Retailers already have access to all the customer data they need to attract — and retain and build — loyal customers. It’s just a matter of knowing how to use it.
Pro tip for telling the difference between retail marketers and dairy farmers: say the word “churn” and see how they react. In retail, churn is a dirty word; also known as “attrition,” it’s defined as the rate at which customers do not return for repeat purchases. Ever again. Sayonara, suckers.
Whatever a customer’s reasons for leaving your brand behind, the longer they’ve been gone, the lower the odds of ever winning them back. That’s why a lot of companies take measures to reel customers back in before they drift too far away — most commonly, these take the form of “We miss you” emails accompanied by a discount offer and coupons, which appear in a customer’s inbox after 90 days without a purchase.
But do they actually work? Short answer: no. Long answer: let’s imagine that customer A makes a purchase from you every single week for years, while customer B always buys around the holidays. By the time your 90-day email reaches customer A, they’ve skipped their regular purchase a dozen times — you’re way too late. But when you send customer B, who was all but guaranteed to make another purchase come holiday season, that “please come back” email, you’ve just offered up 10 or 25% off for no good reason.
Clearly, standard rule-based customer data strategies aren’t cutting it. But when your analytics teams are so bogged down with requests that their marketing teams have to wait three months or longer to get their questions answered, the type of personalized marketing that’s necessary to effectively retain a diverse range of customers is nearly impossible to achieve.
That’s why machine learning technology is proving to be indispensable to so many marketing teams. Machine learning models are tracking individual patterns to pick up on anomalies in behavior, indicating that the customer relationship is actually in danger, rather than relying on a one-size-fits-all approach to churn.
But while preventing churn is a huge step in the right direction for your brand (unless you’re selling butter or ice cream, in which case, awk), it’s actually just one of many ways that accurate, timely customer data can be implemented and turned into profit before Q4. And by many, we mean 11.
Achieve Better ROI:
Identify High-Value Customers
The top 5% of a company’s customer base can account for 30-40% of its total revenue, meaning that you need to know who these customers are — and make sure they stick around. It’s similar to the concept of incrementality that is being lauded as a new way to measure impact with regard to ad spends by marketers. Take a cue from leading men’s apparel brand Bonobos, who used predictive CLV analysis to identify its highest-value customers.
With that list in hand, Bonobos was able to implement low-cost but high-impact practices like including handwritten thank-you notes in these top customers’ shipments, ultimately increasing the predicted lifetime value of its customers by 20%.
Reduce Promotions
Many retailers feel so confident that a 15% discount will outperform a standard email that they’ll blast out a promotion to a 10,000-customer email list without testing a control group first.
While it might seem intuitive that including a discount will increase sales, it’s also possible that an email alone will be enough motivation for some customers to make another purchase. A lot of customers are highly motivated by promotions and coupons — but many are happy to pay full price, as long as they feel loyal to your brand. So, rather than offering a discount to a whole swath of customers, start small and test the outcome against a control group.
Increase ROI on Marketing Campaigns
For a lot of customers, direct mail can be an effective means of increasing repeat purchases. Tiffany & Co., for example, saw a significant increase in revenue with direct mail — but only after carefully targeting their mailing base. Obviously, you don’t want to send a bunch of flyers around to customers who are just going to (hopefully) recycle them immediately, but preemptively identifying which customers are likely to respond to snail mail can make it one of the most effective marketing tools in your arsenal.
98% of marketers agree that personalization helps advance customer relationships, however only 12% of marketers are satisfied in the level of personalization in their marketing efforts*
Improve Loyalty and Make Your Brand Relevant:
Personalization — to an Extent
In an ideal world, brands would have the marketing budgets and creative capacity to treat each of their customers like special flowers, with just the right water-to-plant food ratio for each. Unfortunately, that’s a little difficult to achieve in real life. With the right data, though, you can fake it until you make it.
98% of marketers agree that personalization helps advance customer relationships, however only 12% of marketers are satisfied in the level of personalization in their marketing efforts* — which means that with a little extra effort, your company can stand out. Breaking your customer base up into specific segments, understanding where they are in their journey from awareness to repeat buyer, and keeping easily-altered parts of your communication open to individualization are all ways to drive sustainable growth.
Predicted Product Affinities
Let’s go back to what we said about “keeping easily-altered parts of your communication open to individualization.” Predicted product affinities present one opportunity to do just that. If a customer makes a one-time purchase of a pair of yellow-and-red striped capris, it’s within your customer data wheelhouse to make the discovery that other customers who buy that product often also tend to go for your red foam nose, curly rainbow wig, and giant shiny shoes. Slide in those DMs — and by DMs, we mean your customers’ email inbox or Instagram feed — with suggested products that they’ll love, backed up by data.
Speak to Customers at Various Stages of the Funnel
Think of customers’ relationships with your brand as a funnel. The widest part of the funnel represents the largest group of consumers: those who are aware of your brand but haven’t interacted with it directly. As the funnel narrows, we get to the interested group; these people have perused your website but haven’t yet made a purchase.
Finally, the bottom point of the funnel is populated by your actual customers, whom you want to keep coming back for more. Most advertising channels let you utilize customer data for advanced targeting — take advantage of this feature to serve relevant ads to each group of consumers.
Increase Your Efficiency
Build Customer Segments in Seconds
The ability to discover and build customer segments is foundational to any marketing strategy — it’s how you predict whether an email promoting your new running shoes, your new ball gowns, or your new clearance item is most likely to attract the attention of a given customer. The more specific the segment, the more powerful your communication to that group will be.
To build these segments, you must have a sophisticated pattern-finding tool. Enter machine learning, which is all about picking up on patterns (a million times faster than a human could). Pair this advanced technology with your customer data, and you’re well on your way to segmentation that can cut through the noise and deliver results for your brand.
Measure Multi-Channel Marketing Holistically
Marketing teams think of Snapchat, Instagram, Facebook, and email as massively different beasts, but guess what? Your customers don’t. These platforms are all available to them on their phones as they go about their day, and the messages they receive from you on each contribute to their overall impression of your brand. Meanwhile, however, your brand is struggling to integrate the data that you glean from all of these different channels into one well-rounded portrait of a single consumer.
Customer data can be used to develop unified customer profiles and automate your campaigns to deliver consistent, highly tailored messaging to each.
Achieve Consistency
Remember your mother’s advice: “You’ll catch more flies with honey, but you won’t catch any flies if you mix honey and vinegar together. Sure, vinegar might be great in salad dressing, and honey might be great in tea, but salad and tea aren’t the same thing, kid.” Oh, just our moms?
What we’re trying to say is that your customer data can be used to develop unified customer profiles and automate your campaigns to deliver consistent, highly tailored messaging to each. This can help keep you from dousing your salad with honey, by which we mean accidentally advertising running shoes to your ball gown customers.
And Did We Mention…
Ease of Use
One of the biggest pain points that keeps marketers from effectively putting their customer data to use is the lag time between submitting a ticket with the analytics team and actually receiving the requested information. By the time the analytics team gets to your request for holiday shopping insights, you might be well into the new year and already looking towards spring clearance sales.
Marketer-friendly customer analytics platforms and quick testing mean that your analytics team can spend a lot less time crunching numbers and a lot more time helping the rest of your company figure out what those numbers actually mean — and how they can be translated into growth.
Integration into Marketing Tools
Close your eyes and imagine a world in which creating custom targeted ads is easy. Now open your eyes. Congratulations, your dreams are a reality.
One of the greatest pain points involved in creating Custom Audience ads and email campaigns is that they take two to three hours from start to finish, but direct integration with customer data to your email service provider (ESP), Facebook, and Google Ads means turning those hours into minutes. That’s huge — and that’s what’s possible with machine learning technology.
*Source: Researchscape; ‘Trends in Personalization Survey’, 2018
This article was featured on Target Marketing.