Email marketing best practices: Predictive email marketing replaces batch and blast

April 13, 2015

Email Marketing


While email marketing still continues to be the dominant form of communication, if you're still doing batch and blast, here are 3 reasons why you might want to re-consider:

1. You're losing subscribers the more you increase your email frequency. And the less engaged your email list, the higher the unsubscribe rate and number of spam complaints.  You may be saying, “but it’s working!” But while it may be working for some, can you say beyond a certainty it’s working for everyone? You’ve spent a lot of money acquiring your email base. Once they are unsubscribed, you either lose that audience for good or have to take more expensive measures to engage them (Facebook retargeting, display retargeting, catalog/direct mail).

60% of customer lifetime value is lost when a customer unsubscribes from your list. 

2. List fatigue will impact deliverability. If you’re like most marketers, your email list my exclude people who’ve bounced, but when’s the last time you removed inactive subscribers? At AgilOne, we call them “Sleepy’s” and “Phantom’s”. They may not have unsubscribed - and their email may not be bouncing - but they’ve stopped engaging. And as your engagement declines, SPAM complaints increase, and ISPs may start filtering your messages. Both impact whether your emails get delivered to that individual or to anyone at all.

It’s important to keep spam ratings low or you’ll forfeit deliverability for that ISP. Even if you’re following CAN-SPAM guidelines, you still can’t completely avoid spam complaints because some customers find it easier to hit the “mark as spam” button than unsubscribe. But, if you are over-emailing disengaged consumers, you’re increasing the odds of that happening. General thresholds should be no more than 1 email marked as spam for every 1,000 emails delivered.

Public ISPs like Gmail, Yahoo, AOL, Hotmail, etc. filter your messages based on engagement. If a consumer never opens or clicks your emails – even if they opted-in - ISP’s will auto-filter your emails and push them to the junk folder. So instead of driving someone to unsubscribe, the ISP's are basically doing it for you. And just like any unsubscriber, you’ll have to take expensive measures to re-engage them through re-targeting and re-marketing.

3. All the while, you're paying for it.  At the end of the day - to add insult to injury - you’re still paying to email consumers who aren't interested in hearing from you. Over time, the cost to send emails to inactive customers will grow, and the return may not necessary justify the price tag.

Email marketing best practices: Improve email performance with predictive marketing

Overall, data-driven marketers who are leveraging engagement and predictive analytics to segment and drive personalized campaigns are seeing much better results in their campaigns (open rates, clicks, conversions), lowering churn (unsubscribe, spam, etc.) and increasing incremental lift/revenues. Here’s how:


AgilOne email engagement

 AgilOne provides a breakdown of subscribers based on email engagement

1. Optimize email frequency based on engagement: With a Predictive Marketing Cloud like AgilOne, you’ll know who your email Enthusiasts are vs. Sleepy’s vs. Phantoms as an example. With this predictive insight handy, you can:

  • Optimize # of touches: Many marketers have a suppression rule to cap the number of emails their subscribers receive. Define the right threshold for each engagement level.
  • Reduce email costs: Target those who you know actually open their emails and don’t waste your email sends on those who don’t. If you have to stay within a specified email volume (which 99.9% of marketers do for budgeting reasons), make sure you’re not wasting it. In the above example, I have 74,000 Sleepy’s (8% of my list) and 120,000 Phantom’s (13.5% of my list). That means ¼ of the emails in my list are either inactive or have low engagement.  That’s ¼ more emails available to send to people who are engaging. How many of your subscribers are inactive (have stopped engaging with your emails)? How many emails are you sending these people per month? At what cost?

2. Optimize conversion with likelihood-to-buy (LTB): Similar to the email engagement segment, you’ll have access to every subscriber’s likelihood-to-buy (first time & repeat). Prioritize email sends to those who have high likelihood-to-buy and eliminate those who are least likely from receiving irrelevant promotions. Optimize your margin and response rates based on LTB by reducing discounts and offers to those with high likelihood, while sweetening the deal for those who may need more incentive.

3. Leverage distance-to-store and preferred shopping channelSince our Data Quality process is verifying addresses, merging duplicate contacts, plugging into National Change of Address & US Census, geo-targeting in our Metrics will allow you to see who is actually near a store and target them as opposed to those who will never come into a store regardless. So why email them?Additionally, since we track transactions across all channels, we can target those who prefer to buy in-store (or have bought in-store) vs. those who only buy online.

4. Cluster based segmentation vs. traditional batch & blast/basic segmentation is more effectiveLast but certainly not least, using clusters around brand affinity, product affinity, and shopping behavior allow you to truly segment and personalize campaigns as opposed to applying segmentation based on RFM, transactions, email activity, etc. 

Personalized email marketing