Eduardo Ruiz de Pascual: Ensuring Customer Success and SFMC Advice | Episode 11

Edu: DESelect, so one that is very interesting for me was something we built for a Spanish retailer, and they were struggling because they were managing different languages for every single sense. So every time they want to send a newsletter, it’s four versions that the marketing team has to build. And when you want to do some personalization for different regions or different things, or events, what happens is that you have to multiply by four. So three versions being 12 emails being managed by the marketing team.

AL: And quick question, you say for languages like Spanish, Catalan, Euskera is another language in Spain and Gallego, Galician from the Galician Zone of Spain.

Edu: Yeah, that’s even within the country. You have these many variations, of course.

Anthony: In Belgium I recall similar issues with unique French. You need Dutch, you need German. But I think people sometimes forget there are so many differences.

Edu: And what we did was instead of focusing on the communications that we’re sending, we built a custom solution inside Marketing Cloud with custom landing pages, with script, with tons of AmpScript actually, where they have like a library of content blocks where they could define a period of time that content blocks should be active, the audience that should receive it the priority. So a lot of added information to each content block, and then they only define one template, one email, and it’s dynamic for each person depending on the segments they were in. And it would select the content blocks that should be displayed and order them by priority and even change the subject line.

Anthony: And so how many multiple variations do you have in your four languages? 

Edu: So yeah, it was crazy. I don’t know how many because for the segments they only had like 12 or 13 attributes they could use only. So it’s already a good number and then the languages and the priorities. So yeah, it was, it was very complex. But at the end, what I mean obviously technically it was already a challenge and it was amazing to build, but then you see how it impacts their strategy, like how the marketing teams start talking about the library and how they change their relation with providers. Because before it was like, okay, I want to send your email to 100,000 people, for example, but now I can make you pay more if you want to be on the first place or in the second place, because we also build reporting around how many times they’re viewed, how many times they’re clicked. So they could have a lot of data around that. And it was truly, truly cool.

AL: And I’m sure you learn a lot from that. For those out there starting a similar path or having projects that are as complex, are there some lessons learned?

Edu: Well, I would say developing something in Salesforce. It’s not always easy because there’s a lot of different languages involved, and the reporting, the error reporting sometimes is hard. So it’s better to follow a strategy of split and conquer. So you go with a very basic approach. When it’s working, then you try to add functionality on top of it. Because if you try to address the end to end, it’s very, very complicated. In this case, I don’t know how many. I think it was like five months’ or six months’ project, just trying to figure it out. Every single thing that can go wrong at the end. I think the template, the main template, has like 1000 lines of script. 

AL: Now another thing that you learned during your consulting times, but that you’re definitely applying now, and this is where you impacted our roadmap so much is the issue of saturation control. And then of course manifested itself in our recent product Engage. But can you first start with saturation control? What was your experience with it?

Edu: Yeah, I feel like saturation control for many companies it was like this New Year commitment that you never fulfill because every time you bring it up in a conversation (and I don’t know if you had the same experience) is like, “Oh yeah, saturation control, we should do more about it.”

But nobody’s happy with their current approach to saturation control because it’s a very complex process. First of all, it impacts everything because sales have something to say as well, because you’re reducing the funnel, right?

AL: So maybe you should follow the audience, we should take a step back. What is saturation control, what’s the basic problem?

Edu: So basically the idea is control. How many emails or how many communications you send to your audience while you make sure that you don’t saturate them and they finally unsubscribe. Because even though every message sounds important, it’s better to send two emails a week than five emails a week and then lose them on day six. So yeah, the idea is trying to control how many emails you send, but also taking into account for the complexity of your company, like the priorities and the things that are going on and what messages are important. Because not all messages are the same, and there’s no solution like [Engage] right now. And it’s very complex to build something like this.

AL: Yeah, absolutely. And that’s of course, where we came up with Engage. You’ve been at the cradle of this project. What do you consider the key capabilities today as it stands, and do you have a favorite feature?

Edu: Well, definitely. Also listening to customers’ feedback, the calendar. Having a calendar not only to know which campaigns are you going to send (sometimes it’s done manually or in an Excel file), having that added component of Engage telling you for each campaign that you add there, this is the percentage of the audience that you define that is going to be oversaturated, that should be excluded based on your own rules.

So having that vision gives the customers more arguments and more ways to solve the problem. Because if you have a very easy solution, then the answer to a campaign being saturated is binary. It’s either you send it or you don’t. That’s it. But now with the calendar, if you see a campaign is going to be affected by saturation, you can decide to move it and test it and refresh. If you move it to Tuesday instead of Monday, it seems that more people are going to receive it. So that’s giving them more tools to solve it.

AL: That’s your favorite feature then?

Edu: Definitely.

AL: For me, that was sort of the interesting conclusion of that. Well, not conclusion. The project is still early and we have so many more things in mind. But first, understanding the problem of saturation control.

I sort of imagined the solution differently than a calendar, which it now is. But I think that that’s sort of a very practical, easy to grasp thing. And it’s kind of cool because it’s it’s interesting because we make it seem very easy. But underneath the hood, I know our developers have done a crazy amount of work to make it look that easy, and I think that’s the same with Segment, right? People are like, “Oh, this is easy, but the kind of SQL we can spit out on the backside is amazing.”

AL: So getting back to Marketing Cloud more generally speaking, maybe aside from saturation control, how do you think marketers struggle? Where are the pitfalls when you start your Marketing Cloud project? And maybe for non consultant types like the actual users, where do you think are the biggest struggles with Marketing Cloud, maybe with marketing automation in general?

Edu: Yeah, I think that as a manager I had it at Deloitte used to say, “When you get Marketing Cloud for the first time, it’s like purchasing a Ferrari without a driver’s license.” And sometimes it feels like that because you have a lot of power, you know, you have the fastest car or the fastest solution or the best solution out there. But you have to learn a lot to get to the point where you can take full advantage of it. So what I really like about DESelect is that we’re flattening that learning curve of the problems you’re going to face. Okay, let’s reduce the complexity of the ones we can, so you can spend more time on other things or you can test things out. So yeah, I would say normally that’s the biggest pain point. When you start out there, you don’t know where to go. Like you have a lot of studios and it’s hard to decide.

AL: You need to get your driver’s license or, you know, go to track and get some laps.

Edu: Exactly.

AL: You mentioned DESelect addresses that. Can you give our audience an example maybe for those who are a little bit less familiar with our offering, what’s your favorite aspect of our product or service?

Edu: Well, we’ve talked about DESelect Engage, about DESelect Search, being able to find things faster sometimes and maybe it’s not so much in the onboarding, but later when you start having a huge library of content of data extensions, it’s hard to know where you save everything unless you start doing that from day one. That’s maybe good advice. Also have a solid structure of folders inside Marketing Cloud.

Anthony: You know, I totally remember that from my first marketing automation project. Like just having a good folder governance and naming conventions. It sounds almost too simple or kind of stupid that you have to put your brainpower into that, but it makes such a huge difference down the line.

Edu: Yeah, definitely. So with Search being able to find these data extensions super fast and being able to click on them, when you search for the name, you click on it and you see it right away. So that and also with DESelect Segment, with all the segments, it comes to segmentation, but also reporting.

Sometimes it’s hard to navigate through all the data that you have available and being able to combine it with drag-and-drop and doing previews. I think it speeds up the process to get to the result from the okay, “I want to build this segment,” or “I have this idea of something I want to try.” It really speeds up the process to get to a result so you can validate faster, and that is something that I think a lot of people don’t realize about writing queries with Salesforce Marketing Cloud experts or using DESelect Segment, that the queries are not often, often not only for segmentation, but they can serve a lot of purposes.

AL: I actually used it for reporting myself back when I was working as a consultant and I would build these queries to compare performance of certain pieces of content across clients. So it was a combination of using the send log, using script to log which locale, country, language, all that stuff, and then the content piece. And then to get that data, you had to write big queries. And then of course the volumes were large. I had to use the FTP. And then I like to call this data prep.

AL: Have you seen other interesting use cases that maybe we tackle less today for SQL beyond segmentation or reporting?

Edu: Data validation. That’s pretty common as well. I’ll give you an example. We had a customer recently that uses a specific email when they don’t have an email. I think that’s pretty common when you have a form and it’s required. It’s like mail, add, note, email, dot com or something like that. All right. So being able to get the list of the people that has that email, it seems very simple.

But when you’re doing something complex like what you were mentioning before, it makes sense maybe to do a huge investment in time or you expect to do a huge investment in time. But when you’re validating data, you want the results in two minutes. It’s like how many people are without email and see how long it takes to build every single element that you need to get to that result.

So yeah, I would say with DESelect it’s easier to validate data and we’ve seen many customers that during the onboarding they start calling their CRM colleagues like, “Hey, we’ve identified this error,” that’s pretty common.

AL: So this is something that even our customers are actually using DESelect for: data validation not just segmentation.

Talking to our customers, the issue is often upstream, so maybe they’re using Salesforce, CRM, they’re using Snowflake, or whatever. And they come to Marketing Cloud and then the marketers have to deal with that.

Do you have a point of view, an opinion on the problem from your experience with customers, maybe what they architecturally do? What do you think about that?

Edu: Well, I think the main problem for marketers or the main pain point is that when you identify those errors, maybe that’s blocking a campaign you have in the next few days or something that you want to try.

But for the CRM team, it’s going to be just another task, just another thing they have to do, and it needs to fit a sprint, obviously, because it’s a very complex solution that’s used by everyone in the company.

So the problem there is that you identify something that is a stopper for you and then you have to wait in the queue until it’s solved. Right?

Yes, And the good thing about DESelect is that you can fix it on your own. So the idea (and that’s something that we share with customers) a lot of times is you can make that request to fix it, but you can also build something to do a patch for the time being until it’s solved. You scale that and then you work from there. And that way you can bypass all the errors that you may find without impacting your strategy.

AL: How you’re explaining it is that for some customers, they essentially get dirty data into their Marketing Cloud, but the marketers have to use it so they do some kind of staging of their data using queries, segments, whatever.

I’ve even noticed that some customers say they don’t even have a proper definition of what is Segment A, right? So if they sell multiple products, they just don’t have good data governance around how to fix that.

Who do you think is the owner of that problem in your organization? I know it’s a hard question. Maybe no right answer, but I’m kind of curious what your opinion is.

Edu: As I’ve said, maybe the owner of defining the segments, it’s not the person that is using the segments later. And what I mean is that maybe the CRM needs to have clear data, right? And clear attributes and flags that can be used later in marketing. So marketing is just a reflection of what you have in the CRM.

So maybe the owner of making those segments clear or making those tags, it’s in the CRM, but at the end of the day, if that’s not well defined, marketing is the one that is suffering, is struggling with that. So I would say it’s co-owned, the CRM team obviously long term needs to fix that and make sure that’s in place. But short term, marketing needs to have a solution to address it because it’s, at least in my experience, so common.

AL: I think you kind of touched upon something interesting. I think what’s sometimes happening is that supposedly the owners of the data are maybe sometimes data stewards, data scientists, people owning the data warehouse or the CRM admins, but the people suffering are someone else downstream. So it’s kind of an interesting ownership problem. I don’t think anyone has really figured that out yet. It’s sort of the modern version of IT versus marketing, which is often a conflict.

If I can take even a broader question, how do you think about technology augmenting human intelligence? There’s so many interesting things happening. We’re, of course, on the vanguard of pushing data enablement, but there’s also just recently ChatGPT getting super popular and big. How do you think about that? How is technology enhancing, enabling, augmenting human intelligence?

Edu: Yeah, I think it’s a great question. And I really like how it’s formulated because for me, something that tech can do and that humans will never be able to do is process huge amounts of data objectively and jump to a three bullet conclusion or to a series of next actions. Right. And that’s very hard for us, obviously.

But at the same time humans have something that ChatGPT will never have that is context about the sector and based on your own experience and that gut feeling, that decision that was based on probability, maybe sometimes you take a path that you know it’s not the best one if you only look at data, but that subjection, it’s so positive, right?

So I really see technology doing that part of the job of collecting the data and presenting it in a way that humans can make better decisions. But I don’t believe in a future where technology or AI replaces the whole chain of decision making from end to end, like black boxes.

AL: I don’t know if we’ll get beyond that. And that applies to, you know, not just for ChatGPT, which is literally a chat bot where you can ask questions, but also for: What segment? What should I segment? What data should I use and look into?

Edu: For example, like the approach we have at Engage where, okay, data saturation can be different for every person, but we let customers set their own rules. So the machine is there working for the human and then applying what the human says. And you can go beyond that and do a 1:1 personalization, but only always keeping in mind hard rules that maybe a human has set up.

AL: For someone who’s not a Salesforce Marketing Cloud expert, but just a newbie, they’re just getting their hands dirty. What advice would you give them?

Edu: Well, first of all, welcome. I would say the best thing that Salesforce has, it’s definitely the community. How many users are there and probably any problem you’re having right now was solved by someone before you and there’s a blog post about it or a video tutorial on how to solve it. So my first advice would be when you log in, when you don’t have the driving license, right? And you have your brand new Ferrari, instead of starting navigating through Salesforce and trying to figure out how everything works, that you take a step back, you go to Google, and you start searching for the things you want to do, the things you want to implement. And there’s a lot of information out there very easy to reach, and, I think that’s the real game changer with Salesforce.

AL: Do you have some favorite resources? Maybe you’re using with your own team here at DESelect, or was there a specific module or community or group that you found super helpful?

Edu: I don’t have any specifics because as I’ve said, most of the time, I start from Google and then I start looking. But, Elliot Harper obviously, there are so many good videos about him.

AL: Very, very first interview on Heroes of Marketing Cloud.

Edu: Definitely a hero. Also, I’ve seen videos from the early days. There are some videos in the DESelect YouTube channel that are really useful. If you start digging, there are some communities on Slack. There are forums specific for Salesforce.

AL: What if a customer layers on DESelect? What is the number one blocker or number one advice? What’s the most important thing to do to have a successful onboarding with DESelect?

Edu: So I would highlight a couple of things there. First of all, internal alignment. This is very obvious and it applies to every project, right? But once you identify the pain points that the DESelect platform can solve, I think it’s important that managers select the solution while listening to marketers, and marketers embrace the solution proposed by managers. So that’s alignment where everyone is trying to go in the same direction.

That’s very important for change management, but also curiosity. I would say we have customers with no technical experience, with a lot of technical experience, brand new to Salesforce, and Marketing Cloud experts that have years of experience that were in Marketing Cloud when it was ExactTarget.

The thing that successful customers have in common is that they just go in and try things out. I said before that it’s hard to address Marketing Cloud if you start going by the different studios, and sometimes it’s scary, right? You go into Email Studio and you don’t want to send an email to 1 million customers by mistake, right? But the good thing about DESelect solutions is that they are free to try out. They are safe.

Especially in Segment. You can go in and start building segments, try it out, try the different types. So what happens if I use this join type? What happens if I use this filter type? You have a support portal where you can even ask your customer success manager, and they will be there to help. So yeah, I would say those two things.

AL: There are guardrails to it, so maybe the risk of accidentally sending an email to a million people is lower. So what you’re saying is, that’s what the most successful DESelect customers do. Is that something that is exclusive to maybe your more technically versed customers, or would you give the same advice to non-technical marketers?

Edu: To everyone, actually. There’s a customer that has been with us for three months now, and I’ve always talked with them about how advanced they are in a very short period of time. It’s really impressing because they are doing crazy reports. They are building scoring models from week two, and they don’t have technical background. They are all marketers without technical backgrounds.

And I asked them and they said, “We’ve been wanting to do this for a very long time, but we couldn’t do it because we didn’t we didn’t know how. And now we have DESelect!”

It’s like they have a door that we can explore and that’s why they’re trying out. So I think this applies to to everyone.

AL: Very, very cool. I’m sure a lot of our customers have thought about scoring models. Do you have some insights in how they’re thinking about it? What was their approach? Because clearly they had thought about it before DESelect. So how do they do that?

Edu: So normally when a customer wants to start building a scoring model, there are three steps I always try to share with them because normally when you have a scoring model, you go to the last step, that is, define the attributes you want to set. So label level A, B, and C, right? That’s the outcome. But that’s only the last step.

The first step would be to identify the events that need to be processed for that scoring model. It can be engagement, it can be if they’ve purchased something. So all the events that you want to tackle, and that would be your first step. Identify them, and see if you have the data there to be able to identify it.

Then the second step is giving scores to each event. So each event can have a score, or you can prioritize them or however you want to deal with it. And then once you have that, you can go to the third step of saying, okay, with this score level A, with this score level B, etc. So it’s splitting the process in these three steps. It’s clear for customers normally to address it.

AL: And it simplifies. And I think it’s also key that, I think customers are also a bit worried about trying to do it too. Initially, maybe you can have a scoring model, but it doesn’t need to be actionable right away. What I mean with that is the scoring model doesn’t really have to change lead assignment rules or new journeys. Like you can just let it run and then kind of check afterwards. Okay, what are these scores now? Do they make sense? Do we all agree about that? And, and iterate from that.

Maybe just a bit of a fun question to round up the conversation. Is it true you were part of the Real Madrid youth academy?

Edu: Yeah, I played there for a year when I was 9 or 10, something like that. So very young, but it was a great experience.

AL: So what was it like?

Edu: Well, I’ve always been a Real Madrid fan, so it was a dream. Fantastic. Yeah, it was. It was very fun.

AL: So if we’re going to have soccer games with the company, you’re going to whoop our ass? 

Edu: I’ll try to at least. Yeah.

AL: Great. Maybe this is a funny comparison, but how does that experience compare to your DESelect experience?

Edu: Well, at the end there’s a team component that is shared between sports, business, and DESelect. So basically celebrate the big together as a team, and if something’s happening, if someone needs help, make sure that the whole team can react to that and help. So yeah, I think that’s a mentality that is shared in Real Madrid and DESelect.

AL: Well, thank you so much for being here. You’ve had a massive impact on our roadmap. You’re doing tremendous work with your customers and your team. So thank you so much for being here for the ride with us and being here today for the interview.

Edu: Thank you very much.