When it comes to personalization in email the time for excuses is over

The email marketing landscape is constantly shifting, changing and adapting to new developments and challenges. Mobile is firmly cemented as the most popular reading environment, and has permanently changed the way consumers think, engage and interact with content. The ever-connected individual creates an ever-growing digital footprint where every email and site interaction is tracked, measured and analysed. Technology has had to evolve at a phenomenal rate to keep up. Hundreds of businesses have been started in the past decade to assist marketers in delivering the best experience to their customers. However, personalization, in its truest form of individualized experiences continues to be somewhat elusive, particularly in email.


Personalization is repeatedly cited as a key factor in driving open rates, click-through rates, engagement and ultimately sales and revenue. In every prediction blog for the last decade it’s been right there at the top, and that’s still the case. The latest figures still show personalization as one of the most popular responses from marketers when asked what the most exciting opportunity for their organisation is. Both as marketers, and as individuals not wearing our ‘marketer hats’, we know that a personalized experience is more enjoyable than a generic one. Yet despite all the excitement and hype there’s still an embarrassingly low level of product personalization reaching consumers in their inbox.


I believe part of the problem is the confusion around the difference between segmentation, dynamic content and personalization. As a solid approach to email, segmentation is still very much alive and kicking, and for some brands they see that as personalization. They identify a user as female, so they ‘personalize’ her email to show womenswear products. Or, they have five product categories to promote so build dynamic content to show shoes to people that said they like shoes when they signed up and so on. Whilst this is step forward, it really can’t be considered personalization. It’s dynamic content and segmentation. Personalization is giving each individual product content that’s just for them. Not for them and the 100,000 other people who showed the same interest.


Why are so few brands really delivering personalized content in email? The main reason is data, data, data. There’re tonnes of it (in the ecommerce platform, email platform, Google Analytics, Adobe – the list goes on), but there’s a serious lack of access to it. It can’t be unlocked, it’s not in the right format or it’s not updated in real-time. Batch processing and delays between systems is a real pain in the ass. Simply put, the inability to access marketing ready data I think is the single greatest barrier to true personalization and the reason why the majority of brands simply aren’t delivering. Data is spread across multiple systems (those listed above) which often don’t interact. Better communication between e-commerce, data management and marketing teams can iron out the creases but it’s generally a massive internal project that can take years and many, many people to see it through. Marketers need an intelligent, easy way to deliver personalization at scale.


Internal challenges, siloed teams and the vast quantity of data means that the majority of email marketers are playing catch-up (and have been for years). Reliance is firmly placed on the data captured at sign up, click data or transactional data e.g. last category purchased so there can be lags in experience. On a personal note, on the rare occasion I have seen ‘frequently bought together’ or ‘things you might like’ in email it’s obvious they lack depth and authenticity – especially when the person sitting next to me received the same ‘recommendations’. Another very distinct issue is that processing lag; I’ve received emails recommending products to me that I’ve already purchased. There’s a communication breakdown, or lag, between purchase data being fed into most product recommendations. Traditional recommenders perform statistical computation offline, which causes delays. To actually deliver powerful recommendations, marketers need to rely on real-time data modelling.


So, what’s the answer?


I believe the answer lies in the power of AI. Specifically, leveraging data in real-time in conjunction with AI, to serve product recommendations in email, on site, or anywhere. There are two important elements here - live and AI. AI can be used to predict the best products for a customer based on their individual behaviour. The recommendations can be served up wherever needed. Let’s stick to email as the example – the recommendation can be served into an email at the moment it’s opened, meaning the recommendation is 100% live (no lags, no delays and no products the customer already purchased). Taking real-time data from site means that purchases, viewed, add to basket activity as well as site-wide behaviour and patterns can all be computed instantaneously to serve the live product recommendation. The real-time generation is crucial to improve the quality and effectiveness of the recommendation; the AI system can include more variables including stock availability, latest price changes and offers, the time of the day and users’ approximate location as well as weather conditions - this enables recommendations to be tailored even more precisely.


Truly personalized product recommendations have many benefits. They bring value to the customer, help them to find what they want easily, thus reducing friction and increasing satisfaction. For email, product personalization increases open and click-through rates, along with the average transaction value for purchases made following an email campaign. On that note, if consumers are receiving the right content, at the right time, and spending more as a result, then marketers really won’t need to be sending as many emails as they are right now. Instead of a batch-and-blast approach, it will finally be a case of quality content over quantity.


Fundamentally, we know consumers value a more personalized experience. Kickdynamic’s latest development is an AI prediction engine called LYNX. To empower email marketers to successfully implement real-time personalization, we have released a product recommendation module for email, powered by our revolutionary AI prediction engine. A one click, AI powered live recommender for the email channel and beyond. We’ve been working hard to connect the dots for personalization in email for a long time. So, what is everyone waiting for? A fresh start beckons: the time for excuses is over. Finally, the technology exists to enable brands to do personalization properly.

Back to our blog

Let's connect:

Follow us on Linked In   Follow us on Instagram   Follow us on Twitter

Recent Posts