Resources
Recovering email deliverability when reputation goes bad (Part-1)
Written by
Mathew Vernhout
Mathew Vernhout
0

Recovering email deliverability when reputation goes bad (Part-1)

Introduction

Open rates are falling, clicks are almost non-existent, conversions have totally dried up. Do any of these sound familiar? You may have a significant email reputation issue on your hands. But how do you fix your reputation issues? Unfortunately, there is no straightforward answer to this question; like for most things deliverability-related, the answer is “it depends.”

In this series, we will cover a few things to help you diagnose your issues and resolve them, as each of these items takes a bit of time to prepare for and manage. We are assuming a total reputation rebuild is in order. However, many of these may help, with some modification, to address reputation issues that are early on.

Tools you may want to consider while looking into your reputation rebuild – we will cover these later in the series:

  • Google Postmaster tools
  • GradeMyEmail
  • Inbox placement monitoring tools
  • DMARC monitoring software

Part 1 – It’s all about the DATA

It’s not surprising when I look into delivery issues that I start by asking about data sources and data practices. Historically, fixing the data issues fixes many delivery issues as well. But where do you start? 

First things first, suspend sending to any data provided through a data source outside of your direct control (ex: partner, co-reg). Anything you didn’t collect should be segregated from your mail database, put on the shelf during the rebuilding process, possibly forever. We will get back to that later, though.

It’s also important to fully understand the sources that are adding new names into your email program. Are these data sources clearly setting the user expectation for emails that they will be receiving? Are you aligning your branding appropriately – ex: subscription at domain.com, with emails being sent from mail.domain.com? Are they using the correct consent models you expect them to?  

A good opt-in process precisely sets out these expectations for a subscriber:

  • Only use sources of data where you directly control the opt-in and are consent-based
  • The frequency of the mailings that they should expect to receive
  • Explain to the subscriber the types of content that they should expect from your organization
  • Security features to prevent form abuse are implemented, including but not limited to; 
    • Rate limits on form submissions, reCAPTCHA, real-time email validation
  • Provide easy-to-find access to your organization privacy policy and contact points for the brand

If you’re not already tracking subscribers by collection source, there is no time like now to consider adding a SourceID into your data set; it will help you measure each opt-in source’s effectiveness against others. This allows you to identify underperforming sources and compare them with your high-indexing sources. 

Pro tip:

  1. If you’re having significant data and delivery issues, look at implementing Confirmed Opt-in (COI) for all subscribers. 

Now that we know where the data originates, let’s look at how the data is being managed.

Data Management

Data flows, journeys, and sunsetting, Oh my! 

Meeting your customer expectations should be your focus when setting up your email programs. This includes sending emails from brands that your customers recognize, sending from brands that the user remembers and wants to engage with.  

Understanding how your subscribers are engaging with your brand is very important for this next part. This could include changing email cadence, message content, or even sunsetting certain segments from your subscriber list. 

Determining segments to review can be divided into many different classifications; for simplicity’s sake, we will look into time-based methods now, and revisit other segment strategies later. 

Example segments you can use as part of your reputation rebuilding process look a lot like IP warmup segments. 

  • Segment A: Unique opens/clicks + purchasers in the past 7 days
  • Segment B: Unique opens/clicks + purchasers in the past 14 days
  • Segment C: Unique opens/clicks + purchasers in the past 21 days
  • Segment D: Unique opens/clicks + purchasers in the past 30 days
  • etc… up to 90 days

This works best if you pause your mailings for a short period (2 or 3 weeks ideally) to rest the domain and get your data in order, for progressing with the rebuilding process. 

Send to each of these segments in a progressive order over a series of days. Keep each segment separated on their own day but watch that your volume doesn’t grow faster than 30% a day. 

Yes, this is slow and can be painful, but we are completely rebuilding your email program’s reputation. 

Since these are your most active and engaged subscribers, they are going to be the most interested in hearing from your organization during this process. Follow this process for adding in additional segments and subscribers each day as you work towards the 90-day segments slowly, with no more than 30% list growth from day-to-day.

Keep a close track of your email metrics and be ready to hold volume or segments back that are not performing well. At some point, it is common to see metrics plateau or even get worse as you add more data. This likely indicates that you have reached the outer limits of your customer’s tolerance for your standard email program, and a shift in tactics might be necessary.

Aligning your data inputs and your segmentation strategy is the first step to a total rebuild of your email marketing program’s reputation.

Next time, we will cover welcome and sunsetting programs you should consider implementing to engage your subscribers early and keep them engaged longer.

Unlock unmatched customer experiences, get started now
Let us show you what's possible with Netcore.