Unlock this video + Thousands more !

Start your free 30 day trial today!

Join Free Sign in

Using Customer Data to Craft Email Personalization at Scale

Video Summary

Using Customer Data to Craft Email Personalization at Scale Marketing is all about the data these days. But gathering, analyzing, interpreting, and using data to create relevance and timeliness in email campaigns can be easier said than done. This session will explore the 6 steps to take to turn data into a useful tool for creating marketing, including the latest methodology and technology you need to gather customers insights and transform that data into personalized emails at scale. Buckle up and learn how to: -- Acquire data so that it's ready to use in marketing and sales -- Use data to determine the content of your email marketing -- Use data to determine the timing of your email marketing -- Use data to determine the impact of your email marketing


Each day, I get to work with clients doing big-idea-driven, multi-channel campaigns to weave in the best of what data-driven strategy can do to build and boost customer relationships. I get genuinely excited to share best practices, test new ideas, and see awesome results.

Starting with my role as Digital Marketing Evangelist at emfluence, I've been preaching the good word of great email marketing in both online and offline communities for years. I lead email marketing and customer-driven marketing strategy for Barkley clients and get to celebrate trends and highlight smart stuff the public, speaking for audiences from Las Vegas to New Orleans to New York, and in March of 2013 on the line-up at South by Southwest Interactive in Austin.

A May 2005 graduate of the Mizzou Journalism School: my first foray into content production. I started my career in media sales for a Top 40 radio station in Kansas City. I was recruited away to lead season ticket marketing for Theater League. There, I learned direct mail and email marketing and brushed up my stage skills.

In October of 2008, after evangelizing the product for a year as a client, emfluence asked me to join their team to launch an in house marketing department to oversee their external relations and brand management across all channels.

For three years as Marketing Manager, I managed partnerships and opportunities for emfluence with other Kansas City marketing communities and businesses as well as a comprehensive and cohesive digital and email marketing campaign, promoting the emfluence Marketing Platform, and our strategic services. In 2013, I finally got to put the truth on my business card: Digital Marketing Evangelist!

In 2016, I accepted the exciting role of Director of Data-Driven Marketing at Barkley, working as lead CRM strategist for fortune 1000 companies like Dairy Queen and Spirit Airlines, soaking up all the smarts of my fellow partners and helping integrate data.

Specialties: email marketing strategy and tactical execution, integrated emerging media campaign planning, social media strategy and planning, networking and promotion, email marketing design and copywriting, database analytics, results tracking, research and reporting, event management including trade show exhibition.

Conversational language experience in French.

What you will learn?

  • Understanding how to send an email series
  • Knowing the techniques in using customer data to craft emails
  • Techniques to reaching out to your audience at the right time


Okay. So, in this case we’re saying the item fits the audience without being over about the content.  How about content personalization? If you have no information about your subscribers, if you literally got an email address and nothing else. I would argue that you probably have something else. But even if that’s all you have, we can use the engagement somebody has with your early emails to determine the rest of the series. In this case this is that same welcome series right, and on the first email, we have a home mock-up for a Meritage Home including all of the benefits of buying a green home. I want to clean my house last. I don’t want pets in my home. I want to insulate the noise from my neighbors. Depending on what you click on for more information on that model, the rest of the series then extends that messaging through the end of the series.

[00:00:59] So if I said what can I afford? You can afford the dream was the next email that I got.  It’s a little bit more over right like that one. I can guess how you knew that about me. That’s all right, because it’s relevant and I think especially again not a GDPR expert but in the United States mostly consumers have given up just a little bit of their privacy in exchange for relevance, in exchange for convenience for crying out loud. So, all right, all of you can probably do some of those right? A little bit of a stretch. Let’s get into 2.0 which is segmentation and actually segmentation doesn’t have to be a stretch either. It’s just sort of the next iteration of how we can possibly use data.

[00:01:40] In this case, easy, easy, right? Two versions of an email one with the slides from a conference that they went to as the first thing that they see, and the rest of its the newsletter. On the right the example, that’s just the newsletter. Why would we do that? Why wouldn’t we just send the slides in a separate email?

[00:01:58] Yeah.

[00:01:59] Audience response [inaudible]

[00:02:07] Excellent point. I only want to send the slides to the people that were at the conference. Everybody else has to fend for themselves. And the number one highest open rate every year. Oh, did you know?

[00:02:20] That’s a good one. I didn’t actually think about this is a pretty engaged audience. We’re sort of cheating with a group of people that came to an event right but. It’s possible that my total frequency shouldn’t go over a certain number because then I start hitting unsubscribe rates I don’t like, but truly the highest open rate we see every single year is on that slides email. Like when I sent you guys slides from this presentation. I’m going to send it through Gmail. But if I sent it through an email platform, I bet that I could see that my open rate is stellar because you’ve asked for the content you went to the conference.

[00:02:51] You’re obviously engaged with that content. While you’re there eat your vegetables. Right, I’m going to sneak in a couple other things in that email because I know that that email’s already going to get a high open rate. Now be careful with this, you have to test and see if that content detracts from the effectiveness of the content that they really signed up for.

[00:03:09] On the right-hand side, this is a save the date for a, actually this is coming up, fundraiser in Kansas City and notice that the difference here is, you came to this event last year. I know you were there. Come again this year. You’ve got a two-week head start to buy tickets because tickets always sell out.

[00:03:27] So simple, simple, segmentation in that people who came last year love to be the exclusive recipients of an invite for this year. It’s only two weeks earlier nothing fancy, but the way that we message that the way that we sort of sell that or offer that can make a big difference.

[00:03:44] All right. So back to Hallmark Baby. By the way, I do appreciate, did anybody go to John’s presentation, John Halls presentation that was in this room? I mean, of course John actually shows a picture of his wife and kids and I’m like, you cheater. Easiest way to get an audience on your side, show babies. I don’t have my own, so, I show Hallmark. In this case, this is to prove a little bit of a point. Again, Hallmark has the information about who you are, how old you are. They didn’t know who their audience was going to be, when they went from greeting cards to baby clothing, to expensive kind of baby clothing. They’re like, you know, it’ll probably be hip moms, Millennial moms, first time moms, West Coast, East Coast, probably, right?

[00:04:32] Okay, so we ran segmentation on their list without changing anything like this was just to figure out who is doing what with the emails, how often people are opening among different segments. And when you sort by our best performing emails seven of the top 10 were to women 55 plus. That’s not first-time moms. First time grandmothers love Hallmark baby. It’s clothing as soft as a baby’s hug. I mean, they love Hallmark baby. We can send twice as often and not see diminishing returns.

[00:05:07] Excellent. They want it. We got it. We listen we give it to them.  Sweet. I love my job. My job is cheating. All I do is look at the data and then do what it says. This is a little bit more complicated. This is a retail example. And on this calendar, you can see the three types of campaigns that we do, and then the cadence of those campaigns on a single calendar. On the right-hand side, you can see how many emails is our full database getting, versus how many emails are those who are opening and clicking pretty regularly getting. It’s different number. We’re only sending to the full database once a week, but those people who have opened or clicked something in the last three months, they get five emails a week and they open them right.

[00:05:48] If we can tell that people are still engaging with that content, we’re going to keep sending it. If somebody hasn’t opened something in three months, they probably don’t check that email address anymore. You can accidentally open something in three months or they’ve sent us to a folder where they’re not viewing us. So, we’d rather not overwhelm that audience we are not sending the right type of engaging content yet. So, we save our very best email of the week and that’s what goes to the unengaged audience.

[00:06:16] All right. So, I have a lot of food clients. I should just apologize for that right now. Strawberry Hill Pub and Pizza uses old-school RFM segmentation. We remember RFM from like marketing 202 textbook.  RFM segmentation recency, frequency, and monetary meaning those people who bought more recently, those people who bought a bigger order size, or those people who bought more times over the course of the year. For their different audiences, they get different scores and for those people who are there “A” buyers, like thousand dollars spent or ordered five times in a year. Pub and Pizza by the way, which is delicious is a sweet baked bread type product, and you either know about it or you don’t. So, those people who buy five times a year. Those are our people.

[00:07:08] They get a free shipping offer on Black Friday regardless of order size. The people who have tried it but not yet tried a second time or maybe ordered for a friend or received it as a gift but haven’t come back, those people get a special offer if they order a certain amount or free with purchase, right you buy one for somebody else, you get one.

[00:07:27] Okay, the secret here is the group “B” in the middle, they get a free shipping with a certain level of purchase, $100 and you get $30 off or free shipping or whatever.  Secret for us is again pull back the curtain, the average order value for our “A” audience is $137. So, we don’t want to tell them to spend a hundred because then they’ll spend a hundred. So instead we let them buy as much as they want and offer them free shipping.

[00:07:54] All right, segmentation got it. Segmentation really rolls into a lot of things like targeting as well. So, we’re going to roll that into some 3.0 stuff. Automation and I already kind of told you if you’re in the last session, you know that I already love automation but automation meaning, data informs the timing of our emails.

[00:08:11] First up, there it is. Stick Wood, the repurpose or reclaimed wood that you can do an accent wall, things like that. They’re abandoned cart. We don’t want to see you go right abandoned cart all this time. Okay, it’s as old as the.com bust probably but they’re abandoned cart email performs very well for them. Their welcome email drives three times the revenue of their abandoned cart email.

[00:08:42] So you don’t have to have an abandoned cart, you know how to have an e-commerce site. This to me is that reminder that you should never forget the power of your high open rates. We talked about it a little bit with the KCDMA email, but frankly your welcome email is likely to be the highest open rate you will ever get from somebody. How are you using that space?  As a Content marketer, what’s the next best ask? You said yes to date number one, I gave you a welcome email. What’s the ask? What’s the next date?

[00:09:11] Birthday emails again, a little bit more of a promotional example, although John Hall would say probably should know the birthdays of the people on your resources list, this is coupon driven. Water.org is a little bit different. Theirs is as part of their welcome series. They’re asking you to add your birthday to their data set their record, right? We don’t take birthdays for granted. We know that not everybody has the luxury of knowing that they’re going to be at their next birthday. So, we’d like to send you a little something special to remind us to think of you on your birthday, but we need to know what it is. This is progressive capture, right they signed up for email.

[00:09:47] They probably only gave us their email address. How do we send them more stuff? We ask for more information. So, this is when the welcome email becomes a welcome series and we start asking for a little bit more information in each email.  This one is actually on the customer side. So now we’re talking about you didn’t just sign up for email. You are a new client, you are a new provider, in the case of legal insurance with our ARAG. They have Network Attorney signing on all year long and for a network attorney, it’s kind of like being a doctor with medical insurance. Okay, great. What next? I’m part of the network. Like what do I do? So now you’re going to hear from people in the network that are interested and here are the benefits or reminder of the benefits of being part of the network.

[00:10:34] You’ll see new customers that you wouldn’t have seen before or the example from Influence on the right-hand side. You’re a new user of our email marketing platform. Here are the three things I can just about guarantee that you’re going to ask me in the next 90 days, right? I will give you the answer to your frequently asked questions before you even frequently ask them.

[00:10:56] Super super useful and again, all of this can be automated. So hopefully our support team can answer this sort of weird questions instead of the frequently asked ones.  Okay, step it up a little what could it really look like? This is a little bit more like a decision tree. This is the Spirit flyer Journey.

[00:11:13] Once you’ve booked in the top left you get your sort of automated confirmation email. You make any changes you get a different automated confirmation email. If you have not flown with Spirit in the last year, you get an education series reminding you that bags cost extra or reminding you of our unbundled fare tends to be 30% less expensive than the next least expensive airline.

[00:11:36] So even if you have to pay for your bags or pay for your boarding pass it still on average 30% less it so we had to do those education pieces. If just in case you’re going to be mad just in case you’re going to get to the airport and forget that that’s how it works right. Starting 72 days out, we start by the way, you don’t have your bags you saw the 24-hour email in the last session check your bags and also print your boarding pass and do all these other things that save time and then here’s my travel.

[00:12:05] You’ve arrived or sorry that didn’t go as planned. If we have a flight cancellation, they get a very different message than a congrats you’ve landed in Fort Lauderdale. That’s probably a good day. And this is probably not a good day. So then after flight assuming that you were on a flight and that it landed on time again, lots of data variables here.

[00:12:27] We send you a satisfaction survey. The NPS survey net promoter score would you recommend Spirit Airlines to a friend or colleague based on how you say, Yes, are you a 9 or 10? That’s a yes. Are you a seven or eight? That’s a right. Neutral is like an 8 is a neutral in the scale by the way, and then a 6 or below is like no, they’re actually actively telling people not to fly Spirit at this point.

[00:12:54] So until we resolve that we’re super not going to ask you for things like you want the credit card. No, they don’t want the credit card. This flow is something that is always evolving and something that we can kind of go back to and reitterate. The important thing that I’ll mention here is start with something. They actually waited a very, very, long time to launch this partially because it has a lot of moving pieces.

[00:13:19] But even if you just launched two or three of those emails, it doesn’t make the rest of it incomplete. So, start with the things that make the biggest impact on your bottom line or your customer satisfaction.

[00:13:31] And then this is the messages at each level. So, sort of education piece, the 24-hour piece and then the post-event survey. One quick note here, the styling on this survey is intentionally low branding like it’s less branding in this email because we want people to feel comfortable being honest about it. In email marketing, we have to accurately identify ourselves. So, we can’t tell you that it’s not Spirit, but we want you to feel comfortable being completely honest in this survey.

[00:14:00] Often than not people feel pretty comfortable being honest.  Okay. So, the I miss you campaign, this is the opportunity to reengage unengaged subscribers. Again, everyone in this room, no matter what email platform you’re on has the ability. Well, there’s one, okay. Probably you have the ability to determine the last time somebody engaged with your emails.

[00:14:21] And now we can either send you less stuff or we can take an active approach to saying sort of the we miss you. We’ve noticed that you haven’t come in a while right. In this case, it’s Twinkie the kid, how do you say no to Twinkie the kid? I miss you. By the way, I asked them for a yes and no button. Like do you still want email from us? Yes or no, and they’re like we have it under control. Yes or no.  Okay. Okay, that’s fine. What do I not like about this email? And I love Twinkie the kid and all he represents. It’s one big image. You guys are so good one big image. So, I have a group of people that hasn’t opened an email in nine months and my best shot at getting them to open is to send one big image.

[00:15:08] Come on, guys, we can do better than this. All right on the right-hand side. This is actually a somebody who’s donated for water.org. And then it’s been 13 months since they donated or opened an email and we send a very specific ask to them.

[00:15:23] All right. How about data that determines where to go next? Data in testing, data in optimization. So, in this case again split test in this case, we were testing a Friday versus a Sunday I believe. Whatever days those were February was ages ago. Friday versus Sunday, how many people are going to take the desired action if they get it on a Friday versus if they get it on a Sunday or if they get it on a Saturday, which is a great day for retail email versus if they get it on a Monday. In this case, we’re asking people not to go in and shop and we want that too, we’re asking them to complete their profile.

[00:15:59] These are people that got signed up and all we have is their email address, but we really should have had like their first name and what their favorite store is. So, if we can get a little bit more information from you, we’ll give you a 10% off coupon. So, we’ve got our value all loaded up. This is a test to see whether people are more likely to do this sort of administrative tasks on a Monday versus a Saturday or vice versa.

[00:16:20] This is a look into Hub spots A/B automated split testing. Okay, so I hate when a test doesn’t go my way. So, I had to put it in a presentation. On the left, the subject line is all the new school happenings. Thrivent student resources is about a new way of lending and a new way of borrowing for higher education that doesn’t leave people with $27,000 of college debt upon graduation. That’s the average number. So, their newsletter goes out every month and all the new school happenings could have been the subject line of any one of those newsletters. I was like, can we get a little more specific like what’s in it for me?

[00:16:56] Why would I open this on the left? And then on the right? How are ping-pong balls like student debt? BuzzFeed helps us explain.  That’s a great subject line. I love it. That is literally says BuzzFeed. It’s a Buzzfeed headline right. How are ping-pong balls like student debt? Drew Davis would love that subject line and it’s a video of a bunch of students in plastic boxes with ping-pong balls piling on top of them at the rate of their student debt every year.

[00:17:27] It’s very fun campaign actually except for if you’re claustrophobic, that would have gone poorly, but we let everybody out. So, we do the test 30 percent open rate versus 26 percent open rate.  That’s not what I wanted.  Okay. Well, okay. Well then how about clicks? Like which one gets the most clicks through on the video or on the Articles?

[00:17:49] Okay. So, YouTube, there we go. Now, I hate the way that HubSpot lays this out because you guys can’t see this, but the second link up here is the YouTube link Version “A” got 13 clicks and Version “B” got 15 clicks. Now we’re on track. Okay. Now I see what we’re doing. Further down Thrivent Student Resources is a cohort of the Thrivent Federal Credit Union and their goal is to help people borrow smarter and they offer student loans.

[00:18:20] It’s not nefarious. This is an open relationship. But how many application clicks did “A” versus “B” get? In this case, more clicks for Version “B” once they’re in the email than for Version “A”.  Those numbers are actually small. Okay, I’m out of time.

[00:18:42] So I have like two more things setting up for success with data. And the reason I put this in here and I want to leave this with you. Again, if you want your slides, just say slides to or something or if you already gave me your card, let’s just write it on there. Maybe I’ll just give everybody both sets Let’s not make that harder than it has to be. The reason I put this in here is if you’re going to use data it has to be in the right system for you to use it, and sometimes that means your email service provider.

[00:19:06] Sometimes all you need is HubSpot as your home for the data that you need to use in your emails. Sometimes it needs to be your CRM platform. In business-to-business, this is probably your main customer database, right? This is your sales CRM where people take notes or add customer service or whatever.

[00:19:24] Then there’s a third option and this is something that was new to me a couple years ago. So, I want to share with you guys, so you’ve got it. Data management platform, customer data platform. We got so many acronyms in this system. But the idea here is that you have a custom database that is specifically for types of data that are not necessarily aligned by contact. What I mean is and I’ll actually flip to an example because this will help explain it.

[00:19:48] Noodles and Company came to us and said, hey, we have all of our transaction data. We need you to tell us anybody who ordered mac and cheese, and customized it by adding chicken, but then customized it by adding buffalo chicken.  Which is like two tables lookup reappend, whatever. That’s what a custom database can do with your data and when you’re ready for that, especially when you got a really long sales cycle like Merida Tomes does or something similar to that where you’ve got a lot of different types of content that could go into that equation.

[00:20:22] That’s where this type of data setup starts to make sense. So, I just wanted you to have it. This is another quick example, and this is actually a vendor that you can plug into. I don’t make any money on this vendor called One Spot and this is where people go on your website and what they read on your website.

[00:20:37] They can propose the next most likely interesting article and you can use that on your website in your emails or in your ads. Super, super cool stuff that you can do with data here and I apologize that I ended a little bit over time again on this one. I’m going to give the room up to Lee for the next session.

Thank you, guys so much for being here.


This website uses cookies to optimise your online user experience. Some of the cookies we use are essential for the site to work.
By continuing to use our site you agree to us using cookies in accordance with our Cookie Policy.