Are You Sending Price Drop, New Arrivals, and Back in Stock Emails to The Wrong Customers?
Create revenue-generating Price Drop, New Arrivals, or Items Back in Stock emails with our proprietary AI models that will easily find your customers who have a high affinity to an item. Keep in mind, this feature is not the same as our product recommendations! Instead of grouping items from a user, we figured out how to do the inverse. Think about it. If you’re sending out a discount email, you want to find the most efficient way to group users who all have a high affinity for a specific item.
Here at ReSci, we call it predictive product segments. What this means is that our proprietary AI models are baked into our lifecycle stages. Predictive product segments are an inherent function of our Price Drop, New Arrivals, and Items Back in Stock.
Target excess inventory to interested buyers
Retailers may run into an issue with excess inventory due to over-ordering. What’s your solution? A marketer might begin an email or ad campaign targeting users who have interacted with your product within the last month. The campaign might feature a discount to entice prospective customers. You blast an email, but you end up targeting users who simply aren’t interested. Worse, you miss a large segment of users who have never encountered the product but otherwise would have been interested. That’s a lot of potential lost revenue.
Marketers will discover that simply targeting a cohort of users who have browsed an item doesn’t necessarily translate to true product affinity. Engagement doesn’t equal personalization. To solve this problem requires the necessary relationship between customer behavior and data mining. Without the ability to properly extract and process information, a marketer’s ability to understand customer behavior will be flawed.
Let’s go back to the original problem.
Your last campaign produced some results, but you’d like to re-target your excess inventory based on available feedback. You may want to blast a price drop email, but the only available feedback is your last campaign’s performance. What you’re left with is a group of users who have a chance of not sharing the same item affinity. You simply don’t know.
Predictive product segments solve this problem by analyzing your entire customer base, not just a single campaign performance, then groups users who have a high affinity to that item. Marketers deal with math every day. When you’re looking at your email’s CTR performance, you’re placing your faith in data that has an algorithm behind it. You’re limited, however, to insights into individual human behavior. Most marketers aren’t building advanced systems to leverage their campaigns but they do have a repository of tools they simply aren’t utilizing.
Tools such as predictive product segments take into account minute variables in individual behavior and then wrap it in an algorithm to build an effective strategy. A strong feature of our AI is being able to target customers with high item affinity, but now we’ve added the wish list as an additional feature. For example, you may want to send out a price drop email to potential buyers and also target users who have liked or stored an item in their wish list. Marketers can now do both without the guesswork.
The end result is your ability to target highly engaged users and dramatically increase the chances of re-engaging previously churned customers. Even more, you can upload a list of customers to Facebook Audiences to run a highly targeted ad campaign.
Maximize interest in your new arrivals
You have some cool new items for your customers and you want to maximize user engagement and interest. A marketer with a standard ESP platform would include these new arrivals in a weekly blast to every user in their list, whether it is relevant or not to the user. You not only risk over-emailing your customers, but you also risk losing them and damaging your email reputation.
You may be wondering how our predictive product segments can successfully target users if there’s no user interaction with the new product. Good question. Semantic similarity is a powerful metric used where meaning between different terms determines their likeness. We use title, name, meta tags to drive semantic similarity, and we also look at co-purchase activity.
With predictive product segments, your email campaign becomes highly profitable by targeting users with the highest predicted affinity to the new item. Prediction is the operative word here, and it might sound scary. After all, predictions can be wrong. But it’s important to realize marketers make predictions every day. When you decide to email blast a list of users who you believe to be interested, you’re making a basic prediction, albeit with very limited data.
Consider the fact that our AI processes 2K actions per second, and then consider these actions are very different from each other. More data is always good, but it can also leave you with a greater margin of error. The advantage of our AI is that it’s able to parse through enormous amounts of data to make an informed decision. In other words, as a marketer, you want to reduce the chances of error.
Start sending new arrival items to the correct user by leveraging your email campaigns using machine learning and predictive AI modeling. Moreover, find new customers by pairing your email campaigns with a Facebook lookalike audience who have a strong affinity to your new product. In return, you reach potential customers in a highly targeted and unique way. Many eCommerce companies are taking advantage of this targeting method.
Recapture lost revenue from out of stock items
Unfortunately, you ran out of stock of a hot product. There’s nothing worse than losing revenue due to lost inventory. A few weeks later, your supplier ships you a thousand new units. What’s the most efficient way to get the word out to make up for the loss?
A smart marketer could run a campaign to include users who viewed the product on site. This may produce a decent result. But again, viewing a product doesn’t provide the marketer any deep insight into the actual behavior of the individual. In the long run, you’ll be weighing between best guesses rather than data that’s meaningful for reaching your goals.
A marketer equipped with predictive product segments and our wish list feature will have a highly targeted campaign. Rather than relying on a manual trigger program, you’ll be equipped with powerful software that will increase your chances in recapturing lost revenue, retaining and acquiring customers and future customers.
Tips: Use our segmentation capabilities to give your VIP customers a first shot at newly restocked items, again leveraging predictive product segments. After a few days, you can then reach the rest of your customers who may have an affinity for the product.
You can find the details of this predictive AI algorithm here: Sending Item Alerts to the Right Users.
Want to learn more about the power of predictive segmentation? Read how Perfect Snacks saw an 18% increase in average order value.
Did you find this useful? Share it on social media.
ReSci is a team of marketers and data scientists on a mission to democratize AI. We make powerful recommendations and predictions accessible to brands. Find out how we can help you connect with your customers.
