What Actually Makes a Good Email with Naomi West, Senior Product Marketing Manager at Customer.io | Ep. 395

0:00Today's episode is brought to you by customer.io. You have customer data, but turning it into personalized marketing at scale? That's the hard part. Customer.io connects the dots. Feed it your data, start sending messages that actually feel personal. Learn more at customer.io slash TMM. Welcome to the Marketing Millennials, the no BS marketing podcast. I'm Daniel Murray and join me for unfiltered conversations with the brains behind marketing's coolest companies. The one request I tell our guests, stories or it didn't happen. Get ready to turn the f*** up. We are back with another episode of the Marketing Millennials Podcast.

0:51Today, I am with Naomi West, Senior Product Marketing Manager at Customer.io. We're going to talk all things email, which is a topic that everybody talks about. But I feel like this goes back and forth. What's good email? What's bad email? What should you do in email? And I think Naomi has some myths to bust in this podcast. But Naomi, how did you get into marketing and where did your love for email come from? Yeah, well, actually, I got rejected from my university's marketing school probably three times before I decided to just do my degree in political science, graduate, went into working in recruiting.

1:31And then I was so bored at my recruiting job that I took on a side hustle through AngelList doing email marketing. And I, at the time, had no idea really what I was doing. I was emailing individuals directly, trying to get them to sign up to an app that we were building. And then I realized that you could automate it. And I thought I discovered like the next best piece of technology before shortly learning that that was email marketing and email automation. And that was back in 2015. I ended up quitting the recruiting job where I was working kind of as a receptionist or onboarding coordinator, joining the tech startup full-time, and I fell into this weird world where I just loved email marketing, consent-based email marketing.

2:19So when I'm talking about emailing today, I'm not referring to like cold emails or salespeople emailing people out of the blue, which is cold email or spam or whatever we want to call it. I'm kind of talking through examples where you sign up to an app or a website to get a 10 % off discount code, uh, where you get kind of a, an email checklist of things you need to do before being able to use the tool. And that's really my, my bread and butter and I haven't fallen out of it. So, yeah. The angel list story is pretty cool. I think, and I love, I love a good, um, rejection to like climbs. Those are the best.

2:58That's what motivates me. Sometimes all the people who said I couldn't do something. And I love that. I love that. And it was all because my calculus grades weren't good enough. And here I am like doing marketing because I wanted to do marketing and I'm not using calculus. I'm using AI. So take that. I wanted to ask you. So when someone is judging an email and they say, this is a great email, what are they usually getting wrong when they say this is a great email? They're looking at it without the context of how someone's receiving that email. So for example, what's the trigger behind that email?

3:38Is someone receiving it? Is it a newsletter that goes out on Tuesdays at 9am? Or is it something that's event triggered? So has someone done something in an application or a platform or have they purchased something and then immediately received that email. I think another thing that people often forget to realize is, hey, this email is beautifully designed, but it's actually all images just sliced together. And if that email goes to a spam folder, or if that email goes to someone on a certain version of Outlook, none of those images are actually going to load in the email. So you've just sent them a blank email.

4:14They have no idea what you're saying. They have no visibility into the context of what you're trying to communicate because you've only used images. You've tried to mimic web design, which is email's kind of like cousin of sorts. But email is so limited in design capabilities that you look at something thinking it kind of seems like web design, but email, you can't have that kind of one-to-one parity. It's not apples to apples. So context and kind of the actual design and application of the email is something that people often overlook. I want to go deeper into, you talked a little bit about web design and people try to mimic an email with web design.

4:57So what are some of the web design trends that fall apart inside of an email? Custom fonts. If you think about it, there's only like six to eight. Someone can correct me on the exact number. but there's really only like six to eight stable email fonts that are applicable across all email clients. Email clients being Hotmail, Yahoo, Outlook, Gmail, Apple Mail on iPhone, mail.ru. There's so many email clients out there and they all support different things. So for example, if I'm building a beautiful website and I'm using a lovely Google font, I would think, oh, oh, Gmail is owned by Google. Surely that same font would work if I send this email to Gmail.

5:45Wrong. Gmail won't even load its own fonts. It'll fall back to something like Arial or Helvetica. And so fonts is a big one. And it's super disappointing because a lot of individuals' brand identities are tied to the fonts that they use as their typeface. And when they don't end up loading in the email world, you're like, what? this is so weird. Like I hate this Helvetica or Georgia or whatever these like safe fonts are. So that's a big one. Certain things like GIFs don't even work on certain versions of Outlook. They will only load the very first frame of your GIF. So you try and make very simple things like a GIF work in an email.

6:30And the reality is, is it's not universally supported when it reaches and inbox. Another thing that doesn't work great is things like rounded corners. In some versions of Outlook, that's not supported unless you use something called VML. You can't play video in a majority of email clients. So people will often say, how do I insert this video into an email? And I have to say, well, use a GIF, put a faux play button on it, and then link out to wherever you've supported that email, whether it's YouTube or Wistia or Vimeo or one of these millions of platforms that exist these days. So there's a million things that are like super limited when it comes to email marketing.

7:13And it's really sad at the end of the day. And there's a lot of education that needs to take place because I think you go into email marketing kind of with these rose colored glasses on thinking that you can create this beautiful design piece. and you spend a lot of time attempting to create something that looks great in an email builder or in your marketing automation platform only at send time to realize that a lot of those things aren't supported. I also think that there's also a overly emphasis on a well-designed email and an under-emphasis of what actually is in that email and it doesn't matter what matters.

7:49because like what really matters in emails is relevancy and like to the audience and if they are they getting something useful out of it they don't really care how good designed your email is i mean i don't go terribly designed but they you're not gonna they don't care if you try to make it super pretty they're trying to get something really fast out of this that is useful for them and then they're going to the next email they're not trying to yeah no i i think the nicely designed email so like 10 15 years ago when you got into marketing just like i need to do that now it's just like yeah let's get the content quickly um i want that what are you trying to say in that email yeah i had i had like a eureka moment i guess four ish years ago where i came into this small startup called parcel uh which was an html coding platform for email specifically.

8:45And I am not a graphic designer by any means. Like I'm an email marketer. I can talk strategy all day. I can talk a little bit of code, but I am not a visionary. I've got no visions happening up in my head and I cannot execute a graphic design. So I always default to either an existing template, email template that someone has designed for me, or I default to a text-based email, something simple that I can get out the door same day. And so for kind of the first year of me doing marketing with Parcel, we would send text-based emails. And I had always this large dream of, hey, we're, you know, an HTML coding platform.

9:25We should be able to walk the walk and talk the talk, talk the talk, walk the walk and have like really cool emails. And finally, someone joined our team. His name is Mark Robbins, and he's just a really talented email developer that I knew could bring some ideas to life and help me execute on these design ideas. So we ran an A-B test where we had variant A, which was the text-based automations going out, kind of like the control, and then variant B, which was these really cool emails that had dynamic content and gradient backgrounds and marquee text that would move in the clients that supported it.

10:05And the results came back and like the text-based variant won for both click-through rate as well as like conversion rate. And my hypothesis of, wow, we should invest in these really nice emails was completely squashed. And that was the moment where I was like, okay, maybe keeping things simple is the way to go. And like, thanks, Mark. Sorry, Mark, that your designs couldn't actually be super beneficial to us. But text-based emails I can get out same day. And people love them more, apparently, for the specific business case. It's probably also email-dependent-based, too. Because I feel like if you're sending out a newsletter and you have like a warmed-up audience that loves your email, you could start adding a little bit of flair in that email and they still were like, it still will be received well because people are fans of the content you're saying.

10:58But if you're sending like those transactional emails or you're sending those automated emails, it's always better to lean on under-designed and easy to read than over-designed and hard. And some images, like I look through my inbox every day for one inspiration, but I like to read my emails. And I'll open an email and it'll just be a gigantic image. There's no text in sight. I don't know what the main call to action is. The subject line doesn't tell me anything. And that to me, I just delete it. Because I'm like, am I going to have to scroll for days to get what the main message is? Like beautiful image.

11:35But it loaded like a fax machine because maybe my Wi-Fi is slow. But then I also don't understand the content. So sometimes like keeping it simple saves graphic designer's time. It saves you as a marketer time and it results in way higher engagement. What's like a fair way to fair way to AB test that is it just doing one element different? Is it doing the whole email different? Because I'm wondering what is a fair AB test when you're trying to test text-based email or something with a header or something which works better? What's a fair test that you can run in a day? Yeah, it's tough because even like the removal of images from an email creates like an apples to oranges test where you're not necessarily comparing the same thing in the same message.

12:29The way that I measured it was in, I think it was a three email automation over the course of a seven day period. It was our welcome series where I had variant A, the three emails were text based and then variant B were the design ones. I looked at the overall kind of click-through rate across the three emails on both sides, as well as the overall conversion rate on both sides to try and understand, like, does the variant A that only contains text-based pieces result in overall higher conversion than variant B? I also wait a period of time before I gather those results and see if they're statistically significant.

13:06In this specific example, I waited a full quarter, so three months to see how engagement was over that period, removing kind of like any fluke timings or like periods, weeks, where maybe people were just oddly more engaged with text-based emails. I wanted like a really healthy sample size to go in and remove elements of seasonality, make sure I had good volumes of individuals on both variants to pull results from. And that's what I would recommend marketers do is kind of automate your tests, put them in an automation, an event triggered workflow, have those emails go out. And then after a longer period of time, go and look at how individuals are reacting to them or engaging with them.

13:54If you're going to test it over like a one email, like a newsletter, I usually like to repeat the test multiple times. so if I send a weekly newsletter week one I would test that out week two test it again week three test it again just to see if the results are consistent towards being statistically significant in one direction or the other I think it's a that's a good way to because for marketers this thing just to have a fair test also I think it's a good way because one thing that a lot of email marketers run into and executed marketers run into is um i want to send a text-based email because it converts better but the brand team saying that's off out of brand like you said fonts too and colors and all this stuff so how do you combat that i one way to combat it is do that fair test over seven days and see what have you done in the past if someone if that's ever happened to you that yeah of course of course there's always some someone saying we can't do this it's against our brand guidelines and in some scenarios you're unfortunately just stuck in that world i see it in a lot of more like bureaucratic or larger enterprise organizations where unfortunately the marketer lacks a lot of create creative freedom um i am a pusher i'm a pusher internally.

15:21When I like to test things out, I push, push, push until someone finally says absolutely not. But I think it's going in with like the value first. So for example, if you wanted to test a text-based email saying, hey, here's an idea I have, here's how the impact would result in change within our org. It would allow me to decrease my email design time from seven days to one day. The QA time against our brand team or our design team would be shortened by X amount of time. Here's the results I'm hoping for, a higher click-through rate, a higher conversion rate. And here's what that ends up benefiting our overall organization with, if you can respond with numbers in any fashion, whether it's directly tied to monetary value or just saying, hey, if we see X amount of more conversion, that means X for our business.

16:13ideally a management would see that and be like okay it's worth testing um unfortunately there's always scenarios where people are like no our brand standards are elite and we can't risk someone seeing a text-based email from us like that would ruin our reputation get a new job move on time to find a new place where you can be a little bit more creative but yeah i think it's just putting the value first and hoping that someone has their head on right where they see that and they're like okay let's test it one thing too i wanted to ask you is i know you've worked with a lot of different esps in the past um so like how much does platform matter versus strategy when it comes to esp is like do you need the perfect platform or i mean for some things probably you need some automations that some can do but what is the most important when you're weighing out those two things.

17:13You're sitting on a gold mine of customer data, and it's doing nothing for you. That's not a data problem. That's a tools problem. Customer.io connects your data to every channel you already use. SMS, email, push, in-app. Then it uses AI to build workflows that used to take weeks. Every message feels like it's written for that customer, not blasted at them. 8 ,000 plus brands already made the switch. Learn more at customer.io slash TMM. Yeah, there are a million ways to get the same thing done in all of these platforms. A lot of teams will lean on external kind of like vitamin tools that help them create great emails because their in-product solution sucks.

18:04A lot of email editors inside of marketing automation platforms are the bare minimum. I think because the product teams are focusing on building out better data orchestration and segmentation and people can get what they need done as is. That being said, basic text-based email can be accomplished in every single platform. No issue there. And then kind of a well-designed email, in my opinion, can also be accomplished in like 80 % of platforms. Where people start to get a little bit itchy is when they want to customize an email design based on the email clients that will be receiving it. So once you get past this generic designed email and start leaning into, okay, what does Apple Mail support?

18:55They support interactivity, maybe they support marquees, maybe they support gradient backgrounds, email designers will potentially lean into that and want to customize it. And so there's external tools for that. Some marketing automation platforms like Customer I.O. and Alternatives support those things in the product themselves. So I think if you're wanting to test out the text-based email piece, you should be able to do it on any single platform. it's when you want to kind of lean into the the nitty-gritty uh kind of targeting based on client that's where you can kind of start to do that in either external tools or poke your marketing automation platform to see if they offer those capabilities but i've been in a lot of platforms and the platforms that i like there's certainly a list of platforms that i avoid um the platforms that I like to use and the ones that I work with support the basic email design functionality.

19:52It's really the event triggering and the dynamic conditions, segmentation, their reporting that differs on a platform to platform basis. Let's go into some of the things that lead up to an email that are important that people think about. So what are, when you're designing an email campaign or a live cycle campaign where abandoned card flow, whatever you abandon form flow, whatever you do, what is your like one, two, three step? And how are you planning out from, I'm going to do this to I'm going to execute it internally. Cause there's a lot of like moving parts sometimes too. If you, if you're changing fields, if you're, um, marketing things, so what is like your end to end look like for a campaign?

20:41Yeah, there are a lot of moving pieces. Um, so one of my favorite campaign to, one of my favorite campaigns to work on or automations to build is like very specific education, event triggered emails. So for example, CustomerIo recently launched our MCP server, which allows individuals to basically connect Claude, ChatGPT, Cursor directly to their CustomerIo workspace. So I can go into Claude and I can be like, tell me about Daniel at Marketing Millennials. and it will pull your user profile data. I will ask it, what are some things Daniel's done inside of CustomerIO recently? And I can fetch all of that data and Claude can summarize it for me, which is great.

21:27In that specific scenario, my job as a product marketer at CustomerIO is to explore how people are using the MCP server. I want to know their feedback on it. And I also want to educate them on how to use the server better. With that, I start with defining what I want the trigger to be. In this scenario, it's enabling the MCP server inside of their customer account. When someone does that, they trigger an automation. And the first touch point is kind of congratulating them on enabling the server and reaching out, letting them know that I'm here for feedback, I'm here for questions. My stance on email marketing is it has to be a two-way street.

22:14I have gathered so much valuable feedback from users over the years when they reply to my emails. So that first touch point, I want to welcome them, congratulate them, give them a couple best practice resources for where they can learn how to use the server effectively, and then offer them a way to contact me with feedback. So I kind of strategize around that. I mull around the copy. My main goal is attempting to get them to use the server for the very first time. So that's going to be my primary call to action, driving them to connect Claude or Cursor or ChatGPT. The follow-ups, I want to know how their experience was.

22:54And so let's say after a certain period of time, three days, five days, seven days, you can play around with kind of the day delay to see if one results in more feedback. feedback, the touchpoint after that certain delay is going to be trying to gather that feedback. And so my copy in that email is how'd the experience go? Was there anything you wished you could do? Do I need imagery in those emails? Potentially, I might throw in a GIF showing how to connect the tools, just a really simple visualization of how to use the product. I might throw in a couple example prompts, giving individuals a starting place.

23:34And then from there, see what kind of results I get back. So it's a first pass. And that's kind of how I approach email building. I put out a general feeler of what I want to accomplish as a marketer, what I want in return. And from there, either it could be like me working as an IC to develop and build those emails myself. or I could be working with my product manager, the rest of the marketing team to kind of shape the brand voice and tone if it needs to be more brandy. And then I make sure those feedback loops are in place to translate that feedback back to Slack channels so others can have visibility.

24:13There is a QA process with the email. So once I've built the emails, I submit them to our brand team for review. They do same day review because we have a pretty well-oiled email design system and then I launch it. I think some people forget when they're thinking of the grand scheme of emails. I think you forget about the ones that you could do for individual product managers or individual reps that could help them save a lot of time. They think about the card flow. Also, I was going to add, too, tools sometimes matters if you're trying to do product triggers. because it's sometimes hard. Yes.

24:53Sometimes hard when you don't have a tool that's easy to connect to your custom product, then you're having to have engineering build, something to go into. And engineers ignore us. They ignore marketers. They're like, I'll be like, hey, can I, do you guys have an event for this I could use? And they're like, la, la, la, la, la. Or six months later, it's, it's live. Six months later. Do you need this? Yeah. No, that's why I'm a pusher. If I don't get same day replies from an engineer, I'm in there. calling them, pinging them, pulling them into like a Slack little video. They don't like video. Engineers don't want to be bugged.

25:28So if you're a marketer out there and engineering isn't helping you, you just have to push all day, every day. Even with IT, I remember there was like certain guardrails when I was in marketing ops that they weren't allowed certain fields inside. And we needed just one more field. And it was the end of the world to add a field that I needed on the record so I can do an automation. Yeah, it's weird. I'm fighting because they have their eternal rule and I'm just, this is not going to affect anything and it's going to help us increase revenue. Like, please. Yeah, it's so odd. And it's such like a universal, like if you're a marketer and you haven't encountered a tricky engineer or like tricky IT, it's coming soon for you.

26:16It's like a rite of passage. Everyone has to encounter this certain type of personality. that like has to abide by the rules. No, you can't have that data. You can have it in one year from now. I don't know what, what's up with that. I also want to ask you, I mean, let's say, I mean, you probably are, since you're in email marketing, you're judging a lot of emails that come in your inbox, other people are sending you, but I want to ask you if like someone came in you and one of those emails fell in your inbox and you said, and they said like, tell me, they said, tell me if this email is good. What are like the three things that you're looking for to check if this is a good email, if someone is sending it?

26:56Cause you're judging them. I do the same thing with marketing. Yeah. If I had someone email me being like, is this a good email? And I'm the recipient, let's say it's sent to me. And then the brand is like, was that a good email? the relevancy is is top of mind uh how familiar am i with the brand as is did i just sign up to them and it's part of their welcome flow is this a brand that i've you know been acquainted with for a year from now or a year at least a year relevancy top of mind i feel like there's a lot of brands out there that forget to create these like predictable touch points and just ghost subscribers for six months and then they're like, Oh, I need to send a newsletter.

27:40They send the newsletter. I receive an email and I'm like, what is this brand? I've never signed up to you. Spam. I'm a huge, I'm huge on the spam. Spam, spam, spam. I don't even unsubscribe. I'm like, I don't know. There are some email providers that auto subscribe you to, which is, that's a horrible practice where you get auto subscribed to a newsletter you never subscribe to. I hate. yes unfortunately i think that's like legal in the states maybe so i apologize for that down there in canada you can't do it there has to be like either implied consent where i've like purchased from the brand previously or like signed up for like white paper or downloadable you can't just email me like that it's illegal i'll fight them i don't like that and i think that's also illegal like in europe and stuff so i don't get it relevancy just collect consent from your subscribers and you'll have a good time.

28:32Make sure you're communicating with them like fairly frequently. Monthly is fine. Weekly could be too much. Depends on what you're saying and what you're trying to sell and what you're offering is. So relevancy and then the content itself. So what is it that you're trying to get me to accomplish? Is it a newsletter that I've opted into? In that case, like, is this content relevant to what you said you were going to deliver to me? and am I opening the email and reading through it and not unsubscribing? Is it a newsletter that is trying to get me to click on blog posts and read further? In that case, are people clicking through?

29:12And then the design, I think back to the design once again, is the content legible? People who send out emails with size 12 font size, I can't read that. If I need to make my face wrinkle to read an email on my mobile device, it's a bad email. So basic design and accessibility best practices. That's the trifecta, really. A good email is one that's relevant, has good content, and is legible and accessible. I want to also go into, I know you, in some of your workflows, you talked about that you have cloud and all the AI tools that are connected to your platform. But what do you think? Because, I mean, email's not going away for a long time, but what do you, what do you think is going to separate like a good email market from a great email market in the next three years?

30:03Um, what are some skills that they should double down on to make sure that they are going to be competing? I mean, obviously one of them is like knowing how to do stuff with AI, but like maybe go deeper into how, what are you thinking about that? Yeah. Um, the best types of email marketers that I see today and maybe just like marketers in general are ones that are open to testing new ideas, not people that just stay, you know, working the same wheel over and over again. It's people that say, hey, it's been, you know, six months since we've tested this last, should we test it again? What do we want to find out, you know, from our audience this time?