Speaker 0: Good afternoon, everyone from our dreaming. Hopefully, you got to attend that awesome executive keynote with Salesforce team, um, and learn a little bit, uh, about AI and where the marketing cloud is is going. Uh, but today, we’re gonna continue that trend talking about marketing cloud, and we’re gonna be joined today by Jacob Hayes from Silverline. And he’s gonna share his topic on exploring data cloud for marketing, architecture, and orchestration. So with that being said, um, make sure you throw any questions you have in the chat, use your q and a, and give Jacob your undivided attention. I’ll see y’all later. Take it away, Jacob.

Speaker 1: Thanks, Marcus. Uh, good to see everybody. Yes. Lovely long title for an intro of exploring data cloud for marketing architecture and orchestration. Yes. Jacob Hayes. I lead marketing cloud personalization as a practice lead at Silverline, also dabbling in data cloud and adjacent products related to marketing cloud as well. So first off, as my screen catches up with me, really appreciate the fact that Marjorieven kind of invited me. I’m doing these sessions. I think the Circante team, the Salesforce team themselves, sponsor’s fantastic. Um, I wrote up a little something on the blogs a while ago with the team that basically say this is a really interesting version of a conference, and I’m really excited to talk with you all about some pieces that I’m seeing in Data Club.

Alright. So with that being said, what are we gonna try to accomplish today? I’m gonna do my best to speak as quickly without word vomiting at you of lots of different topics. And so the thought process here, I wanna do a quick overview of data cloud. You guys have probably all heard some sort of a view. I’ll do my own flavor of it. I then wanna run into and jump into a visual ization that I like to talk about when it comes to architecture and orchestration for data cloud for marketing. Then jump through best practices, how to approach a user experience use case with data cloud for marketing. And then finally, we’ll ended up with, uh, key takeaways in the q and a. I am gonna do my best to leave time for a q and a at the end of the session, so please do put your questions in, um, throughout. Alright. So with that being said, let’s dive in.

Alright. So what is data cloud? When I like to look and discuss data cloud overall, I like to say and talk about and what you probably see in this overall slide is, you know, how data cloud connects to and supports the ability to upload any and all available data sources as you see on this left hand side that sync cell files to on prem systems, to cloud based data lakes. Many of these connections are supported with out of the box prebuilt connectors to web and mobile data, all of the Salesforce clouds, AWS, and Google Cloud. But data cloud also comes with the ability for users to build their own connections via integration APIs, streaming APIs, as well as more connections via MuleSoft.

Now one big thing to talk about is all this data within data cloud is on what we call a bronze layer, which, for those who don’t follow fun data architecture lingo, means that this is raw, untransformed, unmodified data or as is from source systems. Now Data Cloud can take this raw data and start to transform it via batch or streaming transformations into what we call silver tables or data lake objects within the platform. Silver tables take the data from the bronze layer and start to prepare, match, merge, conform, and cleanse just enough so that the silver layer can brighten enterprise view of all its key business entities, concepts, and transactions. That alone is not usually enough because what happens often is client enterprises, in client enterprises, that schema of all those objects are very different. For example, right, where I work a lot with HLS at Silverline is called a patient when I’m in a clinical setting, but I’m a member when I’m in the context of health insurance setting. But think about kind of how each one of these individuals are thought of in your own clients’ organizations or your own organization yourself, and think of how you would conform data cloud to that perspective.

Now this process enables, you know, customers and who are using Data Cloud to relate data across their entire organization and translate into a single language. And that single language and why it’s important is because these items are often called, labeled something different. And so once it’s complete, Data Cloud provides clients with the options to match and reconcile this data. Data Cloud also offers AI and deterministic methods to create unified profiles or individuals so that clients know that this person is the same one. That opened a marketing client email, has an open service case, etcetera. And now this provides harmonized, unified data for your enterprise. What this does is unify that data to create a total customer profile, customer graph, or even a more robust single source of truth. From there, this profile can be used by any AI, BI tools, including all of Salesforce products for predictive analytics, recommendations, data discovery. And then finally, this data can be acted upon to save time, automate, and process actions as you can see in the activation set on this far right of the screen.

Now the main concept to take from this whole diagram, which again, I again, I’m sure a lot of you have heard this, is the power of the platform and the power of the hyperscale of the cloud is all wrapped up in this data cloud platform. And it’s this new version, new thinking process of the Customer three sixty. Or is it lead? In my personal opinion, this leads to we’re always trying to connect create a connected experience of either marketing moments, service moments, or sales moments. And if we can take those discrete moments and then connect them together, that’ll either strengthen or weaken our customer’s preference, loyalty, and ad and advocacy. And this is where data cloud heart it really hums, and it takes the next best step of this platform. That’s why we’re spending a lot of time today.

Alright. So with that being said, I’m gonna pivot a little bit. So let’s dive into a fun little visual that I have related to, uh, the customer three sixty. Let me stop sharing. I’ll share a different screen.

Alright. So, hopefully, everyone can see my screen. When I look at Data Cloud for marketing, specifically with architecture and orchestration, I’m really gonna cover these subsequent visuals and the viewpoints in these two sections. These are my personal opinion of how we should be focusing and looking through the customer three sixty lens. And that customer three sixty lens looks at both sales service or what we call Salesforce core, marketing cloud engagement, which used to be just marketing cloud, marketing cloud personalization, which formerly known as Interaction Studio, data cloud, marketing cloud intelligence, which is formerly Datorama.

Now my main focus in this as we go through each one of these little bubbles and subsequent slides or subsequent screens is how I believe these platforms can be implemented in a well constructed manner that allows your clients or the customers, uh, to scale with the customer three sixty overall platform. Now this is starting to be especially important as customers are starting to purchase more of the Salesforce technology stack. And as Salesforce consultants, um, from SI partners, including myself, are being asked to handle more multi cloud engagements.

So let’s dive into it. And that first piece that we wanna dive into is data architecture, which my bubble comes quick. Alright. There we go. So much of the platform that we know about or we have been working through in the past has been facilitated via the person contact ID that is generated between sales and service, the first bubble, um, and Marketing Cloud. This ID allows us to facilitate synchronized data extensions or object replication between these two platforms via the Marketing Cloud Connect, which is my fun little arrow, um, to Marketing Cloud. This allows functionality to be brought back from sales and service to Marketing Cloud, for example, suppression of journeys, they have an active case, or from Marketing Cloud to sales and service to provide agents with engagement data on their specific records. In the past, we used to really stress how critical it was to stand up sales or service so the person contact data is flowing through, and these individuals are assigned the correct person contact IDs. Given that this ID was, in the past, the only immutable ID generated within the broader Salesforce platform that could be used across multiple clouds to facilitate the brokerage, it was mighty critical. But we now have this new platform in the middle, Datacloud. And this new platform is starting to re it’s going it’s starting to take what we think of as the past and break the established mold. It makes us think, how can we start to build out the customer three sixty without just following the legacy path of sales, then marketing, and then data cloud? And then don’t get me wrong. So I I wanna make sure I I hit this over the top of the head right in the beginning. I am not saying that the Marketing Cloud connects, that the person contact ID is no longer the right path to facilitate brokerages between marketing cloud and sales for service. Instead, what I’m saying is that data cloud is gonna start introducing efficiencies that change the way we deliver the c three sixty and how we have usually delivered engagements between core marketing cloud and other systems.

Alright. I’ve said my my little mantra. So next piece. When data cloud is brought into this architecture, we start to find unique ways in which our data model designs should change in market. If you haven’t seen earlier sessions throughout Marjorie, even I don’t remember who these exact speakers were, but they talked about the Salesforce consent management framework. And if you don’t know a little bit about that, go learn that today. That party model and specifically the individual and consent objects are critical to future success with data cloud plus sales or service plus marketing cloud when they’re all combined. Now I’m not gonna cover those party models in more detail. That could be its own 45 session. So let’s just go back into data cloud to open up let’s go back into data cloud to see where this is opening up some interesting data avenues for clients.

So as I look at data cloud and you can see the bubble on these these things connecting together, uh, there was a really interesting analogy I heard last year from Derek Ellis, a director at Salesforce, a data cloud success architect. When he presented on data cloud, it really stuck in my mind. He used the term data cloud is like a Rosetta Stone for marketer insight. Loved it. Really, we’re trying to translate all these different data models and source applications from either external sources, sales or service, from marketing cloud, and source these fragments of consumer data our clients have in their own common fragments. We harmonize them into a single data model. That’s CIM model within data cloud. That then allows us to create extrapolations on top of that new standardized model, get better insights, and discover trends.

So when we talk about data cloud and then we introduce working cloud personalization as well, we’re really talking about two different types of problems we’re looking to solve. We’re starting to say data cloud itself is a system of insights. We’re trying to collect insights across different systems of record by capturing and harmonizing data, unifying that data, and then segment into specific consumer and customer segments. If you remember going back to my slide decks a little bit of ago, it’s trying to get those marketing sales service moments to create that connected experience. Now personalization, on the other hand, is designed around capturing real time behavioral engagement data in the moment it occurs on a client’s digital property or via digital market campaigns. Now both of these platforms, we look at data cloud and personalization, work in harmony to capture critical data about our clients, customers, and consumers, which allow them to have a complete view. Now I do understand, and I’m not, you know, living in a bubble here, that some pieces of data cloud and marketing called personalization platform may start to seem similar. But in my opinion, they should be combined to provide that holistic system of insights and engagement. Ideally, we’re talking about if you bring both of these platforms together, we’re talking about behavioral engagement data from MCP, passing it into data cloud to build even more robust, uh, profile that unified individual, which then opens up even more use cases, use case avenues across the data cloud platform, the marketing cloud platform, and the personalization platform, and then, obviously, vice versa as data flows in between.

Now if you’re following me and you you’ve hopefully followed the diagram so far, you’re probably like, that’s a lot of bubbles and arrows, Jacob. Well, there’s more. At this point in the architecture die you know, discussion and diagram, we need to start discussing MCI or marketing cloud intelligence and its value within the c three sixty. So MCI shines when we’re looking at the effectiveness of media spend, return on investment from marketing campaigns. MCI can pull in emails, Salesforce marketing cloud emails. It can pull in data from personalization. It can pull in data from other disparate sources as well. And we’ve started to see clients aggregate all their sources within MCI from an analytics perspective because there are powerful machine learning capabilities within the Einstein functionality features that allows us to say, how are we performing towards certain KPIs? Now an established use case that I’ve worked with a lot in the past is, for MCI, is utilizing the platform to help clients understand data quality and build out a structured, consistent campaign tracking taxonomy. So the whole point of using this marketing analytics platform is to enable valuable insights across channels, touchpoints, and platforms. This cannot be done unless the client and their agency partners are aligned on a standardized, consistent structure. Now success with MCI. It usually allows, most of the time, the ability to harmonize data historically within the platform UI, which means clients have a better way instead of custom design hacks to analyze year over year metrics when transitioning reporting systems.

Alright. So you followed me so far. I’ve introduced more bubbles. I’ve introduced more arrows. Great. Let’s ground ourselves on three main areas of this diagram before we go into an example campaign structure. So at the highest level, I’m trying to explain out and this is my own little fun diagram design of the insights, the experience, and the engagement layers of how the c three sixty works together and how each one of these platforms can conform. Before I go into the little details about each one, I do wanna make sure that everyone understands, like, there’s this is the perfect state enterprise client who’s buying everything. There are going to be plenty of opportunities where data cloud works with just sales or service, where data cloud just works with marketing cloud, etcetera, or there’s there’s no data cloud involved in the middle. But we like to think through and design out a solution that looks for the future state in building out a constructed future state to bottle and design that allows us to scale with the platform. Alright. So I said my little mantra there as well.

Now when we look at the insights layer, I I look at this in in three main ways, connect, prepare, and model. So the idea here is we’re wrangling variety and volume of data in this layer between data cloud, um, and its engagement and transactional data that’s streaming from Salesforce core. We’re preparing it. So we’re trying to structure, clean, transform it into variables that we can use for predictions. And then we’re trying to model it out to say, how can we select only the data features required for a specific use case to develop predictive models?

In that middle layer or the experience layer, we’re starting to say, what is the audience? So how are we planning and segmenting the customers for target activations based based off data cloud data or just how we have built out marketing cloud or personalization? Then how are we gonna action? So how are we going to leverage AI automation to deliver personalized engagement? We wanna be able to get to that one to one level, but if we can’t get to one to one level of communication, we wanna at least be able to get to one to many.

And then finally, the engagement layer. The engagement layer to me includes post the customer, MCI. It probably could conform itself around, but I’m not that great at designing diagrams here. But what we’re saying in the engagement layers, we’re trying to deliver data driven personalized experiences that move beyond batch and blast. It hurts my soul when everyone says batch and blast, And into a more connected customer journey, uh, with the goal, again, of optimization and intelligently updating campaigns to service customers’ needs.

Alright. Hopefully, that makes sense so far. Yeah. And the architecture design for March is starting to get, uh, folks thinking a little bit more about that.

Let’s dive into segmentation and orchestration. So I do wanna walk through one specific example. So let’s discuss this in the bubbles in the top. We’ll move from left to right. As we think about engagement data for marketing cloud, it becomes important to understand the type of engagement data that, you know, our clients or your customers are using to show intent. You know, for years, we’ve used this idea of open time email or sorry, opening of emails themselves. Uh, but we also know that with the new Apple and Google privacy changes, the likelihood of relying on only open engagement is starting to go away. So we started to transition clients to think about what are the intent metrics that people are doing? Filling out forms, clicking links, providing client additional information via progressive profile, etcetera. All of which we store within marketing call engagement, marketing cloud personalization, or a combination and pull into Data Cloud. We then use this data and insights to understand the customer engagement that has occurred. It’s extremely important when we build out a robust profile of an individual within Data Cloud.

So the next two bubbles engagement. The next two bubbles between service cloud data, which is case data, and propensity modeling, let’s talk about it in this way. So for service data and sentiment data that comes through, as an example, like an AppExchange partner like Medallia, uh, within service cloud, it’s very important to understand how our clients are looking to interact with their customers. For this specific example, we can look at maybe the open case or they don’t have an open case. But we also look at examples such as, you you know, a retention journey where the customer has a recent case with an x number of days, which should cause us to potentially build out a different profile for that unified individual or segment within Data Cloud. On the propensity modeling front, hopefully, this doesn’t get too long winded, but where we’ve seen it is most clients we’re working with may or may not have some level of data science happening at the enterprise level. Um, now we commonly see Snowflake here ideally built on top of AWS or GCP, Google Cloud Platform. We’re also seeing Azure and Databricks in here to help with these propensity scoring. But as we think about propensity to buy or churn when looking at a retention journey, it’s important that we understand how that how that value is being calculated from that source system or source systems, if it’s possible, where at least the values that are being provided so we can be more prescriptive as to how we would be include them in a segment within data cloud. An example. For this one, I say propensity to buy greater than than eight. Um, that propensity to buy greater than eight, that could come from their specific data lake or warehouse. Again, this could go many different ways with a bubble.

Final one, we’re gonna say maybe a point of sale system, a customer database, or that could be product telemetries, you know, from another system. Again, put your own touch and view as as you work with clients looking to go into data cloud. So this could come a couple different ways. This could be product telematics. So I say two x per week of that user who’s using the product. So from a manufacturing lens, that could be who’s actively using the product two more times. If it’s an automotive industry, that could be pulling telematics for a specific service or even selling telematics as a service if the client has a b to b, you know, side. So let’s then touch on the kinda critical component with the source examples, which how much data would we be gathering from that point of sale system or any systems and bring it in the data cloud. So be careful. Uh, we do not want to throw all and every one of our data points into data cloud if it doesn’t meet a specific use case. So, again, be careful to look at and understand what data is being factored in and relate it back to a use case.

Alright. So now we have the sources. We’ll start to create that unified profile. But it’s important for us to think through, like, how would we build that, uh, profile within data cloud, and how are we going to get it towards the end state goal in mind, which is we wanna activate it as well. So, typically, we’re looking at a couple of different things. So from a cross sell, upsell perspective, we’re looking to take what we know about this customer, so that unified profile, and suggest a complimentary product for them. Maybe that’s some additional it’s like, there’s some additional examples we could look at here. And I’m trying to use examples from all industry verticals to connect with folks is let’s use Fins now as an example. So let’s look at a cross selling and mortgage as an example here, which is we build out a segment within data cloud. Maybe not exactly what I’m showing on the screen, but have they do they have a checking account? Have they logged into mobile banking or their web? And do and do they actually have a mortgage already with us? If we can do that, we could start high level and start building down a segment to then activate to engagement or personalization. Now the reason why I keep belaboring this section for a unified profile is we wanna start high level because we can always get more granular as we build out data cloud. So with that mortgage example I just explained, we can get down to the regional level or suggest other products if we need to get more granular. But utilize this kind of idea and thought process as you work with clients and with your customers on how to dig deeper into each segment until you get to a level where the segment membership is, uh, too diminished to provide value. So, hopefully, that makes sense.

Alright. Let’s talk about the next two bubbles, which is the activation. And I’m doing okay on time, so I’m I probably rushed a little bit through this section. So marketing cloud engagement. I think we all hopefully understand marketing cloud engagement. The idea here is we’re going to be pushing segments and activating them down from data cloud into marketing cloud, which stores itself as a data extension. So there’s going to be some post processing SQL work that’s usually needed within marketing cloud to make sure we’re checking consent because you need to be able to understand, do we have the right contact point activated to the correct view so we can make sure we’re double checking, can this person can this person receive the email, have they not opted out, and what is their preferred preferences? At the enterprise level within marketing cloud, there’s you know, that’s check it all subs that we always have as our final safety net, but it is important to know this as you look through data cloud and work through it, which you will probably need to do some last mile checks.

On the personalization side, when we talk about activation, and I’ll go through this very quickly, a lot of times we’re looking to better understand our segment bases or our unified individuals within, uh, personalization with data cloud information. Because data cloud is going to, most of the time, always have more level of granular information than what we’re going to be capturing either from a website, a mobile app, an unauthenticated site, or authenticated portal with an MCP. But if we can augment our segments within MCP from that robust dataset, in data cloud, you’ll be able to get to that much more granular of a level of a one to one communication on web in real time.

So then near the end of this, no campaign in modern marketing is effective unless we measure the effectiveness against the hypothesis and a meaningful KPI for your business. Now it’s vital to have your conversations very early in this process to clarify your objectives for these campaigns and the KPIs for the year that we can measure within this whole design. Now if we can get multiple themes even at a macro level, we can then start to understand how these experiences and KPIs can be measured within MCI. And I threw in here as well if MCI is not the platform of choice for your client, how could we visualize it in Tableau to drive the as they go in mind? And my last arrow is that talks back to data cloud. So we always wanna be doing this full three sixty of understanding the use case, are we meeting the need, and then how do we tweak, iterate, and design.

Alright. That’s enough of bubbles and arrows from Jacob. Let’s jump back to the slide deck. Click in Windows. Click in Windows.

Alright. Hopefully, everyone can start seeing my screen again. So we’ve we’ve talked about the three sixty. Now let’s get into some best practices from a data cloud perspective.

So one of the first things that I always like to talk about is begin with the end in mind. I think you’ve heard me explain it a couple times in the diagram and other areas of this presentation so far is we wanna be able to understand what are the pain points and the solutions that we’re trying to get to from a client’s perspective. My own personal mantra here is, uh, I believe that by starting with the user’s needs, their goals, their motivations, and working backwards to the solution, you increase innovation, adoption, and return on investment of the technology solutions that you’re working with. And so how does this work? At the highest level, it looks at people, business, and technology, or I also like to refer to it as people process and technology with our center point being a solution, but looking at data at every moment in time along this continuum. So you’ve probably heard or seen this, so I’ll go through this quickly. We like to start with the people. So what is our alignment? What are our goals? What are our ideal and non ideal customer experiences? What’s the team expertise internally? From a business perspective, what are the value drivers? What are we analyzing? And what are those business requirements that then lead us into, can we feasibly do it with the technology at hand? Will we have to extend the technology from platform best practices? Will we have to include some third party apps from the AppExchange or other tools out in the market? Excuse me. Or will we have to do some custom development? And at the end of the day, we’re always looking to say our solution will power some impactful moment in time.

Alright. That’s my little people process and change moment. Let’s get into something interesting when it comes to these six steps that I recommend when implementing data cloud itself.

First one, why and what? So initially, you’re well, initially, at the beginning of every implementation, you should be going through robust, uh, discovery and understanding what are the initial segments and audiences that we wanna build. Again, I’m gonna belabor the point. What goal do we wanna have in mind, and how are we going to get the right segments and audiences to meet that goal?

Let’s talk about taxonomy, which is how do we collate necessary attributes required for segmentation, uh, that align to the cloud information model? And then how do we identify gaps and design the extended data model within data cloud and go beyond just the out of the box bundles and model itself.

From an auditing perspective, this is where the most projects are not as sexy or interesting, but we do have to explore the the data sources available. Uh, most of the time, we’re trying to explore as many of those as possible that align to the user experience use cases, but you do need to identify the required feeds, the existing data relationships, and and to be established as part of the integration. But other pieces, how is this latency going to work with data and which source is being updated at what time, which that gets to the profile strategy perspective. So if we can clarify those data points and need to be consolidated into that unified individual, we also need to understand the hierarchical order of the resolution reconciliation rules to tie back into the audit to say, which source wins? If we know our source of truth of data is coming from sales or service, but we also have three other external sources that are going to populate through, which one wins during our unification and reconciliation?

Next is computations. So what are the necessary calculated insights that can address this segmentation and activation requirements? Um, these are a little tougher. Sometimes you do need to get some of the data into data cloud and start looking at the gaps, uh, when you walk when you talk about the data audit and the profile strategy. But this is usually where we say, alright. What missing pieces could we have gathered? Hopefully, we haven’t gathered or missed anything. How can we then use data cloud to power and solve for those missing moments and and collaborate and build out more granular attributes on the unified individual level?

And then finally, because I haven’t said value enough, let’s hit you over the head one more time with it. Clarify your KPIs for use cases, establish operational operational reporting. How are we gonna visualize data? How are we gonna report on it? And then how are we gonna provide the recommendations for the return on investment measures and gaining audience insights. Again, at the end of the day, we have to prove out the data cloud is providing the needs that the client is looking for.

Alright. So then let’s dive into next section, which talks a little bit about a high level implementation approach. Now let’s talk a little bit about reviewing and knowing what you should start with before you dive in to clicking buttons into engagement. Excuse me. Just getting over pneumonia. Super fun.

Alright. Let’s take it in this way. First, we need to establish an individual ID that is unique, enterprise wide, or was represents an individual. This might be CRM contact ID like we talked about before, particularly if our client is using Marketing Cloud Connect, where they’re trying to expand off of an established Marketing Cloud instance, or this could be an MDM ID. There is an interesting use case out there where MDMs are providing that golden record directly into data cloud, and data cloud is then actioning off of it because MDMs are not built to action quickly. The other item here is, yeah, consider attributes and the data sources that need to be ingested into data cloud to support the client’s marketing efforts. Again, align these attributes with the standard data model or identify those gaps to extend, uh, to extend the data model to a more custom model itself. Think through your primary keys. And remember that you will need a primary key for each DMO that you’re mapping to inside data cloud. Identify formula fields for any transformation that will make the date will will make data cloud easier for the client’s marketers to use or help with standardization. Uh, it’s good to know this before you start streaming data in to the platform. And then lastly, do not forget that there are required mappings that you’ll need to do, um, that you need to do that are relating to identity resolution activation. This means you will need to make sure that customer identifiers are mapped to the individual ID, party fields, and the contact point objects.

Let’s dive into that a little bit further in this next section. So this can be a confusing topic and could be its own session outside of what we talked about today. I explained that a little bit in the Adobe InDesign, uh, document earlier. Now these items do sound semantically the same, but in fact, they’re quite different in use. So we have two different ways the word party shows up in the platform. One is the word party as a field typically on DMO, and that field is actually representing a relationship back to the individual that has managed the data cloud relationship model. Example here could be the marketing cloud connect or sorry, marketing cloud contact key. That is a party value that has a relationship back to the DMO itself. The second place you’ll see party is in party identification. This is a separate standalone data model object and used primarily for identifying resolution purposes. This is a way for data cloud platform for the data cloud platform to take a common unique identifier that might exist in multiple systems and leverage it in our match rules. Now remember, take time to learn how these fit into the Salesforce CIM model and your own clients’ model itself, and then how your clients could potentially leverage these objects within Data Cloud.

Alright. I think I’ve belabored a couple of implementation approaches. Let’s see. Let’s get into approaching a user experience use case.

So I have a six step method that we’d like to talk about when we say getting started with Data Cloud. Give me one second drink. First piece we’d like to discuss, so, again, hitting you over the head with these these things multiple times is start with that user experience use case. But think of these three bubbles as you’re starting to discuss it. So I bolded the the piece here, which is decide on key value drivers, but you also need to understand what is that experience. And if you look at a full customer journey map, where are the pain points that happen throughout a customer’s experience with your brand or with the company that you’re working with?

And then how are you gonna measure it? What data do you need? What sources? What attributes? Do you know which ones are gonna be the source priority and most important? Well, it has to be cleansed and transformed before or during ingestion within data cloud. Very key point there.

Where’s the sort? Is it in a single CRM work? Is it in multiple works? Is the data in a warehouse, etcetera? With the big key point here is, is the data structured or unstructured? You can you can accept both in data cloud, but you do need to understand the level of effort behind it.

Then how are we gonna match this together? What data are we gonna use to connect this altogether? Is it gonna be an email? Is it gonna be a customer identifier? Make sure to focus on matching by more than a couple of contact points. That increases the data quality itself within our match and reconciliation rules.

Last one second to last piece here is, are you gonna roll the data up into segments? Uh, the bolded section I have here is think about timing, uh, when it comes to segments, when it comes to near real time or batching. Uh, we have to align that to the use cases and how data latency occurs from source systems in the data cloud, identity resolution, and then act action as itself. And then how far you wanna go back with the data, and then what do you currently have, and then what are you gonna need to create that use case?

With the final goal in mind of where are you gonna activate this thing to? It’s great to have all the data to live in data cloud, but you do wanna action off of something. Is that gonna be an email journey within marketing cloud or within account engagement? Is that personalizing your website via marketing cloud personalization? Is it creating a service case in CRM so the service rep or agent can better understand where this person was before they called in? Um, are we unifying that patient profile in the CRM, again, to enable that agent or sales rep or someone? And then finally, one that I find very interesting is, are you gonna trigger flows between connected CRM orgs? Data cloud can trigger something that occurs in one sales or service org to another sales or service org and have those two start to talk to one another. So there’s extreme power in that.

Alright. Let’s put it into practice real quick. So I think I my example here, I think, I wrote it out. It’s on the screen. Uh, we currently handle and this is specific example, it’s for service, um, and handling specific calls. So in this example, we’re saying, we currently handle too many calls at the call center around failed logins to our online portal, which has taken time away for more serious issues that our team should be focusing on, especially when it comes to cost to serve.

What data do we need? We think based off of this high level conceptual, uh, use case that we’ll need some online portal data to know where the customer has filled log in. We’ll need some customer data related to their name, address, email, phone number, phone data to reconcile who they are. And we’ll probably need some website data to understand if and when they were on our website, maybe on an FAQ or we’re trying to figure this out on their own.

Then we start to say, where’s the data stored? It’s probably stored maybe our online portal platform. Maybe it’s gonna be in some of our CRM instances, maybe one or multiple. And then, potentially, it’s, uh, stored within marketing hub personalization platform if we have that available to us. We’ll probably try to match it with email, phone, customer identifier, last name, online portal ID as an example. We’ll need the date that they have their failed login. We’ll need to know whether it’s on a PC or mobile device depending on how we can, you know, troubleshoot and help them. And this way, we can create that segment called failed login.

Then let’s activate it. So probably activate it from a digital campaign’s perspective to a website pop up, maybe to help deter from a cost to serve perspective, push them to FAQ documents so they don’t call into the call center. They will do some email journeys or an email journey that’s actually sending out an email that says, need help logging in or mobile text depending on the preference of the end state individual and how they wanna be communicated with.

Alright. So that’s enough of Jacob’s TED Talk for today. Let’s take some key takeaways.

I think my biggest one is, uh, don’t be afraid to start. Uh, I don’t know how else I can say that other than dive in. Let’s start trying to design out and see what interesting use cases we can provide for our clients and for your customers.

Uh, do not underestimate the importance of user experience use cases and proper data discovery when discussing data cloud.

Uh, I touched on the operational excellence for only one slide, but that could be its own session as well. Understand the people in process. Effective communication and coordination and coordination leads to success with any technology implementation.

And then what I like to say is test and learn, which think about this as the beginner’s mindset. Don’t be afraid to rethink tribalized knowledge and tribalized processes that are occurring either in your customer or within your own organization.

And then finally, be nimble. Plan for ongoing change, and don’t get caught in my favorite term, analysis paralysis.

Alright. So, hopefully, that all makes sense. I thank you all for this session. I’m gonna go double check now into the q and a. If anybody has any questions, please feel free to post them now, and I’d be more than happy to answer them. If not, um, a little bit ahead of time, which is good. Alright. Don’t see anything in the q and a at the moment. Don’t see anything in this. I’ll stay on just in case anybody has any other comments. But if not, I’m a great consultant. I gave you five minutes back, and I was right on time. So I’ll sit here. Feel free to ask any questions. If not, thank you again for your time. Appreciate you guys listening to what I had to say. This is the fun, awkward moment where it’s just like Jacob will dance. No no problem. Thanks, Kathy. Thanks, Patricia. Thanks, everybody, for for the quick little comments in the session q and a. Alright. I’m gonna stay on for one more minute, and then I’m gonna stop, and we’ll leave stage. Awkward silence, my favorite thing ever, Carol. It’s my favorite. I guess I should have prepared my own, like, mock interview q and a questions with Jacob. I thought there might be some interesting ones. That’s fantastic. I’ll stay on for another forty seconds, and then I’ll I’ll say goodbye and appreciate you guys’ time. Alright. Well, thanks again, everybody. Enjoy the rest of my dreaming, and feel free to reach out if you ever need anything on LinkedIn. Slides will be provided, and there’s a couple of ways to communicate with me. Thanks again.