Making Money On-line: How to Personalize Emails at Scale — While Sticking to Your Brand's Creative Rules

Monday 4 November 2019

How to Personalize Emails at Scale — While Sticking to Your Brand's Creative Rules

The more brands a retailer works with, the more complex sets of rules they have to navigate in the design of their marketing assets. This is particularly true for luxury retailers, who work with brands that have spent decades building up elite reputations — and maintain correspondingly strict rules about where, when and how their names can be used.

Say, for example, that a luxury retailer sells items by Gucci, Louis Vuitton, Prada, Burberry and Tory Burch. Imagine (hypothetically) that Prada won’t allow the retailer to show their products next to Louis Vuitton in any marketing email — while Gucci won’t allow their products to be featured next to any sale creative.

Rules like these make it impossible for a retailer to personalize an email for every one of the millions of subscribers on their list, while adhering to every brand’s rules across every variation. Just designing one single set of emails within these intricate constraints — let alone an entire email campaign — would consume hundreds of hours of manual work.

This is precisely where automated email campaign personalization demonstrates its value. Machine learning personalization software is not only smart enough to learn complex sets of pre-specified rules — it adheres to those rules even as it learns from real-world customer interactions, and uses that data to guide each subscriber on their own personalized journey through the product space. Here’s how.

Treat Each Individual Asset as Its Own Separate Piece of Creative

Email marketers at many retail brands have a tendency to think of entire emails as pre-designed creative assets. This line of thinking runs especially strongly at luxury retailers, whose strictly cultivated aesthetic standards date all the way back to the days of full-page advertisements in magazines and newspapers.

Once a retailer enters the realm of individualized email personalization, however, the inherent limitations of this approach quickly become obvious. Even if the retailer automates an entire campaign, their “pre-baked” creatives prevent their software from reaching each customer with the messaging and visuals most likely to generate a click-through.

At the same time, many retailers are reluctant to hand over full control of personalization to their automation software, for fear of inadvertently sending “Franken-emails” filled with nonsensical combinations of creative and messaging. As a result, their marketing teams may spend days or even weeks manually combing through emails, triple-checking that every creative and message is aligned with the brand’s voice and values.

Thankfully, a third approach solves both these challenges. In this “middle way,” retailers treat each asset — whether it’s a sale banner, a new arrivals message, or a lifestyle shot — as its own piece of creative. Their designers shape the creative’s visuals and messaging in alignment with the brand’s look and feel, then hand it over to the marketing team, who instruct their machine-learning software how the asset can be combined with others.

Label Each Creative Asset with Tags that Specify How and Where It Can Be Used in an Email

Just as pre-designed creative assets enable a brand to maintain full aesthetic control while achieving personalization at scale, guardrail rules ensure that every asset is used only in the right locations and combinations within each email.

One easy way of setting guardrail rules is to label each creative asset with tags. These tags may specify that a given asset can never appear in the same email as certain other creatives — for example, that ads for Gucci purses may be shown with Prada ads, but not with Louis Vuitton. Alternately, a tag may specify that certain creatives must always appear together — for example, that every email advertising summer swimwear must also contain ads for sandals.

Tags can also be used to specify how long each creative is allowed to remain active — preventing spring dresses from appearing in emails sent as autumn approaches, for example — as well as which subject lines and messaging can (and can’t) be used in emails containing each creative. In all these ways, guardrail rules neatly avoid the problem of the “Franken-email,” by setting specific parameters within which the software can generate email variations.

Use Machine Learning and Automation to Personalize Content Within Those Guardrail Rules

By setting guardrail rules around pre-designed creative assets, brands remain firmly in the driver’s seat of personalization. At the same time, this approach frees automation software to learn from each interaction with every customer, and responsively tailor that pre-vetted messaging in ways that create magical moments of engagement at scale.

Along the way, this approach vastly streamlines the personalization process, freeing designers to concentrate on developing innovative new assets, while freeing marketers to concentrate on the high-level strategies of their campaigns.

In all these ways, the combination of pre-vetted creatives, guardrail rules, and intelligent automation results in campaigns that delight customers — while precisely adhering to the boundaries specified by every brand in a retailer’s product catalog.

Part of personalizing emails is knowing if your audience prefers to read them on a mobile device. Find out “5 Pro Tips for Optimizing Email for Mobile.”

See the infographic