Segmentation or 1:1 personalisation: which and when?

As retail marketers work towards creating tailored experiences for their customers, quite often there is a little confusion around the best way to truly create a memorable customer experience – when do you segment to serve a specific customer group? Is it better to personalise based on the individual customer? 

With customer expectations continuing to shift and rise to meet the levels of brands like Amazon, Netflix and Deliveroo – where do retailers start and how do they decide what to do?

This blog post aims to discuss where and when is the best time to use either personalisation or segmentation. 

To read how to get started with personalisation and scale it to your brands needs – read our post here. For more on how to personalise in line with your brand’s need, check out our latest download “One size fits none: the retail marketers’s guide to newsletter personalisation” and discover how to tailor your broadcast emails in a time of batch-and-blast. 

What is segmentation in ecommerce marketing?

Segmentation refers to the process of using aggregated data to split your target market into groups according to their shared attributes, and targeting them accordingly. (It is is achieved by analysing all the data a brand has within its database and filtering the messages they send based on conditions.)

Whilst it is a form of personalisation, it treats subscribers as segments rather than individuals and therefore may not connect with them on a 1:1 level.

Basic demographic segmentation

Customers and prospects can be segmented according to basic information they provided during the sign-up process, such as:

• Age
• Region/country
• Gender
• Geographic location (this is now mandatory and this is the basis of segmentation, without this you can’t work to take you personalisation to a hyper-individualised level)
• Whether they’re a student or not

In practice, this type of segmentation might take the form of:

• Sending male/female-targeted versions of your newsletter
• Sending customers an invite to an in-store event based on their location
• Targeting students only with an email or ad about a student discount offer
• Sending country/region-specific versions of your emails

Advanced campaign segmentation

As well as segmenting based on basic demographic traits, one of the most powerful and effective ways of segmenting your marketing messages also incorporates behavioural and transactional data into the mix.

This type of segmentation is most commonly applied to triggered or automated campaigns, and may be based on factors such as:

• Customer lifetime value – for example, creating a VIP version of automated campaigns (such as cart abandonment emails) that offer special perks for your best customers.
• Product/category purchased or browsed – for example, sending category-specific post-purchase emails containing content relevant to the purchase.
• Recency – for example, creating versions of triggered emails aimed at reactivating lapsed customers (those who haven’t shopped in a while), perhaps with a special offer to tempt them back.
• Lifecycle stage – for example, segmenting your welcome campaign based on whether someone has made a first purchase or not.

As well as helping your brand show it cares about its customer base, segmentation can also help avoid embarrassing slip-ups, such as sending a new menswear collection to a female subscriber, or a 40% off all items discount to someone who’s just purchased an item full price.

Real life example: Sigma Sports 

Premium cycling retailer Sigma sports wanted to personalise its broadcast newsletters and create personalised experiences across the two gender categories. Using dynamic content, which displayed imaged based on the gender the recipient shopped for,  Sigma sport was able to engage its female mountain bikers and showcase its products specifically tailored to them. Using segmentation, the brand was able to craft a personalised experience for select groups of its shoppers without any additional workload for the marketing team; including offers to activate members of the segmented groups and encourage them to engage with other automated campaigns.

1:1 personalisation in ecommerce marketing

Instead of clustering customers according to their shared attributes, 1:1 personalisation takes a look at each customer individually – using data to provide deeper insights. Here’s an example of what a hyper-personalised newsletter (below showing two options that may be sent to a shopper based on their last purchase or browsing activity) might look like:

Like segmentation, 1:1 personalisation is based on each customer’s unique profile – demographic data, browsing activity, purchases etc. However, instead of grouping customer by these attributes, this type of personalisation uses algorithms to create dynamic content blocks that change according to each recipient.

In practice, this kind of dynamic personalisation might take the form of:

• Category-specific content based on the recipient’s browsing and purchase history
• Dynamic offer banners that change depending on customer profile
• Time-limited content blocks that change after an offer or promotion has finished
• Weather-based content that adapts to the weather conditions in the place the email is opened
• Personalised product recommendations (for example, the products the recipient has viewed or purchased.

So, which one is better?

As always, the answer to this question isn’t clear-cut, and the reality is that ecommerce marketers should be incorporating both into their email marketing strategies and beyond.

Segmentation undoubtedly has its limits – having to create multiple versions of your twice-weekly newsletter for more than, say, basic gender or geographical segmentation would be incredibly resource-intensive and probably wouldn’t yield the kinds of results that would make it worthwhile.

Nonetheless, segmenting automated campaigns based on factors such as lifecycle stage, spend or browsing activity is incredibly important in making these emails more relevant to the recipient.

Is segmentation enough on its own? A few of years ago, we might have said yes. But considering the fact that 75% of consumers feel misunderstood by the brands they shop with, the answer today is likely no.

Personalisation can be applied to any campaign – whether segmented or not – to make it hyper-relevant to the recipient. It’s a step beyond grouping customers and prospects together and serving them content in the hope it might be relevant; combining each customer’s unique footprint with data science to present them with content that is highly likely to be of interest.

In a nutshell: whilst 1:1 personalisation is the future, segmentation is still key. Ultimately, the two are symbiotic; on its own, segmentation may not be enough to engage your customers and subscribers. And yet, without it 1:1 personalisation would not be possible.

To see how Ometria Labs is creating pilots powered by both of these campaign elements, read our case studies here.