There are so many tools of the trade for email marketers, it’s obvious email service providers (ESPs) would be something of a big topic in the industry. Which ESPs provide which services, do they impact your deliverability, how do you decide which one is right for you? After those questions comes an even harder one: How should I migrate from one ESP to another without causing damage to my email marketing program?
Our next Email Expert Series video explores the complicated answer to that question, touching on why someone would find themselves in this situation, how to transition from one to another, and what you can (and should) expect to see in these scenarios. As an added bonus, we have Seth Charles, principal email deliverability and industry relations manager at Iterable, joining our team of 250ok experts, including Anthony Chiulli as host, Luke Martinez as a technical expert, and Julie Coval, director of partnerships, who knows how intricate ESP relationships and use cases can be.
(We’ve found key timestamps and transcribed this video below.)
Total Run Time: 14 minutes
00:47 – Common risks and challenges in migrating an email service provider (ESP)
3:12 – Seth Charles from Iterable provides advice for marketers preparing for a migration
4:28 – Recommendations for the request for proposal (RFP) process when assessing new ESPs
7:10 – Good and bad reasons for marketers to consider switching ESPs
9:10 – Importance of having a migration blueprint and defining team responsibilities
12:04 – Reasons why marketers should or shouldn’t leverage multiple ESPs
Listen and subscribe on your favorite platform:
Hi, everyone. My name is Anthony Chiulli, and I am here live in the Edge Studios in Indianapolis, and I’m joined by a fantastic cast. We are going to be talking about a really cool topic today in ESP migrations, and some of the tips and tricks marketers should understand if you’re going through one of those. Also, some learnings from all of us, being in this industry, about things that you should avoid and mitigate. So joining me, to my right, is Luke Martinez, and to my left, is Julie, and we also have dialing in remote today, as our guest, Seth Charles from Iterable. So everyone, thanks so much for joining me today, and let’s get right into it.
ESP migrations, I think, is something that often is overlooked and misunderstood for any marketer that’s gone through an ESP migration. There’s usually war stories and rolling of eyes about the experience, and rightfully so, right? It’s uprooting your entire marketing tech stack and moving it from one platform to the next, and it’s a daunting task. Luke, from your experience, just in general, what are some of the concerns or risks that come along with a migration from one platform to another?
Yeah, there’s a bunch. I mean, the first few that come to mind, and I’m sure we’ve all fought different battles doing migrations, but timing is a big one, choosing when you’re going to do it. Don’t do it in November and December, when a lot of people are sending the most email. Also, timing it in terms of having a little buffer with your previous ESP and your new one. So don’t have a hard cut-off date and expect to be completely transitioned on that date. You want some overlap. It’s unfortunate, you’ll probably pay two bills for a short period of time, but it’s well worth it because it never goes exactly how you think it’s going to go. You’ll have to dial it back, and then you’ll have to speed it up, and you want to have some redundancy, so timing the end of your previous relationship with the beginning of your new one that’s super important. Clearly, downtime is not an option, and warm-up, again, kind of a fragile thing. You can’t always send as much mail as you want the day you want to. So to avoid downtime, you’ve got to have some overlap and make sure that you’re timing that appropriately.
Julie, what are some other challenges that come to mind in ESP migrations?
Yeah, I think as you’re even considering migrating ESPs, you’ve probably already done a ton of research to decide which ESP makes the most sense for your organization. But as a part of that, it’s going to take even more effort to build out a full integration plan, and think about how much time that that’s actually going to take you. I know Luke mentioned having that ESP overlap time. Well, you really need to make sure you’re understanding how long you expect it to take you to migrate to know, okay, well, how long am I going to need my previous ESP? And also, on top of your original plan, understand that you probably want a remediation plan, right? Because you have unexpected occurrences come up, and you want to make sure, all right, how are we going to sidestep in case something happens that we don’t expect?
So Seth, I’m really interested in your perspective there at Iterable and overseeing deliverability. I’m sure that you’re getting new customers weekly and involved in a lot of these warm-ups and migrations and escalations. From your experience, what are some of the challenges from your seat that you’re seeing?
Yeah, I mean, something that we deal with a lot is just making sure that when customers are kind of scoping out what that might look like when they come over to us, know we’re always wanting to make sure that there’s parity between the feature set and segmentation logic and capabilities. Just to make sure that it matches or exceeds what they had before just to make sure that there’s no kind of downtime or piecing together segmentation logic that they might have to use. And then another big one that I’ve dealt with a lot, kind of, throughout my career is just making sure that senders are able to transfer important aspects of their database, like suppression tables. So that way they don’t come on to a new platform and all of a sudden we lack kind of historical suppression data because sending to that would, from a new infrastructure, would obviously create a ton of issues. So those are kind of the two main pillars that we focus on, at least to start.
Yeah, I would agree. And on that I think it’s an excellent response which leads me into our next question about—even backing up a little bit further, before a migration is occurring is the RFP process in which many customers and marketers go through and assessing which ESP is right for my business. What do you suggest marketers consider when going out for bid in RFP and shopping—or going out for bid for RFPs and shopping ESPs, what are some of the maybe tricks or tips that, from your seat, kind of seeing perhaps what can happen if these aren’t followed? Can you share any recommendations for a shopping-RFP process?
Yeah, definitely. And we go through this a ton, right? So the number one thing that we often chat with customers about is just making sure that they aren’t thinking kind of short term. If a customer is in a position where they’re at the end of a contract or something, and they need to consider moving ESP platforms in a hurry, it’s always important to kind of have the longer-term kind of goals in mind. Because you don’t want to make this significant of a change in all these things that can impact your business for a long time. You don’t want to make that just based near-term goals. You want to make sure and kind of position yourself to be able to scale kind of moving forward. Another huge one is going to be the reporting aspects so making sure that the reporting that they would expect to be able to get and the things that they rely on to help make decisions on their part, match up appropriately and then also help educate them about some of the custom reporting capabilities that they might have that could allow them to even get better over time, which is obviously part of that long-term plan. And then just in terms of the data flow itself is always going to be important. So making sure that your webhooks are in place, making sure that the integrations are happening, so they’re able to update their subscriber database and things like that with as minimal kind of growing pains as possible.
Yeah, I think that’s a great response, and I love when you talk about right sizing for the future and not for today because ESP migrations aren’t something that a brand or a company wants to do over and over again within a few years. I think it’s always kind of, buy above your current needs and wants and optimistically choose a provider that’s going to allow you to grow and continue to optimize your business, so that you’re not facing another migration because of a lack of features or functionalities in the near future.
Luke, talk about maybe some of the valid versus invalid reasons that brands should even consider switching platforms.
Yeah, there’s a lot of good reasons to switch, and there’s a lot of unfortunately common bad reasons to switch. And I think an example of a bad reason is magic IP addresses or better deliverability. We used to get this a lot when I was in the ESP space. Our competitors—and we were kind of guilty of it too, we would say we’re the best at deliverability, and the truth is the sender determines their destiny deliverability-wise. It’s your engagement. It’s your complaints. It’s your content. No ESP has magic IP addresses that just do a better job at sending email. So I always try to push back on this idea that switching an ESP is going to get you a better open rate. If it does, it’s because it’s likely something that you’re doing better or worse in one place or the other. I think that competitive pricing, that’s probably a good reason. Obviously, everyone has a budget. The actual sending of the emails has kind of become commoditized these days, so sometimes it is a pricing reason. But I think the best reasons to switch are about service level. Do you trust these people that are going to be your partner, helping you manage a really important part of your business, an important channel for you? So what’s their service level? Functionality, knowing some of the stuff that Seth listed off like can I do the segmentation I want my audience targeting? Does the campaign builder do the thing I want? Does it have automation workflows? Maybe you don’t even need those. So some of the bells and whistles that might be available, are you going to actually put those to use? So to me, the good reasons are things like functionality, to some extent, price, the service level, the relationship level between you and the ESP. Bad reasons are, “I’m leaving because my open rate’s low.” It’s not your ESP. Your open rate being low is you got to do some soul searching and work with their services teams and leverage a tool like 250ok or something else to help you figure out what’s causing it because it’s almost never your ESPs fault.
Almost never. I would agree. Julie, what about just the mindset and preparation, right? So go through the RFP process, you choose valid reasons to make the switch, and now you’re ready to undergo the planning and preparation for migrating. What are some of the recommendations for that aspect?
Yeah, I think, like you said, you’ve already gone through a lot of processes. You’re probably getting excited because you’ve decided which ESP makes the most sense for you, but I would say you’re at one of the most vital steps in the process, of the fact that you really need to make sure you have a clear migration plan. And the reason for that is as you’re migrating to a new ESP, it takes a village, right? It’s going to be a cross-departmental effort from not only your marketing team, but your IT team, your product team, really anyone who’s touching the ESP or even touching your brand, quite honestly. So you need to make sure that everyone is not only aligned on what their role is, and what tasks they’re doing, but what is the timeline and what are the expected milestones. You need to have set time periods of when you want to get things done because if not, you’re just going to be kind of going with the flow and who knows what might happen. So as a part of that, I would say what I touched on a little bit earlier is making sure you have that remediation plan in place. I mean, hopefully, you vetted your initial migration plan so that everyone is on the same page, but, as I said, you might have some type of unexpected occurrences happen. You need to make sure that you can go in a different direction in the case that something isn’t expected. Something that you wanted to happen didn’t happen on time, and you need to just make sure you have that kind of plan B. The other two pieces that I think are important to be aware of is, what data can I transfer from my old ESP to my new ESP, making sure that you have that within your plan to not only get that data over that you need, but also if there’s certain data that isn’t easily transferable, you have a way to capture that and maintain it, so that you can move forward, as a company, with that data. And then finally, what we’d said a couple times earlier, of making sure you have time with both of those ESPs overlapping, and I think that’s really huge. While it is a bummer that you might have to pay a little extra for a few months, it’s really important to have your previous ESP intact so that you can maintain sending messages out of it, while you can’t send those messages that you want out of your new ESP.
Yeah, I think the importance of preparation and plan, what I call “blueprint,” for a successful migration is paramount. And it’s a huge undertaking, and there’s numerous departments involved, like you said, and having, certainly, a strong PM to oversee that migration, that blueprint, and have, like you mentioned, definitive timelines and steps and statuses to keep everyone on track is crucial. Luke and Seth, I wanted to get your opinion, and start with you, Luke, about this notion that, what I don’t think many people understand is, a lot of senders actually use multiple sending platforms or ESPs throughout their business. Maybe it’s for different mail streams, or different departments. Is this a common trend, and what are the reasons to consider perhaps using multiple ESPs?
Yeah, it’s similar to the last question I took. I think there’s good reasons and bad reasons to do the multiple ESP thing. Start with the good reasons. If you have transactional mail streams and you have a solid email service provider that, more or less, behaves as a pipe, and they’re just triggering messages through them. It’s not about marketing automation or beautiful design templates and stuff like that. There’s certain ESPs where that’s their wheelhouse, transactional email. They have robust APIs. They can send email as fast or as slow as you need to, etc., and then you might have some other more marketing- and kind of nurture-focused things that do require automation and drip campaigns and all this stuff. And you want to use their template editor. Very few do all of these things really well, and I think there is a legitimate case to be made that some are really good at transactional and some not so good at the marketing and some vice versa, so, good reason to use two ESPs. Another one is I think in the event of outages, most of these ESPs these days have four-nines of up-time, they’re extremely reliable. Everyone has outages, though, and if you’re a huge sender and you’re constantly sending email all day, and your ESP goes down, you’ve got to have the—you got to be able to flip a switch and go somewhere else. And if that’s the case, you need it to be warm. You can’t just go zipping up a new account somewhere else and send 10 million messages on day one. But if you’ve always sent a little bit through each, you’ve got some hot IPs that are familiar with your domain, you’ve been using that infrastructure for a while. In the event that you’ve got to switch over, you can do that. And if you’re not using two, I mean, that’s the whole topic is switching ESPs. It’s hard to do, so have one ready, so if you need to, you can switch it over.
Seth, do you have anything to add on to that question?
Yeah, I mean, Luke definitely hit a number of those, so when a particular ESP is good at something, and you want to kind of utilize different pipes to do things that they specialize in, that does make a lot of sense. And then, when it comes to—even when a decision is made to completely transfer, it’s not always a bad idea to kind of hedge your bets a little bit because like Julie has mentioned a bunch of times, warm-ups rarely go perfectly the first time. So in the event that some unintended segment was included as part of warm-up and it really kind of sets you back days or weeks even, it’s always important to kind of have that safety net to where you can kind of default back to in the event that something happens. And you can utilize that infrastructure again because like Luke mentioned obviously it’s already warm. The infrastructure that you send with is already established, and that’s something that filtering companies and mailbox providers certainly look for.
Yeah, I would agree. I think this has been valuable information, guys. And, Seth, thanks so much for dialing in remote. I wanted to thank our audience for tuning in. I certainly think that ESP migrations is something that can be daunting, but if you have the right preparation and resources and intel to ask qualified questions during the RFP process as well as pre and post-migration, it can actually be a successful project. So thanks for watching, everyone.