Speaker 0: Okay. Welcome, everyone, to day three of MarDreamin, our last day. We’re super excited to have you all join us today. Um, my name is Kelly Ryan and I am from Sercante. I’ll be moderating today’s session. Um, before we get started, I have a few housekeeping items to go over. So, yes, these sessions are recorded and will be available on demand after the event, and we will also be following up with them via email. So have a question post it in the Q and A tab over on the right hand side there and we’ll have a couple minutes at the end of today’s session to go over them. And then lastly, use that chat. There are emojis, gifs, and many more things. We want to hear from you. Please be interactive and we’ll have a couple polls during the session today too, so make sure you’re ready. Okay, so now I’d like to introduce you to our speaker today, Jennifer Rivero, who has a great session ready for all about connecting the dots with data. So, Jennifer, I’m gonna go ahead and pass it to you.

Speaker 1: Fantastic. Thank you so much. Um, I would like to get started by just saying a big thank thank you to all of our incredible sponsors. Obviously, thank you to Circante and Salesforce, also to Stencil, PFL, Gridmate, The Spot, and Fionta. I do have a fun session, uh, planned for you today. We’re going to be talking about, you know, if content is keen, then data is clean. Uh, understanding your audience, What is essential for data? And ending on data governance.

So people always talk about content being king. Well, it’s not just about content. Um, it’s about sharing relevant content that is personalized to people, telling people what they want to hear more about, Right? There’s so much stuff out there, so much content out there. Um, what’s really become critical is showing people how to find and discover what’s relevant to them. And what’s easiest is if we can actually deliver content that’s relevant to them. Um, you can see here we’ve got a king and queen shopping for groceries, um, it’s it’s a partnership, right? It’s not just about content, it’s not just about the data, but how they come together and you know what? If the content’s not relevant, uh, don’t even bother. Like grocery shopping, if I don’t like the food, don’t even bother bringing it home.

If content is king, data is queen, right? Um, so here you see P and G, uh, Procter and Gamble and, um, a little bit of how their marketing has changed over the many decades. Um, specifically, here you see Tide and a nice quote from the CEO. And, you can see over the past two decades how companies have been moving from demographic messaging to psychographic groups and more one to one messaging. Um, here he talks about how Procter and Gamble has gone from generic demographic targeting like sending the same exact message to all women aged 18 to 35, to creating more than 350 precise smart audiences. Um, Procter and Gamble, they have more than a billion contacts, and it’s exciting to see that they’re now utilizing, um, AI tools and getting smarter with all of the data that they have collected about their consumers over the last few decades to send more relevant contextual messaging. Um, you can see they’ve highlighted first time moms, millennial professionals. Um, I know that even for myself, I have moved through these different types of psychographic groups. And as I’ve moved through them, I’ve my own, um, shopping needs have changed. And during that time, I have been in that same one demographic targeting group. So, um, it’s great to see that the the data and the capabilities of AI and what companies can and are starting to do are catching up to the strategy.

I want to highlight an email campaign that Peacock sent out, um, last year. Uh, this was a case study that was featured in Marketing Brew and Braze. So, if you have not gotten a chance to read about it, I would definitely say please go and read about it because I hear all the time, hey, is email marketing dead? I get a lot of emails and I don’t, um, read them all the time. Um, my answer is no. You may be getting a lot of emails, but the ones that you do read are the ones that are relevant to you, um, the ones that catch your attention and that those are now fewer and farther between, but they are also more impactful. So to talk about what Peacock did, um, they they, uh, took all of their their viewership data and, um, there there are so many more rules in place now about how privacy needs to be respected and so, um, they were not able to, you know, um, show back to you exactly what you’re watching, but they were very smart about how they, um, identified those, uh, behavioral groups of customers and, um, showing them back what was most, uh, often watched within those groups. Right? So you felt like you were, um, getting a review of your user activity in a very personalized way, but still in a way that respected your data privacy. Um, and this campaign showed business results. So just a highlight of the business results, um, they saw a 20% reduction in churn among paid subscribers, 6% higher upgrade rate for their free users. So people who weren’t spending money went, oh, hey, I should give you guys money now. And I think it’s important to see these numbers in context of their competitors. So for all other premium subscription streaming companies, um, they saw churn rates increase while Peacock saw a 2% decline in their churn rates. So, this to me is just proof that, um, we need to do better with sending relevant messaging to customers and when we’re able to accomplish that, that we see results.

Alright. So when you get started with the campaign, we all know that the the most important thing is to understand your audience. Kelly, would you be able to start the user poll, please? That I wanna ask everyone, how often do you receive a request in your organization for an email campaign, but there are no details on the goal or the target audience? They say something like, hey. I just want you to go ahead and send this email out. Here’s what who I want you to send it to, um, and here’s what I wanted to say. Um, and I don’t know exactly how the the poll works. So are you guys able to see it now?

Speaker 0: If you wanna stop sharing your screen for a moment, I can post this on the screen.

Speaker 1: Oh. Okay. I see you’re starting to

Speaker 0: get some answers in and they’re obviously on the screen right now and everybody can either hit the poll, you can, uh, go to the right hand tab to poll and then you cast your vote there.

Speaker 1: Awesome. It’s kind of fun to see this. What? No one was ever saying, um, this has never happened? Yeah. Um, I know I’m here preaching to the choir. If you’re here, you’re probably a marketer or you work with marketers, and so you’ve been in this position. And we know that emails are a tool, um, for us to enable a larger business objective and for us to further the brand. It’s it’s a, um, it’s an opportunity for a brand touch point. Let me go back to sharing.

Okay. So it’s important that, uh, we’re we have the opportunity to work with the the business stakeholders to create that campaign brief. Right? Define what’s the goal of this campaign? And even if they’ve already got an idea of who that target audience is, um, validating, uh, that that is the right target audience and understanding, well, after they see this campaign, what is it that we’re trying to get that person to do so that, um, we can make sure that we’re all every everything we’re building and doing is driving toward that one goal. And so we have context not just for that one email that goes out or one communication, um, but that we see it in within that customer’s entire journey. Um, I have a client who wanted to send, um, an email for their one particular brand, um, that was geared toward, uh, it was basically like a, um, an ethnic rice and they had made assumptions around well, this is the brand of the rice and so this is who I’m gonna send it to but then when we actually pulled, uh, who is purchasing that particular rice, it was completely different from, um, who they had assumed. And so it made them really go back and think about who’s buying it, who should we be messaging it to, um, should we be reevaluating the whole, uh, branding of it? So it was very interesting.

Okay. Um, so this slide and, um, imagery hopefully is something that is familiar to you. Um, it is adapted from, um, materials that Salesforce uh, uses. So I saw this in a a few different Dreamforce slides and, um, really appreciated the way that they they pull it together in showing how the the enabling technology is, um, foundational to how you engage with your customers. Um, but, really, even if you’re not on Salesforce, this is relevant to you regardless of your your platform or toolset. Right? So you can see at the bottom, the foundation is that data. And it’s exciting to see in the last few years, we’re seeing more and more of AI being integrated into that, uh, with the volume of customer data that’s available out there today, not just customer data, um, but third party data that you can pull in and and use to enrich, um, the data you’re working with. AI is no longer optional. Right? You really need to be leveraging AI tools to get smarter about identifying what’s relevant and have that data work smarter for you. And then all of this is feeding into that holy grail of the single view of the customer. So you can see here, a unified customer profile. Um, this slide makes it look easy. You have all the data just feeding in and it gives you that, um, single view of the customer, well, this is not easy. And I think that’s why people are getting so excited about data cloud, um, specifically from Salesforce and why you see a lot more CDP solutions gaining traction in the last few years. Um, But this isn’t a presentation about data cloud, so I I won’t be, um, talking at length about that. Uh, it is nice to see that the recognition for data driven marketing. I think that sometimes when you’re working with, um, business customers, they they imagine the the days of, like, the the Don Draper style Mad Men marketing where it’s highly creative, um, and there’s a lot less emphasis on on that data side of it. Anyway, once you get all of this data put together, um, this now becomes, uh, an enabling tool for all of your employees in an organization. So you can see on the outside in green, um, empowering your employees so that they can better collaborate, um, power their workflows and automations, and be more agile as they serve the customer. And then here in the middle, you’re seeing the center of all things is the customer. And I really like how Salesforce had highlighted the word listen. We do need to listen to them. We don’t do enough listening whether it’s listening to our customers or to each other, um, and then obviously actioning and engaging.

Okay. Now that we’ve established that we need to start with the customer and understanding the customer, let’s get to identifying essential data. Um, at the end of the day, if we’re running something like an email campaign, what do we need? Uh, and now you might be asking, wait, Jennifer, you were talking so much about getting, um, um, relevant personalized info, uh, content out to customers. How can you do that with just this bare bones data? Well, it is about starting with what we need and then looking at what we have and building on it. So if we start with the absolutely essential data, obviously, we need our email address, how will we reach them, um, subscriber ID, uh, if you’re in marketing cloud or some other way to uniquely identify, uh, that person or record in the system, primary keys, and opt in and opt out status. So, um, this can be different based on how, uh, where you live and how that opt in and opt out has to be handled, But as an email marketer, you have great power and with that comes great responsibility. It’s important to be compliant and respectful of customers. I’m sure oops. I’m sure this has happened to you, um, but I I’ve run into a lot of companies where they might look at identifying, um, their customer base, their emailable customer base, and they just look at their active customers. They don’t actually look at whether or not they have an email address for them. So then when they go, oh, you’re right. If I’m looking at an email campaign, I’ve gotta look at only people I’ve got email addresses for and bring that back to who I’ve got valid email addresses for, that pulls it back significantly. So just a good reminder. And then, um, you can start layering on additional data. So what types of data do you have that’s available to you for demographic uh, information, um, personal preference information? Do you have behavioral data, um, what they are engaging with? And transactional, depending on the the field that you work in. Right? Um, I think here it’s important to understand, you know, what pieces of data you’d like to have and what systems they’re coming from and how you’re gonna connect that all together, um, but don’t fall into the trap of gosh, I have all of this data. I just want all of it in case I might need it. Um, and then you end up with hundreds and thousands and literally thousands of, uh, different pieces of data elements. Um, you are going to end up in analysis paralysis and try to figure out what the data is going to be telling you. As well as from a technical standpoint, your architecture and system performance is going to suffer. That’s why you always wanna start first with understanding the customer, the business objective, and then letting those tell you what data is going to be important to you.

Of course, data quality can make or break the value of your data. Um, so these are just some ways that you may end up having issues with your data. So missing data. I was just talking about email addresses. If they’re missing, they, um, have no value to you. You may also have duplicate data where that data is in, um, multiple places and you run into an issue around, um, what is your source of truth. Misplaced data, um, maybe there is a Boolean field or a pick list value or something where um, it no longer meets the need of the person putting in the data. And so rather than putting in that data into a structured way, they’re starting to put data into unstructured fields or, um, other places so they can capture the information and then, again, it becomes more difficult for you to work with. You want to identify those and, um, figure out how you can address it early before it becomes a problem. Uh, of course, inaccurate data or inconsistent data, um, naming conventions, drives me nuts when someone wants to make state an open text field. Why? Why? We know what states we need to put in there. Why would we want to to let someone write that in? And then, of course, um, a theme that I’ve been talking about today, which is too much data. Too much data is also a problem that can impact how you can utilize your data.

Okay, identifying potential data, garbage in, garbage out. So I like to say that unreliable data is not just garbage in, garbage out. It’s even worse than trash. So when you’ve got bad data, it costs you money. Um, so I really like the way that this pyramid lays it out. Um, the dollar is the cost for preventing bad data. So if you can proactively prevent bad data, um, that may be an additional effort, but it’s definitely going to be less costly than having to then correct, um, something that is a result of that bad data. And then you can see how it continues to grow. Your problem continues to grow. And finally, the last one being, um, the cost for fixing a problem resulting from bad data.

Um, now I’d like to show you an example of this. Okay. Um, so here on the screen, I have, uh, an example of something that happened, um, and why it’s so important to understand the context or the business process around how data gets populated. K. So, um, for, uh, one of my clients, um, they had a b to b sales team and that b to b sales team, if they didn’t know the name, they populated a descriptor for that customer. And this was a client who also wanted to use personalization elements in their communications. So they said, well, why don’t let’s keep it simple. Let’s just bring in their first name. Well, gosh, if that salesperson didn’t know their name, they populated a descriptor, which could be bald guy, fat person, or, uh, you know, you can use your imagination, but I would have just died if this email had gone out. But if you think about the kind of email data validations that you would have built into place, like, making sure that there are no special characters in the name, um, that it’s not blank, um, this wouldn’t have caught it. So that’s why it’s important, um, to also understand that context, uh, the business context around the data.

Um, I would like to pause here and just send that second poll, Kelly. Question is how often do you run into bad data? Feel like this is all the time. I mean, if you weren’t cringing for my example no. Um, just going back to that that pyramid, talking about the cost for preventing bad data and just reminded of a of another scenario that had happened. Um, a very simple one where we had Salesforce integrated into marketing cloud. Um, but the teams that were doing the setup of Salesforce, Marketing Cloud, and the connector, they were all in different time zones. And so, um, Salesforce was set up in one time zone, Marketing Cloud was set up in another time zone, and then journeys were being set up in another time zone. And so there were expectations that customer, uh, that, uh, business teams had around when certain journeys would end up getting triggered and the they were, um, doing a QA review of, you know, well, this is the audience I expected to go in and receive the email today versus tomorrow. But why they consistently had issues of people being identified, um, uh, on day one versus day two. And we ended up narrowing it down to this being a time zone mismatch issue. And, um, after that, we realized that we had been designing around that time zone issue in multiple journeys and scenarios. And so you’d see, um, rather than having prevented that from the beginning, um, they they were continuing to build up technical debt to address the problem. What sucks is that once you have that kind of technical debt, people don’t usually wanna solve, uh, spend money in order to address it. Um, but thanks for thanks for doing the poll. You can see it’s, uh, uh, always and often, it’s kind of painful. 75% said always or often.

K. So going back, um, I’m just showing this again because I feel like this is a great depiction again of just that holy grail of the single view of the customer. And I know before I was saying, well, you know, maybe part of the answer is something like data cloud or CDP. Uh, technology is part of it. It is absolutely part of it, but there’s so much more beyond the technology, um, and that’s all around that data governance at your organization. And will the real Jennifer please stand up? Any, uh, nineties Eminem fans here? Don’t worry. That’s not a poll question. Um, and so here’s where it’s it is about making sure that the the business comes in and owns their data and how, um, how that data should be managed. So determining who that real Jennifer is and how that data all come together, that’s up to the business.

So let’s talk about data governance. Data governance, um, I’d say most companies that I’ve worked with, um, probably don’t have not they probably don’t spend enough time on it. Right? It’s it’s something that is absolutely critical, um, but in most cases, haven’t gotten as much, um, attention and, uh, I think resources as, um, as it really should, especially given how data has just, um, exponentially ballooned in the last decade. So if we look at the key components of data governance, um, the first process is data life cycle. Um, so this is defining how is data created, used, stored, archived, and deleted, uh, and not just data, but also metadata, is how are you facilitating discovery of that data, um, across that data life cycle. Um, I have highlight deleted because we don’t want to necessarily keep data forever. Right? If you were a customer of that company twenty years ago, should they be maintaining that data about you today? Um, um, or is it a is it of value to them, and is it, uh, or is it a bigger risk?

Data quality. We have talked about data quality, but, um, obviously, it’s important to ensure that you’ve got that accurate, reliable data and what are the, um, what are the processes that you’re setting in place to cleanse that data or validate that data to help facilitate better data quality.

Uh, data access and usage. So who can access that data and for what purpose? Um, I joked earlier with great power comes great responsibility, um, and so we as a within your organization, want to make sure that, um, it’s only data is accessed only on an as need basis. Right? Which leads right into data security and privacy. Um, how are you making sure that you’re respecting regulatory standards? Um, from the email marketing world, I know we’ve got Capstan, um, CASL, GDPR, but there’s may also be other regulatory concerns depending on the industry you work in like HIPAA. Um, and then, of course, how are you protecting from, um, unauthorized access and data breaches? We we are seeing more and more, um, headlines around, uh, data, unauthorized data, um, leaks, and we we don’t wanna be the next headline. Right?

And then data ownership. So who in your organization is accountable for data governance? Um, when I work on teams, I like to joke that, you know, hey, we are delivering as a team, but who’s still that one throat to choke or butt to kick or who’s who’s going to keep this moving forward? So who who are those people and what part of the data governance process do they own? This is a lot data governance is a lot to set up, um, overnight. And so it is recommended that you identify short term versus long term action plans. So long term maybe defining that data governance strategy for your organization, but short term, maybe you’re building from the bottom up, first focused on that essential data, making sure that, um, maybe you’ve got fewer pieces of data you’re working with, but it’s all of quality and then creating that prioritized list and an action plan of how you can continue to grow, um, that that data you can rely on to build more relevant and personalized campaigns. But as with everything, focus on what’s important, which is starting first with the customer and the goal of the campaign And, you know, only using the data that is essential and reliable.

Uh, I’d like to, uh, stop right now for questions.

Speaker 0: Awesome. Great presentation, Jennifer. Love the real life examples. They made me panic a little bit. Um, it doesn’t look like we have any, uh, other Q and A’s right now. And we only have about ten seconds left, so great job.

Speaker 1: Well then, thank you everyone.