This is a guest post from Brian Dayman at Kickbox.
So you’ve decided to migrate to another ESP.
Whether it’s through improved email workflow, more complex metrics tracking, better segmentation, or anything in between, taking on an ESP migration can help you connect with more customers and take your marketing to the next level.
But before you can reap the benefits of a new ESP, you’ll need to go through the migration process, which has the potential for a lot of headaches if done with a “fly by the seat of your pants” approach.
No matter your industry or marketing team’s size, there’s a need to manage a myriad of migration tasks like galvanizing a team of stakeholders, managing the project, learning about your new technology, mapping, and, perhaps most importantly, cleaning up your data.
Not the most glamorous task, but one of the most critical
To be honest, taking responsibility for cleaning up a marketing database isn’t the most glamorous of jobs to take on. However, it’s necessary if you want to make the migration process as seamless as possible and reap maximum rewards from it.
Migrating ESPs is a prime opportunity to make sure your data gets the attention it doesn’t always receive. Separate from standard list hygiene, preparing for platform migration is a perfect time to conduct a deep clean of your database. And the reasons for doing so are far from arbitrary.
Cleaner data now is more ROI later.
The truth is that, from small to large, all companies have dirty data (up to 25% of their entire marketing database).
This is especially true of those with older subscriber databases or using multiple channels to collect data.
For example, an ecommerce marketer with multiple campaigns for each product or an agency with multiple clients’ lead data to manage. On average, organizations use 3.8 channels to collect data, each potentially incorporating multiple points of capture.
At Kickbox, we’ve found that 12.5% of email addresses entered on signup forms are initially invalid due to misspelling or attempting to use fake addresses.
Compounded over time, the database decay can cause deliverability and email ROI to take a hit when sending through your new ESP.
Additionally, your new ESP will often analyze your mailing lists for quality concerns when migrating to their platform, so it’s best to be prepared for that by cleaning up any risky data that could be hiding spam trap risks or other IP blacklisting concerns.
Even if your lists are not flagged during the migration process, cleaning up your database can still optimize performance and mitigate any issues that’ll lead to lower inbox placement and wasted marketing dollars in the long term due to poor sender reputation.
And, with increasing data privacy laws and evolving anti-spam technology used by ISPs, it’s more important than ever to keep your relationship in good standing with both senders and subscribers alike.
The cleaner you get your data now, the more benefits you can reap later. One of our high-volume sender customers, albeit through a separate process from a platform migration, recently shared their results for bounce rates for their transactional email after incorporating some data hygiene practices.
As they taught us in the military: make your bed first thing in the morning, because, when you come back from a day of work and you’re tired, it’s going to be much harder to tidy things up. In other words, be considerate of your future self.
To help you with your approach to data preparation, here’s a three-step process to follow when preparing your data for your new ESP as efficiently and painlessly as possible.
1. Commit to being the marketing data champion of your team.
Depending on the size of the marketing team you’re working with, there’ll often be a team of internal stakeholders to make the migration happen.
For example, an engineer will often own the technical parts of the migration. A business operations leader might make decisions on which data is most critical to maintain and how to organize it.
The marketing leader will often own the email database, as well as the messaging and segmentation within the new ESP. A project manager might ensure all roles and tasks are owned and track the deadlines.
As a marketing data champion, you’ll often be dealing with two main categories of data: event data and subscriber data.
Where event data is associated with actions taken within individual campaigns like those associated with specific products, subscriber data is personal data, and is foundational to your marketing. After all, the only profitable contact in your database is one that you can reach.
Before kicking off your subscriber database cleanup, be sure your role as data champion is defined among your team.
You should coordinate with other stakeholders to ensure you can surface future questions about field mapping, integrations, and business goals as you organize the marketing database in preparation for ESP migration.
2. Identify the good, the bad, and the ugly.
Before we can begin the cleanup process, we need to get a gauge on the health of our database and unveil risky data that’s hidden within it.
But defining what’s risky can be challenging if you don’t have a clear definition of what a low-quality address is, aside from engagement rates.
Start with deliverability instead of time since last engagement.
Conventional marketing knowledge tells us to treat unengaged subscribers as risky data. And, while there’s definitely a correlation between unengaged legacy subscribers and risks like spam taps or future spam complaints, simply flagging emails as risky by using only a time-based standard can lead to missed opportunities.
In some contexts, sending an email to a deliverable contact will have a beneficial long-term impact even without a recent open or a click.
In The Email Addiction subscriber study, it was found that over 50% of unengaged subscribers are either waiting for the right offer (37%) or because they expect to buy from the brand again. (24%).
Additionally, data champions don’t always have full insight into the health of legacy email data if they weren’t directly involved in its acquisition (e.g., tradeshow leads uploaded by field reps or leads from another marketer’s campaign).
That’s why it’s so important to start the database cleanup with the Deliverability-Based Method.
As opposed to a time-based method that tosses out emails that haven’t been engaged in a certain timeframe, the Deliverability-Based Method uncovers risky data based on how likely these emails are to reach the intended recipient.
This is done through email verification and serves as a lower-risk method for testing emails than loading them into your new ESP and hitting send.
If verifying legacy email data, then list or bulk verification will be used, and can often be done through integrations with your old or soon-to-be-old ESP to gauge the health of your current email lists. Soft-bounced lists should be included in this process as well.
Once verified, high-quality deliverable email addresses will be distinguished from risky email addresses that could cause issues with the ESP migration process or performance after migration. Examples of risky addresses are:
- Accept-all: the domain accepts all email you to send to it, even if the email address isn’t valid
- Disposable: one-time email addresses often used to receive initial communications from a service (such as an activation email) and are discarded.
- Role: an email address that’s associated with a team or job function instead of an individual.
- Free: potentially risky. Free email such as Yahoo and Gmail can receive lower open and deliverable rates in certain contexts, although B2C businesses generally benefit from accepting free email services.
- Undeliverable: low quality—the email address doesn’t exist or is syntactically incorrect.
Once you’ve verified a list, you should be able to see the health of your marketing database, both by individual lists and at a glance.
By unveiling the deliverability of email addresses in your database, you can have an idea of which contacts are worth re-engaging and potentially preserve the value of the list you’ve worked so hard to build.
Single out the other obvious culprits.
In addition to identifying risky deliverability statuses of your database’s emails. Duplicates, redundant data fields, and hard bounces should also be removed.
This is where exporting your lists and rolling up your sleeves will be necessary, but make the actual data cleanup that much easier.
Now that you have your quality and risky email addresses organized in separate bins, you’re in a much better position to take the next step to shaping up your database for migration.
3. Clean up the dirty data.
Once you’ve identified the good, the bad, and everything in between, it’s time to do the actual cleanup. The cleaning of your database for your migration involves two parts: making decisions on what to prune and preparing your suppression lists.
Decide what to prune.
Undeliverable, unsubscribes, and hard bounces should be a starting point for flagging what should go onto your suppression list.
From there, it’s time to look at your current and new ESP’s segmentation options, analyze the business impact of each deliverability category, and decide which ones should be mapped or suppressed.
While you don’t have to sit on a throne like Cesar and give the thumbs up or thumbs down to each segment or category, consideration should be given to the business impact of each segment before deciding which to keep.
This is where Step 1 comes into play and coordinating with other stakeholders (if there are any) to decide what subscriber data to carry over based on your audience, product, industry, and business goals.
Update your suppression lists.
Now that you’ve done the heavy lifting of identifying and deciding what data to prune, it’s time to clean up your database. This should be done through your suppression lists.
When pruning your contact list, simply deleting the contacts isn’t enough. Instead, contacts should be added to a suppression list as a failsafe to stop the email address from being re-engaged and to comply with anti-spam laws.
Hard bounces are automatically added to suppression lists, so be sure to include those from the ESP you are migrating away from.
Once you’ve cleaned up your marketing database, your team will be ready to take on other possible steps in the migration process like domain setup, authentication, IP warmup, and perhaps further tweaking your segmentation strategy.
The process should be a gradual one, and could be more complex based on your database size and your business goals. But, with a clean database, getting up and running with successful campaigns will be much less of a headache and you’ll be in a position to maximize your email marketing with your new ESP.
Brian Dayman is a Marketing Manager at Kickbox, a technology provider that promotes email best practices to improve deliverability. The mission: ensure customers with opt-in contacts get their message to the inbox and to prevent all the rest from hitting send.