Skip the Generic Intro
We won’t bore you with generalities of how email as a marketing channel is tried and true, and how it will persist for the foreseeable future. Or that some unprecedented scenario requires your immediate attention.
Instead, let’s get into the meat and potatoes – as the grandparents used to say.
Let’s assume you’re already running email campaigns with personalized content, messaging, offers, or product recommendations. Check.
When to Personalize?
(Time versus Behavior)
Any email you send should have a degree of individualization. We’re on the same page about that.
However, when should you send an email? Under what circumstances is it most valuable to send one of your shoppers an email?
To be clear, this “when to send” discussion is not about the optimal time-of-day or day-of-week that tends to get more engagement, opens, clicks, and conversions. A Google search will provide 50 articles on that. (And perhaps almost 50 different answers!)
Instead, we’re talking about a different “when”: behavior-based or contextually-relevant indicators that underscore high-value opportunities to launch email campaigns.
Let’s break that down:
- There are person-based scenarios where it’s more valuable to send emails
- These scenarios are based on the context around an individual shopper
- By triggering or scheduling email campaigns under these circumstances, you increase the likelihood of email engagement and conversion
In summary, opportunities arise based on a shopper’s behavior for:
- Specific triggers you can use to launch emails;
- Valuable email types you should be sending;
- Relevant messaging to include; and
- Timely product recommendations to insert.
What are those different campaigns, content types, and all the behavioral triggers? That’s why we’ve all gathered here today. Let’s dive in.
Value-Based Email Sends
In the following blog post, we’ll be reviewing 9 different email campaigns and tactics to run and use for engaging your customers.
It goes without saying… the cohorts we’ll be targeting are known audiences. You have their email address and thus can communicate with them directly. Because they have engaged with your brand at some point in the past – purchase, form-fill, etc – you have some stored information about them.
For our purposes, it doesn’t matter if this shopper was added to your customer database today or 5 to 10 years ago.
The idea is: you know something valuable about these shoppers. You can attach some history of activity – or new, real-time activity – to that profile to inform how you communicate with them.
What defines a valuable scenario to send an email? We split this into three broad categories:
- The customer has an intrinsic high-value profile, defined by past behaviors.
- The customer’s brand-engagement indicates intent or near-term propensity to act.
- The lack of customer engagement for some period of time exposes an opportunity to explore.
The email campaigns that follow fall into these categories in some way – as should really all value-based email sends.
1. [CRM Customer Base] Calculated LTV Campaign
In this email campaign, we are identifying customers that have an intrinsically high-value profile. Being holistically high value, we can periodically select contacts in this group for a campaign to drive purchases.
The clever way to identify this intrinsic value of any given shopper is to generate a predictive model of customer Lifetime Value (LTV) – a sum of the total dollar value of purchases this customer has made and is projected to make for their lifetime.
Side Note: While some LTV calculations set a timeframe of 1 year or 3 years for the purchase window, using the full lifetime projection is another formula that typically provides a slightly different relative value determination between customers.
The emails that launch to this high-value group can be sent as part of a daily, weekly, bi-weekly, or monthly send. It depends on a few factors, most importantly how large your list is and what goals you need to achieve.
These types of campaigns can also be leveraged when your team needs to meet sales goals for a given period. Sending to these prioritized shoppers reliably drives more revenue and conversions.
An example might be: your top ~1000 high-LTV leads would be targeted each week (who aren’t already in another campaign). LTV segmentation can also be layered with other creative ideas!
2. [Website Engagement] Calculated LTV Campaign
This behavioral campaign combines the LTV calculation from the previous example with another set of data for weighting: website engagement.
Recent website engagement is used to determine another facet of a shopper’s “value”. This weighted value component introduces a time component.
→ Calculate the predicted likelihood to purchase in the next “X” number of days.
So this campaign effectively targets customers based on a weighted LTV score. The multiplier – the type and amount of website engagement – helps you to email customers who are most likely to buy immediately.
The type of engagement your business cares to prioritize will dictate what you actually send! For example, if you are looking to sell non-discounted, New Release inventory, you can prioritize website engagement that looks like: spent 5+ minutes on the website, clicked into the new releases category, and viewed 3+ New Release products. The email copy and creative might promote the most relevant, personalized New Release products.
3. [Post-Purchase Upsell] New Product Discovery Campaign
This email campaign focuses on upsell – capitalizing on momentum & brand familiarity.
After a purchase, within 2-10 days, this email is calling the customer back to the website with an additional set of new products for the customer to discover. It can be triggered automatically based on your timeline.
This discovery campaign should utilize dynamic product recommendations in the email that match the customer’s purchase. Ideally the products displayed in your email are either “frequently purchased with” the shopper’s most recent product transaction, or based on another algorithm that predicts a high degree of relevance for that shopper.
Upsell campaigns are important to maximize your customer purchases, especially given high CPAs and more expensive advertising costs. This helps to meet those goals by keeping your brand – and new, undiscovered products – top of mind after the momentum of a purchase.
4. [Delayed Post-Purchase Upsell] Supporting Accessory Campaign
This is another upsell campaign that has similarities to the new product discovery campaign described above. However, this version has a time delay to allow the customer to establish some real familiarity with your product. This delay might be 30-90 days.
Your store likely has some accessories that go well or support the already-purchased item. This is the opportunity to drive more sales for those supporting accessories.
By creating a delay, you’re creating some breathing room between purchases and assuming there may be reasons your customer may not be willing or able to make purchases in quick succession. Importantly, it also lets your shopper use, experience, and understand the item they purchased better. As a result, they’ll be more likely to desire the accessory you email them.
This email should mention or display the previous product shown, and then include accessories – either based on category, prediction, or even manually hand-picked for specific relevance.
5. [Website Session] Abandoned Browse Campaign
Shoppers visit your online store all the time, and a great deal of these are anonymous. Unfortunately, it’s very difficult to send email campaigns to customers when you don’t have their email address. We’ve tried.
However, you can often recognize a user: perhaps eCommerce software, like Shopify, recognizes their user ID. That ID can map to an email address between systems. Another common case is a web browser like Google Chrome recognizing a logged-in shopper. Obviously a subscription lead form helps too.
Once you know who someone is, you can use the interaction of them visiting your website to trigger a meaningful email to follow up on their behavior. We call this the “Abandoned Browse” – as opposed to Abandoned Cart. For example: a shopper lands on your website and starts browsing within one category, maybe clicks into a couple different products, but ultimately closes the tab and moves on.
This opportunity to reach out might be obvious, but we would also advise to layer on dynamic content. This can be personalized to not only the category that customer was browsing, but also some more specific product recommendations that can be predicted from similar shoppers’ past behavior and other predictive modeling.
6. [Website Session] Abandoned Cart Campaign
This email campaign is fairly well understood and table stakes for many eCommerce stores. Customers indicate high intent to purchase a specific product when they add it to their shopping cart on your website. In these cases, we want to capitalize on the high intent and follow up very quickly.
The email here should include at least one of a few types of content:
- A) a friendly and fun, if relatively basic, reminder that there is something waiting in their cart and what that is;
- B) information to nudge the shopper across the finish line, such as social proof, urgency indicators, or similar; and
- C) an easy way or clear instructions to help them complete the purchase – allowing one-click to land on their pre-filled cart to check out, or similar.
Not only should these email campaigns be sent pretty quickly following a shopper adding something to their cart without a subsequent purchase, but there should also be more than one ideally. Perhaps a same day or next day email would be the first, but 5-7 days later might trigger a second.
7. [Website Session] Advanced Search Filtering Campaign
When shoppers are poking around your site, they leave behind helpful data that can be powerful indicators to understand their purchase intent and affinities.
When they are using advanced search filters – to input their desired specific categories, sub-categories, price range, colors, materials, styles, and more – you’ve just gathered an impressive set of extremely valuable data.
The customer basically just filled out a long webform that answered tons of questions about their current preferences for products on your store.
The next step pretty much goes without saying! This is a more nuanced and powerful version of the “abandoned browse” email campaign listed above. So your job is to target these shoppers with emails that capture their obvious high intent – and couple that with really specific product recommendations, and other personalized content that shows you’ve paid attention to their interests.
8. [Predicted Disengaged Behavior] Subscription Churn-Interception Campaign
While this campaign is certainly the most applicable to eCommerce stores that have a subscription model, a similar model of customer lifetime and “churn” can be generated with one-time purchases.
For this valuable campaign, you are predicting the timeframe and likelihood of a customer “churning”. Meaning, they cancel their subscription with your brand, or go dormant long-term and don’t purchase anymore from you. You can intercept them before the churn and offer them something to prevent it.
Because many stores aren’t frequented on a monthly basis – it may be difficult to imagine predicting this “churn” point for more expensive and less frequently purchased items. (Who buys a Rolex watch every year?) However, the right predictive calculation can approximate this point for non-subscription models.
In either case, the churn prediction interception campaign should use content that convinces the customer to keep things going, either by offering incentives or perhaps by reminding them of past value.
9. [Predicted Disengaged Behavior] Delayed Re-Engagement Campaign
A powerful email campaign to run is to re-engage customers that are disengaged. These email sends are all about good timing.
To successfully execute a campaign like this, you should have some predictive models that look at past shoppers who had engagement – and then layer-in their disengage & re-engage patterns. With this you can learn what to look for in the future that might indicate opportune times to run a re-engagement campaign.
A relevant example might look like this:
A shopper purchases a nice pair of jeans, and then over the next 4-5 months they visit the site a number of times and end up purchasing one additional item or accessory. Then after another 3 months they haven’t engaged at all.
A targeted email campaign with a promotion, sent 6 months after their last brand-interaction, is very likely to re-engage that person back to your store. They will either end up making a purchase – or at worst they’ll interact with some products, filters, and pages. And that gives you data to personalize subsequent follow up emails!
By figuring out the set of previous behavior that indicates likelihood to re-engage in the future, you can make meaningful use of the huge body of past customers that may be sitting in your CRM unengaged.
This one definitely takes a bit of processing power: using a machine learning intelligence engine. But identifying the timing & behavior to trigger this type of email campaign can have big results.
Non-Behavioral Email Campaigns
Plenty of other email campaigns run based on triggers that are not “behavior-based” in nature.
Top scenarios to send emails are based on your internal product and inventory information. For example: low inventory, brand new product drop, or recently re-stocked inventory for a hot item. It also could be related to excess inventory you’re eager to sell at discount.
Additionally, it goes without saying that sales and holidays are essential campaigns that are triggered by internal planning and seasonality, rather than the behavior of shoppers.
Non-behavioral email campaigns are powerful, and typically come from some internal business trigger or event. Advantageously, the campaigns defined in this guide – triggered based on customer interactions with your brand – tend to have great conversion rates.
The email campaigns listed here are triggered from consumer behavior and value: either reacting to website behavior & purchases, marketing engagement, high LTV, or lack of engagement for some time.
The idea is that you’re anticipating customers’ needs and trying to act in real time to meet them.
Some of these emails can be simple triggers from website behaviors and timers, others require more intelligence to be gleaned from all that transaction, website, product, and marketing data. To layer on robust behavioral predictive models, an intelligence engine like Aidaptive makes sense to add to the tool kit.
Not only should these emails be sent and timed based on relevant behavioral triggers, but the content therein must be appropriate and relevant. This takes testing and iterating.
You’ll need some additional business automation and intelligence to generate and experiment with the email content & product recommendations – especially to test and iterate at scale for each individual shopper.
Advanced eCommerce marketing transforms email marketing (which is well-established, understood, and not very creative) into a more effective medium via optimization techniques: from timing, triggers, content, and personalization.
If any of these ideas resonated and inspired you, let us know! We love to hear feedback: [email protected].