Download This Whitepaper
Oops! Something went wrong while submitting the form.
Download the 2022 Email Trends and Observations Report
Oops! Something went wrong while submitting the form.
Retail marketers today have access to an unprecedented amount of data about their customers. Yet most are unable to harness even a fraction of that data to drive smarter marketing campaigns. Instead, they are relegated to using the same rules-based approaches to targeting and segmentation they have for years.
Predictive models by contrast, capable of effectively processing billions of data points, are perfect for identifying behavioral patterns at scale with the help of machine learning. These algorithms can identify which customer interactions lead to a higher future spend, predict the likelihood of a purchase and detect when a customer is about to churn.
Listrak’s Predictive Analytics Playbook outlines twenty ways you can easily put predictive analytics into practice and get smarter about acquisition, customer retention, discounting strategies and cross-channel tactics.Enjoy reading these plays and realize that they are only a starting point for you to evolve all your marketing campaigns in order to drive new levels of engagement, loyalty and revenue.
Most marketers consider their “active” email list to be those subscribers that have opened in the last 6 months(or some other arbitrary measure).
We’ve all been trained to heed this boundary or suffer the wrath of the deliverability gods…
And it’s true that suddenly sending large volumes past this boundary will be detrimental to your deliverability and your business.
But what if we could selectively go beyond the wall to a small group of “inactives” most likely to engage? (the deliverability gods have no problem sending to customers who want to engage).
Most marketers devise a one-size-fits-all approach for their win back campaigns using data “averages” (e.g. “not purchased in 61 days”).
But these averages don’t incorporate critical factors like individual purchase cadence, categories purchased, or seasonality into the campaign timing.
Predictive analytics can understand each customer’s individual buying patterns and determine their likelihood to churn.
By setting recurring weekly campaigns targeting “ready to churn”customers, retailers can increase their chances of engagement, revenue and long-term loyalty with these customers.
Even when we’re on the brink of losing a customer forever, there’s still a chance of bringing them back with the right reactivation strategy.
Again, in the past we’ve been forced to use static rules and variables to determine the timing for the reactivation program.This one-size-fits-all approach generally fits nobody due to individual shopping styles.
By using predictive models that analyze shopping patterns at a customer level, marketers can take a more data-driven approach to reactivation.
Predicted Lifecycle Stage allows you to suppress daily sends to churned customers, and instead send a monthly targeted reactivation campaign specifically to these customers.
Using lookalike audiences on paid search and social is key to targeting prospects that are similar to your customers.
But most retailers set their budget and keywords to drive as much qualified traffic as possible to their site in the hopes of converting them – with no regard to the value of those customers.
And we know that all customers are not equally valuable to us: would you rather attract one-time, deep-discount shoppers or high-value, repeat customers that buy at full-price?
Use Predicted Future Spend to build a lookalike audience of your top 10% of all potential spenders to drive a higher caliber customer to your site. This approach not only increases the longterm value of your customer base but can drastically improve the ROI of acquisition campaigns.
It may seem counterintuitive, but a customer is most likely to purchase nearly immediately after they complete a purchase. Remember that time you remembered something else you should have bought in the first place?
Retailers will often send a single thank you message a couple days after the order confirmation to remind them you care. One email says you care, but only a little bit. Let’s treat potential VIP customers with an additional (conversion optimized) thank you message.
Using the Predicted Future Spend model, we send additional messages to customers with a high likelihood to purchase at the exact moment they’re most engaged.
Abandonment campaigns can be a great place to leverage coupons to push shoppers to a purchase. Anti-gaming logic even prevents crafty shoppers from taking advantage of discounts.
But in the long run, margins suffer if you’re giving coupons to everybody that walks through your front door. It seems kind of silly to hand someone a coupon as they’re walking up to the cash register, right?
With the Discount Affinity predictive model, we can withhold coupons from full-price buyers, especially those with a high likelihood to purchase. This preserves margins to keep profits high without cutting into conversion rates.
Like young love, interest is high early on in the relationship. Eventually your subscribers are going to push you away and stop opening your emails. This lands them on the inactive list, likely to never return (some would say I have some emotional baggage).
Once a contact becomes disengaged in email, the natural response is to stop sending to them. But we don’t have to give up so easily, there are other ways to get their attention.
A smarter approach to disengaged email contacts is building a custom audience of only those Most Likely to Buy. Once there, the ads on Facebook and Instagram will have a much higher conversion rate, saving money and driving revenue thought to be lost from the email channel.
Nobody escapes email list attrition; contacts can and will unsubscribe without warning. Once a contact unsubscribes from your email list, they’re probably gone forever.
Anticipating the moves of your contacts like a chess match can be tricky when all you’re doing is ignoring your inactive contacts. With the value of an email address so high, it’s worth the effort to play a little defense. Don’t just forfeit your king and concede defeat.
If someone is predicted to unsubscribe, retarget them using Facebook or Gmail ads to reactivate them on your email list. The ultimate objective is to get them back on the active contacts list where engagement matters, and ROI is higher.
The battle for the inbox can feel like a last stand, surrounded by the enemy with little hope on your side. One effective weapon is Multi-Pass Messaging which can improve campaign results by up to 30%.
But you can’t take a shotgun approach when you need a rifle.
Sending a second copy of the same email might grab some attention, but it puts deliverability at risk: you are really just sending more messages to a lot of people who didn’t care about your first one.
After the initial campaign is sent, the contacts that are most Likely To Open should be the ones targeted for a follow-up pass. The engagement increases will improve deliverability and boost profitability of daily broadcast send volume.
Conventional wisdom holds that if you send more emails, you make more money. But on average 70-90% of messages never get opened – and that can’t be good for your return on investment.
So while a blanket approach might work to some degree, a more tactical segmentation strategy can offer a better ROI.
With predictive analytics we know which contacts are most Likely To Open and Likely to Unsubscribe. By suppressing contacts that are unlikely to open or are most likely to unsubscribe, retailers can significantly reduce send volumes and boost ROI of daily sends.
Display ads are notorious for rampant click fraud and low ROI when compared to email campaigns. Getting a 2-3X return won’t get you fired, but it won’t get you promoted.
Programmatic vendors offer 3rd party data solutions to help retailers target the right contacts, but your 1st party data trumps their data every single time. And there are ways to stack the deck in your favor using predictive models.
No matter your display ad partner, match your highest Predicted Lifetime Value customers for a lookalike audience that’ll put an ROI smile on your boss’ face.
When you’re selling consumable products, replenishment reminders are a no-brainer. Customers see this as a service and your boss loves the 6X conversion rate compared to broadcast messages.
But these messages can easily be lost in the clutter and, to make matters worse, we might be sending coupons to shoppers already standing in the checkout line…
We can break through the clutter by dropping an extra message to those Likely to Purchase (but they haven’t bought yet).
And using our Coupon Affinity predictive model, we can focus messaging to full-price buyers on scarcity versus savings.
Sending promotional emails to your full list is like an “easy button” for revenue. Subject lines with discounts and free stuff will always get shoppers to take notice.
But these discounts unfortunately become a crutch for many marketers and cut into margins.
To get smarter with promotional campaigns, think about suppressing your full-price buyers, and potentially sending them an alternate campaign that focuses more on new or popular products.
Most people think that once a person unsubscribes from your email list, they’re gone forever. But that’s just not true: they just become protected by the League of Marketing Acronyms, starring Captain CASL and AmeriCAN-SPAM.
The reality is that you can find those contacts in other channels like paid search and social. Just because someone unsubscribed from you doesn’t necessarily mean they never want to buy from you again.
Grab your unsubscribed contacts with a high predicted spend and circumvent the anti-spam force fields by retargeting them withGmail or Facebook Ads.
Testing is one of the most underrated aspects of a successful email marketing strategy. Good marketers are always testing their subject lines and improving their open rates as a result.
Incremental open rate improvements can help your overall engagement metrics, but there’s an opportunity to crank the degree of difficulty to 11. What about a targeted approach to engage your toughest customers, the ones most likely to unsubscribe.
In a Dr. Moreau-like island testing facility, expose likely unsubscribers to more aggressive subject lines to see what can bring even the most challenging contacts back into the fold. Learning from these tests can help prevent future contacts from becoming disengaged in the first place, keeping them“off the island.”
Browse abandonment campaigns are sometimes just what’s needed to push them over the edge with a simple reminder.The most basic series is a single message showing a shopper the last product they browsed. More sophisticated campaigns will send a series of two or more messages and include recommended products.
Once again, we have a shopper who has shown intent to buy, but we’re going to ignore the behavioral signals and stop sending after we hit our predefined message series limit.
Take advantage of engaged window shoppers and add more nurturing messages to a browse abandonment campaign if they’re more Likely to Buy. This keeps them interested in your brand and boosts conversion rates of an abandonment series.
Great marketers are always testing their strategies and tactics. But what are you really learning from your unsubscribed contacts?
By just letting contacts unsubscribe without engaging, you’re losing a lot of great opportunity for future customers. You’re also missing out on a chance to have a dialog and learn what you might be doing wrong.
By setting up a recurring campaign to customers likely to unsubscribe, you can test out a survey to find out what they are missing from your campaigns. The results can give a deeper understanding of customers that will improve your overall email and content strategy.
There are many benefits to sending to engaged contacts: higher engagement improves your deliverability reputation and you’ll simply make more money per message sent. But everyone is afraid to over message their contacts.
Guess who wants to hear from you more? The people who are most likely to open your messages and buy your products. They really can’t get enough, and chances are you’re holding back needlessly.
If you’re only sending a few times a week, try sending to theHighly Likely to Open and Likely to Purchase on several of the “off” days. If you’re already sending daily, this is a good opportunity to add a few extra campaigns to your highly engaged contacts and keep your brand at the top of their inbox.
Relevance is and always will be king for engaging customers. But while personalization of content is what we as marketers strive for, few of us achieve anything close to relevancy nirvana.
And nothing kills brand mojo like presenting women’s products to men, pushing shoes to those shopping for a blazer, or showing Mustang products to Camaro owners.
Using a Brand/Category Affinity predictive model paired with dynamic content, we can make sure that shoe fanatics don’t see sunglasses and Nike loyalists always get products with the “swoosh.” Simply use the model with either saved dynamic content for static header imagery, or campaign specific product and content blocks paired with shopper’s preferences.
Launching a new product line should be an exciting time, and retailers have to make sure their customers know what the latest product drops are. It seems like a basic concept, but a lot of retailers will push their new product line to everybody on their list.
This is appropriate if you’re adding an entirely new category to your catalog (like perfume alongside a cosmetics line) but is otherwise annoying. Sending to everyone will certainly include anyone who might be interested, but also includes everyone who isn’t.
Imagine you have men’s and women’s product lines, why would you announce a new line of sun dresses to someone who has only purchased men’s items? Things can be a little more complex if we talk about less defined product attributes, but if you’re launching a new line of baking cookbooks, you should target customers with a high affinity for baking and exclude or test as mall group of your outdoor grilling fans.
An industry-leading marketing cloud focused solely on retailers, Listrak delivers results for its clients with the power of 1:1 interactions that drive incremental revenue, engagement, lifetime value and growth. Fueled by artificial intelligence, actual human intelligence, machine learning and predictive analytics, the Listrak platform boasts a comprehensive set of marketing automation and CRM solutions that unify, interpret and personalize data to engage customers across channels and devices. Listrak has nearly two decades of digital marketing experience, serves more than 1,000 clients, and works with leading brands.
For more information, visit www.listrak.com.
.png)
.png){kind=icon}
