Great marketing needs great data, especially when trying to optimise personalisation.
There’s nothing that sinks brand positivity and customer loyalty faster than using inaccurate data to incorrectly personalise marketing communications. No one likes being addressed by the wrong name or being sent irrelevant product recommendations!
In the past, we’ve recommended Google Optimize as a tool to conduct experiments in personalisation using zero and first party data. It was a simple and easy way to get started using data to test and optimise personalisation in several different ways.
But with the sunset of Google Optimize in 2023, the time has come to revisit personalisation strategy and how to optimise it.
What is marketing personalisation?
Before we get into how to optimise personalisation, let’s dive into exactly what we mean when we talk about personalisation from a marketing perspective.
It’s very simply the process of creating marketing communications and customer experiences tailored for a specific individual using their data.
The aim is to appear as though you, the brand, are speaking on a one-to-one basis to the customer.
Personalisation can be applied to almost any channel. If you’ve got the data and the mechanism to identify the individual (such as an email address or a tracked URL), you can use it to deliver personalised marketing.
In its simplest and most common form, personalisation is applied in email marketing to address the recipient by their first name and include content that’s consistent with their preferences. But it can be applied to entire websites, individual pages, ads, content and any other marketing collateral.
The goal is that customers feel seen, understood and valued. When implemented well, personalisation goes a long way towards positively impacting customer retention, lifetime value (LTV) and word of mouth.
How to optimise personalisation
Marketing is either personalised or it’s not. But how far you get into personalisation exists on a scale, and it’s for every brand to decide how far to take it in line with what’s best for their customers.
Mass marketing is at the unpersonalised end of the scale, with marketing on a 1-2-1 basis the hyper-personalised approach.
Great personalisation relies on great data, so first things first…
- Clean up your data
It goes without saying that accurate data is absolutely essential for accurate personalisation. Presenting cat-centric content to a dog-loving audience on a pet food site isn’t going to improve how valued those customers feel and in fact, it will probably have the opposite effect.
Don’t forget, we’re talking about opted-in data throughout this blog, you need clear consent from the customer to use their data. So if you have records of unclear or historic origin that there isn’t a clear opt-in against, you need to do some housekeeping to establish permission (or not) to use that data.
- Segment your audience
You will find there are groups of customers who share similar attributes – such as product preferences, geographic location and purchasing habits. Creating segments of these customers allows you to send marketing communications that match their attributes.
In Distil, we also group customers by collections of attributes, which we call ‘Customer Traits”. It means for a personal care brand we can separate the luxury skincare shoppers from the essential body wash buyers. For our materials merchant, we can separate the DIYers from the tradespeople, and for our meal prep brand, we can group the gluten-free dietary requirements from the low-fodmap or high protein needs.
- Individual attributes
Obviously the most important field is the customer’s name. That’s why campaigns featuring individual names are so successful. Remember ‘Share a Coke with…’ and the thousands of unique names from all around the world printed in each region so almost everyone could get a Coke with their name on it?
The power of the name is why almost every email marketing campaign features the *firstname* field.
But hyper-personalisation can go much further. You can get into attributes such as:
- The date a customer’s subscription is due for renewal
- The status of their loyalty points earned and when they need to spend them
- The limited edition product release that they’re invited to because they’ve bought X, Y and Z products and been a customer for *insert number of years*
When you’ve got good data, the power of hyper-personalization is limitless. But remember, with great power comes great responsibility…
When personalisation misses the mark
Have you ever received a communication from a brand that just felt a bit…well, creepy? This happens when personalisation goes too far and it can actually alienate customers.
According to this report produced in conjunction with Econsultancy, 67% of consumers say ads based on location are creepy. And it’s the same story with ads using cookie-based browsing activity and conversations around a smart device – all creepy.
This tweet by Netflix illustrates exactly this. Although not directed at an individual, the comments from some followers showed that using data in this way was a step too far!
To the 53 people who've watched A Christmas Prince every day for the past 18 days: Who hurt you?— Netflix (@netflix) December 11, 2017
Remembering Google Optimize
If you want to see our previous content using Google Optimize to create personalisation experiments, you can watch part 1 here and part 2 below.
While the Google Optimize platform itself may have been sunset, there’s still some useful nuggets in these videos.
In Google Optimize Part 1 we looked how we can use A/B Testing to create an experiment that compared two new emotive Titles for the website were you and I work “Gerry’s Vets Online Hub Place”.
One of the headlines was chosen by you. And one was chosen by me. That A/B Test is still running, and we’re looking forward to seeing who’s idea wins. But now, we’ve been called into see the bosses. Some things have transpired in the competitive landscape … and we’ve been told we need to step up our game.
It’s time to use “data” and unleash some real Personalisation!
For the first video in the series, check out Google Optimize, Part 1 – A/B Testing walk-through – you can see it here