Specializing retention emails using behavioral data

  • 28 March 2022
  • 0 replies

Userlevel 5

This content, written by Andrea Vidovic, was initially posted in Looker Blog on Aug 3, 2017. The content is subject to limited support.

The “user experience” can now be understood to span everything from packaging and delivery, to customer service and troubleshooting, to social content and email communications. People have come to expect impeccable UX across the board, therefore, to remain relevant within a customer’s inbox you must curate highly original experiences at scale. Using Looker, you can identify users ready for various types of nurture, build specific content for them, and showcase value with each and every ‘send’. In this post we'll talk about different types of emails and how to ensure you are sending the right email to the right person, at the right time.

Customer nurture like you really mean it

Retention emails are an integral part of maintaining a solid relationship with customers who have ‘qualified themselves’ as interested (albeit to varying degrees). Customers may express faint interest with your business; requesting more information, submitting personal data for a ‘free trial’, or bookmarking your product and/or service for review. Customers past these initial stages are heavily engaged with your product; using the product, requesting upgrades, and perhaps making several related purchases. Then there are customers who have adopted loyalty with your product, perhaps you’ve even hit the holy grail of evangelism with this segment — these individuals should never be taken for granted. Just because a customer engages with you does not mean they’ll actually come on board and once on board it doesn’t mean they’ll stay, and if they adopt you as “the one” it’s your responsibility not to break their hearts.

Inactivity emails

Just like it sounds, this genre of email acknowledges users who seem to have ‘dropped off the map’. There are various approaches to re-engagement depending on what data you are analyzing and what immediate goal you are trying to achieve. User-behaviour that would be helpful to map in this case would be: how long users have been inactive, at which point users tend to drop off, when they seem to ‘wake up’, and what message inspires them to participate? Here is an example of the tried and true coupon-offer approach:

In this email GrubHub is offering me a deal since I haven’t ordered in several weeks and have been known to order more frequently. Now, I’m not sure whether it’s a happy coincidence or not that they have featured Asian cuisine (my favorite) but I am finding myself super enthused about potentially ordering dinner tonight. If they did do it on purpose then good on them, and if they didn’t do it on purpose then they might want to try personalizing food visuals more often! That is to say, it would be cool if user behaviour was measured to find out which foods are the ‘most-ordered’ or ‘general favorites’ so email templates could be customized to match beloved cuisines to eager foodies.

Within Looker, you can identify friction in the funnel (i.e. seeing where users become inactive) and confidently leverage user behaviour to minimize inactivity by sending nurture at weak points.

Activity emails

Conversely, activity emails can be spurred by a variety user actions (and even some combination of inaction). Below is an email that was sent in response to the fact that I visited multiple pages of this site, technically ‘abandoned cart’ at the end of browsing, later revisited the site several more times over the course of fifteen days to see what was available.

In this email, BarkBox is ‘testing the water’ to see if I might benefit from trying a short-term deal (like a teaser of the real thing). From my spontaneous online behaviour, it seems that BarkBox was confidently able to infer my reluctance to commit to a lengthy subscription (being a new user and all). Their strategy is to empathize with me, allowing me to become more familiar with their service in hopes of convincing me to commit to the standard subscription at a later date, once good rapport has been established.

Upon being connected to your database, Looker enables you to analyze more granular event data so you can examine the user journey the way you want. It’s like you’re granted a 360 degree view of how visitors are interacting with your website, with the ability to drill all the way down specifics to help you piece together even the most puzzling customer journey.

Renewal emails

This email is sent in anticipation of a renewal of a service, and is best sent as a series. An example drip strategy is 60 days before expiration, 30 days before expiration, one week before, the day of, and then a week or so after if you haven’t managed to convince them by then. Looker analysis can help you identify the ‘sweet spot’ to send this email, thus eliminating the guesswork so you can focus on creating more valuable and entertaining content. The email below did a great job of reminding me I should probably touch up my roots:

I do short subscriptions (instead of yearly renewable) with Madison Reed because I’m really fickle; I like to change up my hair color every three to six months. To remind me it’s almost time to select my shade of choice, I get an email from Madison Reed the week before I’m due to color my hair (during the onboarding I designated that I’m on an eight week cycle). This type of email is valuable because it helps busy-bodies stay on top of the purchases they regularly make.

Cross-sells & up-sells

To cross-sell is to promote a different (yet complementary) product/service to an existing customer. To upsell is to promote an upgrade or an add-on to an existing customer... Amazon.com does a lot of both! The company regularly up-sells Amazon Prime services with pop-up promotions, usually while I am at the virtual ‘check out’ on the website. I also receive a weekly digest of items related to my past purchases or based on my personal settings. The email below is a digest of newly added books I might find interesting:

Here Amazon takes into consideration that I have repeatedly bought books from them, I have a designated interest in Political Science titles; and though it has been some time since I have bought a book, I almost always click-through when I get this email!

VIP offers

Promotional offers are always a joy to receive, but it means something more when a company recognizes your devotion (having dropped mad cash repeatedly on their products) and sends you an exclusive perk. For example, GAP invites me to ‘shop first’ and beat the crowd to the big deals:

Offering two different promotions (one online and one in-store) allows me the liberty to shop however I please! Although, this is also an interesting opportunity to monitor what I end up choosing; in Looker, you would be able to see how shoppers move on VIP deals over time. By identifying trends you can organize more thrilling customer experiences accordingly (i.e. throwing special events at stores or gamifying the online shopping experience, for example).


An email featuring an invite to take a survey is sent to ‘check-in’ with new customers to see how adoption is going (within five to thirty days), to gain feedback on service updates accepted existing customers (within fourteen to thirty days of an update), or to present existing long-term customers with the opportunity to personalize their own experience with a service, such as the one below:

Surveys are probably the most dreaded (yet most useful) B2C interaction. In Looker, you can find out what the best timing is to present a survey and keep track of how a variety of incentives fare in stimulating big turnouts. While I found this email to be a bit lack-luster in plain text for Adobe (a company that basically sells creativity), I was happy have a chance to curate an ideal experience for myself regarding the Adobe Create Magazine.

Email metrics can impart unparalleled insights

Email data is a like a treasure trove for behavioral studies all on its own — every customer interaction (or lack thereof) can lead marketers to form a deeper understanding of their performance:

  • Which emails did a user open and/or click on?
  • Which emails did a user ignore? Which missed the inbox completely?
  • Which emails caused users to unsubscribe?
  • What type of offers do users respond to most often?
  • How long ago was a user’s last interaction with an email (days, weeks, months)?

Running a simple study (like the one outlined below) can help marketers more acutely discern what customers expect at various stages of the relationship-building journey.

  1. Begin by sorting users into 3 main groups according to frequencies of interaction: high, moderate, and slight.
  2. Then organize A/B tests for copy, offers, template design, timing, even the levels of personalization within each of these groups.
  3. In comparing the results, notice what resonated within each group and use this data to form best practices.

All in all, the way users behave with email should help shape what kinds of campaigns you build, how often you send them, and inform what stage in the relationship is best for each communication. Business analysis in Looker can help you find the most optimal segment to reach with each campaign, as well as pinpoint opportunities to run spin-off campaigns. Retention emails are supposed to feel like a piece of a grander conversation; keep it personal, keep it fresh, create experiences that build and expand upon the last.

See Looker and Sendwithus in action!

Visit the to learn how the sendwithus email optimization platform can help your marketing team create better email experiences. For a trial of sendwithus, simply sign up .

0 replies

Be the first to reply!