As Toys ‘R’ Us files for bankruptcy protection in the US and Canada, we’re again reminded of the brutal (and overcrowded) market retailers today operate in. The 69-year-old toyshop cited “unrelenting competition from ecommerce and big box retailers” as just one of the reasons for its insolvency.
But it’s not just veteran stores, born offline, that are finding it difficult to keep up with today’s ever-changing retail landscape; according to Fortune magazine, 90% of all online businesses fail (just look at Nasty Gal and Fab, for example).
So, what can today’s retailers do to stay on top of their game and pave the way for long-term prosperity?
There’s no simple answer to this question: from the quality of a brand’s product to the management style of a business, there are numerous factors that can determine whether or not a brand sinks or swims. What we can tell you, however, is that —in 2017— the way in which you communicate with your customers is extremely important.
Today we operate in a relationship economy, where a consumer’s spend is driven by their rapport with a brand; i.e. what that brand does for them, and how it makes them feel. Consumers today want to be recognised by their favourite brands as individuals—whether that’s in the form of personalised recommendations or exclusive perks for being such a great customer.
But how can a marketer go about building and maintaining such a relationship—all at scale?
Below we’ve listed the seven stages of smart ecommerce marketing: starting out with the basic newsletters of yesteryear before graduating to the cutting-edge, game-changing technology available today.
To learn more on this topic, check out Econsultancy’s range of ecommerce training courses.
Once upon a time, marketers approached customer communication through the concept of a one-size-fits-all campaign. In other words, the same message was sent to an entire contact base, with the content shaped by generic information such as: products trending at the time, products merchandisers wanted to push or simply the time of year.
Since the noughties, such campaigns have been sent out via email service providers (ESPs) in the form of a newsletter; an example could be a brand sending out a festive campaign in early December promoting its Christmas party dress collection.
But what happens if 20% of those who receive this campaign never wear dresses?
If we were answering this question twenty years ago, the answer would be “nothing”. Customers just didn’t expect as much from marketing messages back then and, thanks to fewer people sharing content online, their opinions weren’t as important to a brand’s reputation as they are now.
Oh how things have changed. Today, customer expectations are much higher, and if you have an unhappy customer that receives a highly irrelevant email, you’ll know about it. As will their entire Twitter following.
Email newsletter signup form
But the above scenario is outdated; it’s the original, old-school way that most marketers either have moved on from or are (hopefully) at least in the process of doing so.
The next step up from batch-and-blast newsletters comes segmentation; in other words, using a small amount of data to personalise emails by basic demographic information such as gender.
Segmentation was particularly popular ten years ago, when it was seen as an innovative way to boost conversion by ensuring at least some of the content being sent to subscribers was in some way relevant.
However, it soon became clear that – without automation – segmentation could be a tiresome process; by doubling a campaign, marketers were also doubling their workload—was it really worth it?
Luckily, around five years later technology stepped in to help in the form of marketing automation.
Marketers were amazed by the concept of a marketing platform that could automate some of the manual and often technical tasks that were eating away at so much of their time.
One of the first campaigns to take off was the welcome campaign, closely followed by basket abandonment (a campaign capturing people’s baskets and emailing it to them to encourage them to clinch the purchase).
Other automated campaigns soon followed suit, such as post-purchase, browse abandonment and at-risk.
But what did the process of sending automation campaigns look like at this time?
Retailers would typically try to send these campaigns from their email service provider; however, as these providers were typically created around fifteen years ago (for the sole purpose of sending one message to everyone), this would prove extremely difficult.
Consequently, marketers would purchase an additional marketing automation tool that sits alongside the ESP and would have to be connected to the database (because you can’t do a basket abandonment campaign without knowing what’s in a basket).
4) Dynamic content
Everything we’ve discussed so far in this list requires multiple campaign creation (e.g. the example of sending one campaign for men and another campaign for women, and so on).
But the introduction of “dynamic content” changes this—taking the whole process from basic segmentation to 1:1 personalisation.
To recap, dynamic content (in email) refers to content that automatically changes according to a recipient’s customer profile. Based on ‘if’ conditions within the template’s HTML (e.g. ‘if’ recipient has spent [x], the template will display [y]), dynamic content is all done within the user interface (UI)—therefore removing the need for any manual code editing. Common forms of dynamic content include dynamic offers and promotions, ‘hero’ header images and product recommendations.
Most marketers will go about executing this by using a very advanced ESP which can take care of the dynamic content part (not all of them can), but typically a product recommendation provider is also needed.
Once you’ve plugged in both of these things (in order to track the data), you’re left with a pretty sizable tech stack, making things a little complicated.
Dynamic email content
5) Unified cross-channel communication
Hitherto, we’ve only talked about email marketing, but — as most of you reading this will already know — this is far from the only channel consumers are using.
From social networks like Facebook, Instagram, Snapchat and Twitter to offline touchpoints like direct mail and in-store campaigns, today there are so many different ways a shopper can interact with your brand.
Consequently, an effective ecommerce marketing strategy will bring all of the different touchpoints together and look at them holistically in order to determine the best message to send on the best channel at the best time. So if a consumer spots a suede jacket they like on an ecommerce store, they can be retargeted with the same jacket on Instagram, and if they then purchase the jacket in-store a marketer can just stop the campaign and just send a ‘thank you for your purchase’ message on their favoured channel.
In order for this to work, there will be no such thing as siloed marketing teams (such as “the social team” or “the email marketing team”) but instead there will be just one team working together to share a consistent message and reach the same goal.
To pull this off, a retailer will need the tech stack mentioned in the previous point as well as then plugging into all of the social platforms you need. A never-ending tech stack, basically. But bear with us: it gets easier (ironically).
6) Taste profiling
So far everything we’ve talked about in this post has been centered around human based rules and logic, for example: “If a customer bought a waterproof coat, display [x] suggested outwear coats.”
However, this type of marketing (where a human being calls all the shots) is unlikely to be all that intuitive or accurate. Looking at the example above, why should a customer only want to view outerwear coats, just because they happened to buy one five years ago?
If a marketer wants to truly understand a customer’s unique tastes and needs, far more data is needed. For example: “What has this customer been browsing on-site recently?” “What have they bought offline over the past couple of years?” “Which items have they returned?” “Which items have they left in their basket more than once?” “Did they buy a product through an ad or an email?”… And so on.
But even once all of this customer information is available to a marketer, in order for them to actually use it a completely different level of technology is needed: one that can run a model that not only has the ability to unite and store vast amounts of data, but also draw on it to predict future behaviour.
Is this AI? It’s something that can forecast future behaviour through a computer algorithm, so yes. And this is the direction ecommerce marketing is going in.
7) AI Marketing
But it’s not just content that AI will be able to power: the delivery of a message, in a cross-channel world, will also be determined by a machine.
What does the concept of an AI-powered algorithm deciding which message to send to whom actually look like?
Take an abandonment cart campaign, for example. Now, your brand might already be sending this campaign, but how do you know whether to send a single email or a series? Or the number of days to wait after the cart has been abandoned before sending it out? Or the incentive to include, if any?
All of these questions are impossible to answer, for each and every customer, without the help of a machine. It’s a never-ending stream of (important) decisions that machine learning can take care of by rapidly processing vast sets of data (leaving marketers to focus on innovation and strategy).
So, how do you get hold of said machine?
In order to start using AI in a brand’s ecommerce marketing strategy, a marketer needs a whole new piece of technology that can sit as a layer over everything else and ensure that the right message is sent to the right message at the right time.
So, all this sounds great – right? But you’re probably thinking: “a nice to have”.
But is it really just a nice to have?
As mentioned at the start of this article, competition in the retail industry is already tough. There’s already a big fat “unsubscribe” box at the bottom of your emails, tempting unhappy recipients to click. And when GDPR comes in next May, it’s going to get 10 times harder: when someone says they don’t want your emails, you’re unlikely to be given a second chance.