But that is happening surprisingly slowly. Anecdotally, the feeling seems to be that, while there are a few examples of great personalisation, very few companies are really nailing it.

Personalisation: What it is

If generalised is intended for everyone, personalised is tailored for me. At a very basic level, a generalised offer is something like “3 for the price of 2, while stocks last.” A personalised offer would be something like: “during your birthday month, get 20% off all cakes and candles”—the kind of reward many loyalty programs started off with decades ago.Of course, the degree of personalisation has evolved. For example, using historical data, my favourite supermarket could see that I’ve never bought a cake from them before, but I do regularly buy other confectionary, especially chocolates, and specifically one particular brand of dark chocolate. They could then reward my loyalty by giving me points to redeem on one of those slabs for free when I visit the store during my birthday week. Yes, please, and thank you very much!

Hyper-personalisation goes even further, incorporating multiple dimensions of data in real time - for example, geolocation. Let’s say I’ve received the aforementioned offer for a free slab of dark chocolate and head to the supermarket on my birthday. Picking up my live location, the supermarket could detect that I’m in the shop and send a message to my phone: “Great to see you in our store today. Please help yourself to a second free chocolate, on us. And happy birthday!”

Some may find this kind of hyper-personalisation intrusive, but with opt-in permission, it enhances the experience and strengthens loyalty.The degree of personalisation that’s possible has evolved.

Think of it as levels of personalisation, based on how much data is available and how much is known about a customer:

At Level 1: this customer likes sport.

At Level 2: this customer plays sport.

At Level 3: this customer plays golf.

At Level 4: this customer plays golf most Sundays.

At Level 5: this customer plays golf most Sundays at a certain club.

At Level 6: this customer is playing golf right now at this club.

Levels 1 to 3 are arguably more in the territory of segmentation than personalisation – an offer of discounted golf balls, for example, might go to a group of (hundreds or thousands of) people.But you can see how, at each progressive level, it’s possible to tailor rewards to make them more relevant, immediate and individualised.

Why personalisation isn’t ubiquitous

Even more basic forms of personalisation are still surprisingly rare. And poor attempts at personalisation are still common - such as an email from a supermarket with “exclusive deals for you” even though you haven’t shopped there in over a year. That’s not personalisation; that’s poor use of data and ineffective marketing.

So why aren’t we seeing more effective personalisation?

• It’s shiny but new: The allure is clear, but personalisation is still in its early days. Many businesses are still learning how to navigate data and use the tools available. AI adds exciting possibilities but also complexity.
• Not enough data: Personalisation requires data, and it takes time to build up sufficient quality data to be meaningful.
• Too much data: Having loads of data is a good thing but also presents its own set of challenges. Finding relevant data amid the noise requires AI, however many businesses are still figuring out how best to use AI to mine data. A daunting task for most.
• Too few experts: Data alone isn’t useful—it needs to be analysed and applied effectively. This requires strategists, marketers, and developers working together, which not all businesses are structured to support.
• Lack of prioritisation: Businesses juggle many priorities, and personalisation may not be one of them. Risk aversion, complacency, or a lack of familiarity with data-driven strategies can slow adoption.
• Time and budget constraints: Building and maintaining quality data takes resources. While data and AI are hot topics, budgets don’t sign themselves off.

How to start: Small, slow, but steady

Businesses don’t need to jump straight to hyper-personalisation. Even basic personalisation is better than none. The key is to use the data available effectively, personalise rewards, and build up from there. It’s a marathon, not a sprint.