Technological advances in retail happen quickly, so retail marketers have to be able to react just as quickly. In this piece for RetailTechNews, Abigail Davies (pictured below), content marketing manager, Ometria, describes the changes we have already gone through, and how machine learning could hold the key to future customer communication.
In 2018, communication is at the heart of retail success.
Take Boohoo.com, for example. The Gen Z brand is going from strength to strength. Why? Because it knows who its customers are and, more importantly, knows how and when to communicate with them.
Customers today prioritise the relationship they have with a brand. They expect to be recognised as individuals, whether that’s through spot-on product recommendations or unique offers created just for them. For context, 70% of 18-24 year-olds say brands that fail to personalise their marketing messages will lose them as customers.
However, when you’re dealing with thousands, or even millions, of customers at the same time, is this level of personalised marketing actually possible?
A couple of years ago, the answer would be a monosyllabic 'no' – luckily for you guys, things have changed since then.
Below, we take you on a journey through the evolution of smart retail marketing: starting with the best practices of yesteryear, before making our way to impressive tactics you can use today.
The starting point
Up until relatively recently, even the savviest retail marketers were approaching customer communication through a one-size-fits-all email campaign.
Leading brands would send the same message to an entire contact base via an email service provider (ESP) and there was absolutely nothing wrong with that: the generic newsletter was all tech permitted, and customers didn’t know any different.
Abigail Davies, Content Marketing Manager, Ometria
But then technology evolved – and with it so have customers’ expectations.
First there came segmentation, a tactic that involved using small amounts of data to personalise emails by basic demographic information (such as location, age, gender, etc.).
However, it soon became clear that segmentation could be a tiresome process; by doubling or tripling a campaign, marketers were also doubling or tripling their workload.
Luckily, marketing automation stepped in: an innovative technology (back then) that could save marketers time by sending campaigns on their behalf.
Since then, retailers have been trying to power automation campaigns from their ESP; however, as these providers are (typically) created for the purpose of sending one message to everyone, it can prove tricky.
Consequently, marketers often need to purchase an additional marketing automation tool that sits alongside the ESP and can be connected to the database.
The (dynamic content) milestone
Whilst these tactics have offered marketers something more sophisticated than 'batch and blast', they still require multiple campaign creation. The introduction of 'dynamic content' has made things a bit easier.
Dynamic content (in email) refers to content that automatically changes according to a recipient’s customer profile (therefore enabling a marketer to work from one single template).
Based on 'if' conditions within the template’s HTML (e.g. ‘if’ recipient has spent [x], the template will display [y]), it’s all done within the user interface – therefore removing the need for any manual code editing.
Common forms of dynamic content include bespoke offers and promotions, and 'hero' header images and product recommendations.
In order for a marketer to be able to use dynamic content, they need access to a single customer view (i.e. access to all of a customer’s data, brought together and stored in one central place).
If this unified view isn’t possible because customer data is siloed (perhaps due to a convoluted tech stack), it’s far more tricky for a marketer to be able to draw meaningful conclusions about each customer and power advanced personalisation.
Where we are now
Everything we’ve talked about in this post has been centred around human-based rules and logic. The problem is, this approach relies on limited resources, and therefore risks not always being accurate.
Enter: machine learning.
If a marketer wants to truly understand a customer’s unique tastes and needs, a vast amount of data is required – and in order to actually process this data, a completely different level of technology is needed.
Today, marketers need a product that can not only unite and store vast amounts of data, but also draw on it to predict future behaviour. It needs to be able to sit as a layer over everything else, and ensure that the right message is sent to the right person at the right time.
And we’re not just talking about email here. The sort of product we’re talking about will sit across everything, bringing together all of the different customer touch points (social, direct mail, mobile notifications) in order to look at them holistically and determine which channels work best and when.
This type of technology isn’t just a ‘nice to have’ – it’s a necessity.
This content was originally published in RetailTechNews.