements with customers, collects explicit and implicit feedback, learns from the experience, and modifies its strategies accordingly.

Equally, once AI makes recommendations, it’s important that teams share them across the organization. For example, if a company’s marketing team learns its customers are more concerned about sustainability, there are implications for the product design team – but also for the supply chain and sourcing teams. Insights must be embraced, enterprise-wide – across the value chain, to ensure they’re acted on effectively.

Successful implementations also recognize that context is key. Customers demand different things at different times of the day or at different stages of their lives. Their preferences may even change depending on the device they’re using. For example:
A person visiting a website via a laptop may be open to exploration.
If that same person connects via an app on a phone, they’re likely more interested in quickly completing a transaction.

If they’re using a company computer, they may wish to receive communications about work-related products and services but not personal products and services.

Respectful personalization at scale has become a crucial component of any customer-engagement strategy – so much so that my team and I ensure its front and center when we work with our clients to deploy the Capgemini Data-Driven Customer Experience solution. If you have questions or comments about this, I would be delighted to hear from you.

Successful implementations also recognize that context is key. Customers demand different things at different times of the day or at different stages of their lives. Their preferences may even change depending on the device they’re using. For example:

  • A person visiting a website via a laptop may be open to exploration.
  • If that same person connects via an app on a phone, they’re likely more interested in quickly completing a transaction.
  • If they’re using a company computer, they may wish to receive communications about work-related products and services but not personal products and services.

Respectful personalization at scale has become a crucial component of any customer-engagement strategy – so much so that my team and I ensure its front and center when we work with our clients to deploy the Capgemini Data-Driven Customer Experience solution. If you have questions or comments about this, I would be delighted to hear from you.

Author

Padmashree Shagrithaya

Head of AI, Analytics and Data Science
“Managing multiple machine learning models, built by varied teams is a huge challenge. MLOps is a powerful approach to bring all the pieces together and reap larger organization-wide value from AI at scale projects.”