Today on the Salesforce Admins Podcast, we have another deep dive with Josh Birk, who talks to Shoby Abdi, Principal Architecture Evangelist at Salesforce. Join us as we chat about the Well-Architected framework, why you should start using Data Cloud, and what you need to do to get ready.
You should subscribe for the full episode, but here are a few takeaways from our conversation with Shoby Abdi.
Shoby is the Principal Architecture Evangelist at Salesforce, and he’s come on the pod to tell us about the Salesforce Well-Architected framework. And he’s here to tell you that the framework is not just for Salesforce Architects—everyone should take a look at it.
As Shoby puts it, if your customers or users open Salesforce and it scares them, that’s how you know you need to use the framework. It can help you identify anti-patterns in your org and fix them by building solutions that are trusted, easy, and adaptable. Those are the essential ingredients for a healthy org, and that’s what puts you in the best position to take advantage of everything Data Cloud has to offer.
We’ve talked a little about Data Cloud on this podcast in the past, and that’s linked below, but Shoby helped explain why it’s such a big deal. The main point he makes is that by bringing all of your data together in one place, it can start to tell a story. As he puts it, “What is your customer doing? Where are they? And then how do you interact with them next?”
When we talk about Data Cloud, it can often sound like it’s some sort of parking lot for all of the information you have about your customers. But as Shoby points out, it’s so much more than that. It’s more like if your parking lot had a valet that could tell you everything about who’s parked there, what kind of car they drive, how long they’re staying there, and exactly when they plan to leave.
If you’re a Salesforce Admin looking to enable Data Cloud, Shoby recommends working through the Well-Architected framework to make sure that your org is healthy. After all, these tools are only as good as the data you give them.
If you’ve done all that, there’s a lot to get exited about. Data Cloud brings everything into Salesforce as standard objects, which means you can build all sorts of interesting things with flows and everything else you’re already using. And with Einstein Copilot and other AI tools, the possibilities are endless.
There’s a lot more great stuff from Shoby in this episode about how Data Cloud works, so be sure to take a listen. And don’t forget to subscribe to hear more from the Salesforce Admins Podcast.
Josh Birk:
Greetings, Salesforce admins. It’s Josh Birk here, your guest host for this week. This week on a deep dive, we’re going to dig into data cloud and then talk about data cloud. I brought one of my good friends and old colleagues, Shoby Abdi.
All right. Today on the show we’re going to welcome back Shoby Abdi to talk about data cloud. Shoby, how you doing?
Shoby Abdi:
Not too bad. Just doing well. Connections is right around the corner. Just to timestamp this for everybody.
Josh Birk:
It’s a Connections week. It’s a DevOps streaming week. Everybody here is in Chicago and I’m still at home talking to you. So that’s my social calendar.
Shoby Abdi:
Exactly. That sounds very similar. I am also in Chicago.
Josh Birk:
Nice. Okay. So introduce yourself a little bit. Tell me a little bit about your current role.
Shoby Abdi:
Sure. So my name is Shoby Abdi. I am a principal architecture evangelist on the Salesforce’s Architects and Well-Architected team or the other way around. Essentially, if anybody has ever gone to or if you’re a Salesforce architect or if you’ve ever gone to architect.salesforce.com or if you’ve ever heard of the Well-Architected framework, my role, along with a few others is that we are evangelists. So it’s very similar to the work that Josh does or on the admin side or developer advantage to the developer size, except our audience is firmly architects. Architects of all kinds, whether that be our internal ones, customers, partners, architects all over the place.
Josh Birk:
There’s a nice overlap between admins and architects. So actually let’s give you that this is the Tonight Show portion where you get to do your own plug. Tell me a little bit about the Well-Architected framework.
Shoby Abdi:
Sure, absolutely. So the Well-Architected framework in a very simple manner is prescriptive guidance for solution health. When we put it that way… Now remember, it’s a Well-Architected framework. Now, Well-Architected frameworks exist not just with Salesforce, but Amazon, Microsoft, all kinds of other cloud providers have created their own Well-Architected frameworks. What’s really important about the Well-Architected framework is that it’s not simply for architects, right? While it’s Well-Architected, it’s not a well-built framework for architects. It’s a Well-Architected framework.
So the idea is that regardless of your role, your function, if you’re just a customer, if you’re an implementer, everybody should be looking at it. And when we talk about solution health, really what it boils down to is a lot of times when anybody is implementing for the first time, or they’re implementing in an existing org, or even at as an administrator, you inherit an org. You start a job and you inherit some eight, nine-year-old org with all kinds of technical debt happening and you’re like, “Okay, I don’t know what’s good and what’s bad, what to keep and what to deprecate.”
What the Well-Architected framework does, and when you look at it from a baseline perspective, is that it was really developed and designed by architects and other individuals in the ecosystem like yourselves, like the folks listening. It was also firmly designed by our individuals within our product and engineering team. And that’s important is that the prescriptive guidance within it is really firmly built on top of not just base level knowledge, but it’s firm recommendations, firm practices, and we even take the tack of actually dictating specific patterns and anti-patterns.
Josh Birk:
Got it.
Shoby Abdi:
Related to how it goes. And the way that Well-Architected framework is structured is around solutions should be trusted, easy and adaptable. And we firmly believe they should be in that order. One of the first elements of trusted is secure, but then easy, it’s about being intentional and then adaptable. One of my favorite, one of the last ones is composable. And obviously as Salesforce, trust is our number one value. So obviously we’re going to feel trust should be the first thing you focus on.
So really the framework itself is really a… The simplest way I can put it is that if you want to build a healthy solution, if you want to maintain a healthy solution, you can utilize a prescriptive guidance within the framework and you can find it at architect.salesforce.com. It’s available today. It’s not hidden anywhere. You can find it today and utilize it as you need it.
Josh Birk:
Love it. So if you’re an admin and you open up your org and it scares you, this is a good trail guide to get out of it.
Shoby Abdi:
It’s not even if you’re an admin, it often scares you, it’s more of like if you’re a customer or if a business user open up the org and it scares them.
Josh Birk:
Right.
Shoby Abdi:
It’s like, “I’d love to use Salesforce, but I really can’t use it because of the following SB issue. My report-“
Josh Birk:
Because of all the anxiety it gives me, whatever.
Shoby Abdi:
“I really want to use this report. I don’t know where it is, but I really want to use this report, Meredith, but it takes five minutes to run. I can’t use it.” Okay, well, it may be full of anti-patterns. So that’s where utilizing some of the guidance within our performance, topics, around throughput and latency could come in handy to really understand, are these reports I need to modify? Do I need to do skinny tables? Do I need to optimize any queries if it’s custom? That’s what that guidance really is.
It’s not simply an admin even or an architect or developer or anybody looking at it themselves. It could also be based on actual customer and end user feedback of what they’re seeing and what they’re telling you that needs to be fixed.
Josh Birk:
Because that’s the commonality between these three roles is that the users are always in the middle.
Shoby Abdi:
And Salesforce.
Josh Birk:
And Salesforce, yeah.
Shoby Abdi:
There you go.
Josh Birk:
Okay. So we’re talking data cloud today, and let’s take the very, very high level. If you have to introduce data cloud as a product to somebody, how do you describe it?
Shoby Abdi:
The way that I describe data cloud is that when it’s so much like you hear a lot of different capabilities around it. You’ll hear data lake, data warehouse. Really what data cloud is, is it provides the ability for any organization, regardless of the amount of data sources that your organization has to make those data sources and all of that data, that disparate data that can be everywhere doing many things. It makes it useful in the context of your customer.
If you’re a B2C organization that has end customers out there in the world buying things, then it could be useful as far as a unified profile for that person goes. As you’re a B2B organization where you’re more business oriented and account oriented, it could be useful for you as well. I was actually describing this to someone yesterday where when you look at data cloud and its capabilities, and a lot of people will always think of different data sources, data lakes, data warehouses where it’s different and important is that data cloud is not simply a place where you park your data, hope it works, and hope you can find it. It’s not a parking lot for your data. Right?
Josh Birk:
Right.
Shoby Abdi:
What it is when you bring it into data cloud, you can automate it. You can get a view of your customer. Right?
Josh Birk:
Yeah.
Shoby Abdi:
When we talk about that single view of the customer. And really what drives that view is the customer’s interactions and activity. Whether you’re a consumer based organization or a business to business based organization, your customer is going to interact with you in some way that involves data. They’re going to be creating data, whether they’re clicking on websites, they’re paying invoices, they’re paying credit cards, they’re creating cases, they’re telling you that they want to buy 500,000 of these and you’re closing a deal and they’re paying off an invoice and agreeing to an order over email.
All of these actions, all of these functions, all this data as it comes into data cloud starts to tell you what are the behaviors? What is your customer doing? What is your customer doing today? Where is your customer? And then how do you interact with them next? So that’s the power of data cloud is, it’s not simply a, “Okay, I’ve got all my data in. That’s great.” No, that’s literally step zero.
That’s what we tell everybody. Get your data into data cloud. Just get it in. Because once you get it in, then it’s the power of automation, analytics, unification, segmenting, activations. That’s where it starts to matter. So that’s really when we talk about the data cloud, that’s what we look at.
Josh Birk:
Yeah. I love the parking lot analogy because first of all, it’s so easy to think of it like that. I’ve got my data in Salesforce, but then blah, blah, blah, blah, blah. There’s something over here. I’ll use data cloud to connect it. But to kind of extend that analogy, it’s not just a parking lot. It’s a parking lot with a valet and a security officer and a scheduling system. It’s got all the moving parts you need in order to efficiently move cars in and out of that parking lot.
Shoby Abdi:
Yeah, absolutely. It’s not just lot covered. So really what it is, it’s similar to everything that Salesforce does. Like I said before, trust is our number one value. Similar to the Well-Architected framework as well, it’s built with trust in mind. It’s built with security in mind. So because we have that trust first. Now, when it comes to data cloud, you can be secure that when your data comes in, we handle that. We can take care of that part. Then what starts to matter is what you do with it like automation.
Josh Birk:
Yeah. What are some specific use cases that you see? What’s an example of somebody who just clicked and they’re like, “Oh, this is actually the problem that I need data cloud to be the solution”?
Shoby Abdi:
So a lot of it really comes in terms of, okay, data cloud obviously started from its CDP days and it’s evolved significantly. A lot of use cases that we’ll see are simply around, “Okay, I need to understand how do I interact with my end customer next?” So principally you’ll see some of that around unified profile. If you’ve ever seen a demo of data cloud, you’ll see first use case of, “Okay, who is this human being? What websites are they going to? And then finally, what is it that we need to really work them with?”
And then people often ask like, “Okay, I’ve got that element of data cloud and I’ve got all this data in. I see this person, but who’s going to help me next?” That’s where our friend Einstein comes in handy, right? Because you can create an email within Einstein utilizing different prompts, different capabilities, but once you include data cloud data into that prompt, now all of a sudden you’ve enriched it significantly. Where now because CRM data, while powerful is often limited.
It’s limited based on what your business does and how your business updates still that CRM data, whether it’s case data or opportunity data. But data cloud data is your customer’s data. Right?
Josh Birk:
Mm-hmm.
Shoby Abdi:
They’re the ones who are constantly making it happen. So now all of a sudden if your email or if the campaign is a part of, is a reflection of actual real life activity they’re doing and they didn’t need to tell you, “Put me here next,” that makes a big difference in that process. So that’s fundamentally a use case that we see. It’s like while it’s very general, it could apply to marketing, it could apply to sales, in the end, the reason I articulate this specific use case is that it starts to really combine the power of the data cloud with the power of what you probably already have if you’re looking at data cloud, which is your Salesforce CRM.
Josh Birk:
Yeah. Well, and I like… Every now and then in tech, we get the handiness of a phrase that just makes sense. Unified profile, I feel like honestly gives you the solution, but also defines the problem in the first place, right?
Shoby Abdi:
Yeah, absolutely.
Josh Birk:
If I’m trying to communicate with you, Shoby Abdi, but you as a digital identity is actually strewn about 15 different data sources, then when I go to Salesforce, I have to cobble that together before I have an accurate representation of you.
Shoby Abdi:
No, absolutely. A lot of the challenge is, someone even asked me this… Yesterday was apparently a busy day, but someone asked me this yesterday as well, was just around, “Well, can you use data cloud for de-duping?”
Josh Birk:
Right.
Shoby Abdi:
And then I said, “One of the things you have to think about is de-duping is a you problem. It’s not a customer problem. The customer’s problem is you don’t know who they are. So if you’ve got eight different versions of that human being’s name and you don’t know which one is the which one that actually has bought anything, or maybe they’ve changed their name, they’ve changed their address, they’ve changed their phone number, but then all of a sudden when you bring in the data cloud, you bring in all those unified data sources and one of those John Smiths is clicking on a lot of your websites, ordering a lot of your product.
And the John Smith you thought that was in your CRM was the one that was actually doing it, is not that one. Then you really don’t have to de-dupe anything. Then you know which one is the actual human being, John Smith, like the ones you should actually start having that enriched discussion with, that enriched engagement with, whether it’s through sales, service, marketing support, however. That’s where you can start.
Josh Birk:
So data cloud’s not there to try to fix your bad data management, but it can help focus the lens, so to speak, so that you’re looking at the right part of the data.
Shoby Abdi:
No, absolutely. Right? Because by bringing all those data sources in, now all of a sudden your customer is almost in the data driver’s seat because as they interact with your online properties, as they interact with your organizational properties and data created, now all of a sudden their activities start to reflect within those profiles and you start to see and understand what it is they do.
Josh Birk:
To extend to that, how would you compare, because you and I are old school technologists and we grew up in the days where what we are talking about now would mostly be in the realm of extract, transform, and load, right? Like an ETL process. How does data cloud compare or how is it similar or different from those old school solutions?
Shoby Abdi:
Yeah. And I think that comes up quite a bit of like, “Well, how do I do ETL with data cloud? Do I go with this ETL tool or do I go with data cloud? Do I do this or do I do that?” And invariably the power… One of the things that often comes up is it becomes a… When you look at what data lakes we’re looking to solve and data warehouses were looking to solve was like, “Okay, we’ve got all of our data all over the place. We need and to be one specific place.”
And then it’s like, “Okay, we’ve put it now in one specific place.” And it’s like, “All right, now what? Now, are we all of a sudden… Once again, it’s the parking lot.” Are we all of a sudden like, “Okay. Well, this person’s claiming they need to leave the building at 9:00 PM. This person is leaving at 10:00. So why don’t we park their car up here and then with there and then go over there.” Right?
Josh Birk:
Right.
Shoby Abdi:
So now all of a sudden as a customer because of something I really didn’t do, I may have gotten there really early. But because I got there early, I’m getting penalized for leaving early because a lot of people came after me. So the way that I compare it with ETL, it goes back to the ability to create that unified profile, activate on that unified profile, create some [inaudible 00:15:45] unified profile, is that you start to worry less about how to get clean amazing data.
I don’t even compare it to MDM. I’m sorry ETL, but it’s MDM. How do you do data cloud with MDM? And it’s like, well, once again, who is it for? Is it for your internal organization to have its perfect golden record situation when in fact the golden record should be the customer? It should be that customer. It should be the individual account record or that individual record inside of Salesforce. And then what’s really driving it being meaningful golden are the interactions and activities that are happening.
And that’s where that power of data cloud comes in with you bring in from data sources through batch elements or through streaming elements, whether you bring it through our Insights API or any other API or an SDK. As you bring that data in, now all of a sudden you’re simply enriching that profile more and more and you’re able to automate on that more and more as it goes along.
So that’s why it’s hard for me to compare it to base level ETL or anything because the power of it is that just bring the data in. Bring the data in. Where with ETL it’s all about, “Well, before you bring the data in, let’s talk about at least 50 things.” Where now it’s like, “No, just bring the data in. And then utilizing the power of the data cloud itself, the power of automation built within our platform like Flows, and Apex, and Orchestration, others, you can do a lot more with it. But the first step, I’m going to go back to what the first episode will be on, it’s just got to bring it in.
Josh Birk:
You just got to bring it in. Right. Well, and let’s use that as a launching point because you just said automation a lot and you said Flow. So an admin who knows Flow, what are some things that they can use Flow to interact with data cloud?
Shoby Abdi:
Yeah, absolutely. And what’s interesting is a couple of weeks ago we actually put out a blog. So if anybody’s ever read our blog, medium.com/salesforcearchitects actually blog post about this very thing. It’s essentially, if you want to look, it’s called Making Data Cloud Work With Your Existing Salesforce CRM data. And while I’m not good at titles, the end goal of it was really to talk about when we talk about data cloud and Salesforce and our entire platform, our Einstein 1 platform, it’s really all about… One of the things that you’ll see is that shared metadata framework.
And that’s one thing if anybody takes anything away from whatever it is I’m saying, it’s really that all of this is a shared metadata framework. What that shared metadata framework means is that now as we just keep saying, “You bring that data in, bring that data in.” It’s not simply, “Okay, now that I’ve brought that data in, I don’t know what to do with it. Can I actually access it? Can I use the tools that I have in place in platform to actually do anything with it?”
Or it’s like a marketing cloud thing. I need to learn marketing cloud automation. So one of the drivers behind creating that blog post was to say, “No. Data cloud was built with the power of the platform in mind.” It’s the marrying of data cloud capabilities with Salesforce CRM. So when data cloud comes in, it comes in as different kinds of objects. It’s objects. So it starts usually from data streams as DSOs. And DSOs is essentially that data from all these different third-party sources. And this is where data starts to reflect within your data cloud environment, whether it’s zero copy or it’s registered.
It’s essentially there. It’s available to you to actually start to proceed with. That’s when you start to create essentially data lake objects as well as DLOs. And those DLOs is when you start to look at, “All right, what are those specific data use cases I want?” It could be sales-related, it could be service-related, it could be marketing-related, it could be telemetry-related, it could be anything. But then beyond those DLOs, once again, we’re still talking about objects. That’s when we’re going to create other kinds of objects.
So the first kind of object is usually what’s called a DMO. Now, if you’re an administrator or developer architect, when you’re working with data cloud, you’re probably going to be working with DMOs quite a bit because DMOs are more often than not, where a lot of the automation is happening. So a DMO is essentially to harmonized group in those DLO. So you could have those data lake objects. Let’s just say, one of the things about I always miss is those fun old school MarTech diagrams where they’re like, “Oh, let’s show all the hundreds of billions of kinds of applications that exist out there for marketing.”
Josh Birk:
Right.
Shoby Abdi:
It’s true. Right? You may have 55 Google Sheets and ADRPs and etc, etc. As you start to go down the source object, DSO, data lake object, DLO, DMO object in a DMO process, you start to essentially build out, parse out what the object model needs to look like, what that harmonized elements of data needs to look like for your specific use case. And I’ll go over a Sales Cloud, use case that’ll be relevant.
So for DMO, and then you’ll have another kind of object, which is actually a calculated insight object, calculated insights, which is a CIO, now that you can create basically based on essentially an insight based on a query, based on specific analytics like certain… Old school calculated insight or [inaudible 00:21:33] review. There’s a lot of different ways you can call it, but it is essentially based on a calculation that occurs where someone hits this website these many times, or they’ve made these many orders or that kind of thing.
Josh Birk:
Gotcha.
Shoby Abdi:
But those are the kinds of objects. So now all of a sudden, okay. You’ve got all these kinds of objects within data cloud, these data cloud objects. What happens then? So that is where the power of our fliend, friend. Not fliend, but friend.
Josh Birk:
Fliend.
Shoby Abdi:
Friend Flow comes in. So when we look at flow, it’s really the one thing we need to realize about data cloud is that a data cloud instance can exist in one environment. It could exist as a single data cloud instance across multiple environments.
Josh Birk:
Gotcha.
Shoby Abdi:
So if you’re an administrator who’s like, “Well, it’s not as easy for me Josh, because I’ve got 55 environments in this one data cloud instance, catering to all of them.” You could still use Flow. The way that it changes the way that you design it slightly differs. And this is obviously evolving, but for now, this is the way it is at the time of this recording where if your data cloud and your Salesforce CRM are in a single environment, the same environment, that’s where you’re going to create a data cloud triggered flow. And the data cloud triggered flow, if anybody has an activated data cloud org, or if you’ve gotten a data cloud environment from Trailhead, you’ll see the ability to create a data cloud triggered flow, and that’s launched when a DMO or CIO have a change of data. So now all of a sudden it’s like, “Okay. How do I then associate that triggered flow with my Salesforce data?” Well, you do it the way you do it.
Josh Birk:
You would do it normally.
Shoby Abdi:
You create a record or you essentially update a record or you do something, right? Because now all of a sudden that customer activity within that Flow can then drive whatever updates you want within that CRM object.
Josh Birk:
Gotcha.
Shoby Abdi:
So that’s one of the first ways you can correlate, or you can actually directly link your data cloud, data model data that’s being generated to your Salesforce CRM data model data.
Josh Birk:
Got it.
Shoby Abdi:
The second way to use Flow is that if it’s in different environments. Now you may say, “Okay, well it’s in different environments. Am I in trouble now? Now the answer is no. Once again, data cloud is designed for this in mind. Now, there are many ways to actually utilize a data action and data cloud, and what a data action does is very similar to the trigger flow. It looks for any data changes in the DMO or CIO, right? One of the specific ones, you could do web hooks and other means that work with all of our clouds, but the one that people probably most know is a platform event, because a platform event is basically another kind of object.
And that now all of a sudden you can create a platform event and then that can hit a flow within another org, a platform event, and then all of a sudden it becomes, “Well, now how do I associate that data cloud platform event, data to my CRM data in my other org?” Well, you do it the way that you would do it today, right?
Josh Birk:
Right.
Shoby Abdi:
You essentially do it the way that it would work with the platform event today if it came from data cloud, if it came from grandma’s Microsoft Access table, whichever one. So that’s where you can operationally use it, where in those situations, whether it’s the same or different environment, how you can use Flow if you want to create read, update or delete any and all Salesforce data.
Josh Birk:
Gotcha.
Shoby Abdi:
The other path that you can do is you can also… You have the ability to actually view some data cloud data associated with CRM data, and that’s relatively new, what are called data cloud related lists and data cloud copy fields. A related list is essentially the way it sounds where it’s just specific DMO like data model object data associated to a lead contact or person account or a field, or a field data that can be updated from data cloud.
So that once again, the lead contact and person account. The only distinction on that one is that it’s view only at the moment, and it’s really restricted to lead contact person account at the moment. Whereas through our Flow process, any implementer, administrator, anybody can use the existing skills they have from a clicks or a code perspective to make it work Flow.
I keep using Flow. People are like, “Well, what about Apex?” Same rules. Same principle rules. You can create a trigger on platform events. You can create a trigger off of… And utilize the triggered Flow and data cloud in the same environment to kick off Apex. You have a lot of those options as well. So in the blog, I have an example of like, “All right, I talked about it in very ethereal terms.” How do I talk about that in terms of an actual use case, like a useful one?
So the way that it’s described is that essentially we have data like a Sales Cloud one because Sales Cloud is our number one biggest cloud, and we have Sales Cloud data coming in from order management systems and ERPs. And it used to be, “Well, if I’m bringing in from order management and ERPs, that’s all ETL talk. Do I need to do that?” No, no, no, no. You can absolutely bring that data in into Salesforce using data cloud. And once it’s in data cloud, you can create those data lake objects, you can create those DMO objects and even calculated insights.
The example that I gave was if you have all these order management systems and all these ERPs, you can create DMO records that combine all of the order data across all these different sources into a single object model.
Josh Birk:
Gotcha.
Shoby Abdi:
You can combine all of them if they’re, let’s say, direct. In fact you’re selling through like grandmas… I’m really harping on grandma today. Grandma’s like a local convenience store. You can get their data as well. And then you could also create this calculated insight record and you could say you want to group all the order data by an individual across all of them, whether they went into grandma store or they went on your online e-commerce website. E-commerce, That’s a good way to put it. You’re online. E-commerce, I’m going to really age myself.
Then regardless of which direction you go, you can get that data from all those disparate data sources, Salesforce, and that can create an order record and associated to a lead and opportunity or quote that can update data within your CRM. It could do all of those things. You can actually show those copy field data, those related list data associated to that lead contact or person account because it’s probably you making that order regardless of where you go.
So now all of a sudden, that order, whether you went to the store, whether you did it online, showing up under you. It’s coming in from data cloud. So that’s a very base level one example that’s described in more visual diagram detail inside of the blog post. But the reason that we went with the route of a Sales Cloud, one, is that right now, obviously data cloud is activated in everybody’s environment. You could turn it on today.
Josh Birk:
Gotcha.
Shoby Abdi:
The first thing that people turn on, they’ve got Sales Cloud. It’s like, “Oh, let’s see what this looks like with my Sales Cloud. Can I use it with Sales Cloud?” The answer is yes.
Josh Birk:
The answer is yes.
Shoby Abdi:
Absolutely you can and you should. So that’s really the goal of that.
Josh Birk:
Nice. Now, if somebody is listening to this and they’re like, “Oh, this might be me. I use Sales Cloud. I have other clouds, I have other data sources, et cetera,” are there considerations? I mean, other than the obvious stuff, don’t go blow up your production just because you listened to Shoby and Josh on a podcast. But are there considerations an admin should take before flipping that switch and turning on data cloud?
Shoby Abdi:
Yeah, absolutely. A lot of those considerations are obviously looking at the state of your existing environment. If all of a sudden you’re going to open up all the sources of data, the source of very powerful automation, I think people should absolutely not underestimate how powerful is because in many respects is that it’s going to be meaningful. It’s going to be meaningful is because when we talk about bringing those data sources in, it’s your data. It’s your organization’s data. But the end, you want to look at your org and essentially understand, okay, is it in the best place it can be to do that? Right?
Josh Birk:
Yeah.
Shoby Abdi:
And bar another word because I’m setting myself here, up here, am I healthy?
Josh Birk:
Exactly.
Shoby Abdi:
Is the organization healthy? So now all of a sudden it’s like, “All right, well if I got to get healthy, where do I find that?” You could look at the prescriptive guidance that exists within the well architected framework, because while you’ll find a lot of items there that maybe specific data cloud, one of the things over the next few months, our product managers associated with the framework and with data cloud, they’re adding more as it goes along.
But one of the things to keep in mind is that if you’re going to bring that data into Salesforce, CRM, anyway, the guidance within what Well-Architected framework is all about that.
Josh Birk:
Gotcha.
Shoby Abdi:
So I wouldn’t even say it’s more of a, well, before you blow up your org… Remember, data clouds are really designed for such a significant amount of data. You may be listening to everything right now, I was like, “Well, Shoby, I got way too much data. I got way too much data.” I can’t handle this. Once again, that data cloud was designed so that problem doesn’t exist. Because you may say, “Well, I’ve got too much data in Salesforce. How do I solve that?” Well, that’s where data cloud has these amazing CRM connectors. Sales and service. And you can actually ingest your own data and utilize the analytics and automation capability at scale to make it work where if all of a sudden an account is 20 years old, you don’t want to lose out on all the amazingness of the last 19 and a half years.
Only the last six months matter from an automation perspective. It used to be used be, “Well, you just got to archive that data and then hope for the best.” Now, you could still right, but in the end, bring it back in data cloud.
Josh Birk:
Get the insights.
Shoby Abdi:
Get the insights, absolutely. Get that historical insight. All of a sudden you’re… And if you start to inject Einstein into this, because it’s the Einstein 1 platform, you start to inject Einstein into it. That is where all of a sudden you may be using the latest and greatest in technology, but your customers aren’t losing out on the historical context that you are providing them. They will gain. They will absolutely gain from what you’re providing.
Josh Birk:
And since we get paid every time, we can say AI on the air, it also completes that picture. When we talk things about like grounding prompts and putting your data into what the model is going to understand, if you haven’t fixed that lens problem, if I haven’t figured out who Shoby Abdi is or who John Smith is, then AI won’t either.
Shoby Abdi:
No, absolutely not.
Josh Birk:
Data collect can help you point in that direction for the AI.
Shoby Abdi:
And really that’s where all of a sudden, even when we talk about principally AI, we talk about that human at the helm.
Josh Birk:
Right.
Shoby Abdi:
Right? For a long time we obviously called it human at the loop, but now we’re talking human at the helm. And really the whole end goal of the human at the helm is that you in many respects need to co-pilot this. It’s almost like we have this product called like Copilot.
Josh Birk:
It’s like, “We’ve thought about that name first.”
Shoby Abdi:
It’s like, “Oh my God.” And really the end goal of being a co-pilot… And the way that I’d like to think about it principally, especially with co-pilot and how it relates to data cloud, is that when you’re a co-pilot and you’re flying a plane, some large 747, what you’re not having to do is go outside on the plane and the flaps, raise them up, go down to the fuselage, put the wheels down, and then when the plane is in the air, “Boy, now I got to get it for cruising. So now you got to go back outside when you’re 10,000 feet in the air.
Josh Birk:
In the air, yeah.
Shoby Abdi:
But the flap is back down. Raise the wheels up. Our aviation technology has evolved beyond that. So when you look at the power of Einstein, look at data cloud, really what it is, is that you truly become the co-pilot where all of this automation, everything is there to help you as the human being to just do your job. But what it’s not doing right is it’s not adding more complexity to it, right? Aviation technology has gotten to that point where it’s like… There’s been jokes of like, “Well, if I fall asleep, who’s going to flip the switch to land? When we have to land.” It’s like, “Okay.” It’s almost to that stage.
It’s like, once again, but would you ever fly a plane? Would you ever fly in a plane that didn’t have a pilot? I don’t think anybody is ready for that yet, right? So having a human at the helm providing all of this, so increasing the ability for that just makes things more trustworthy. It makes things a lot safer for you and your customer. So that’s why I like to… Once again, I don’t think it’s an AI, but now I just did. But in the end, when I think about AI in terms of all this, it’s really, it relates to me. The data cloud story is a huge part of it, right? The Salesforce CRM story is a huge part of it. You can’t talk AI. You can’t talk Einstein without data cloud and without CRM.
Josh Birk:
Yep. All righty. Well, thank you very much, sir. We will give references and resources in the show notes, including to the Architect site and Shoby’s blog post on this topic. Shoby, thanks for being on the show.
Shoby Abdi:
Thank you.
Josh Birk:
I want to thank Shoby for the great conversation and information. As always, I want to thank you for listening. Now if you want to learn more about this show, head on over to admin.salesforce.com where you can hear old episodes, see the show notes, and also read other resources on the topic. Once again, everybody thank you for listening and I’ll talk to you soon.
Today on the Salesforce Admins Podcast, we talk to Eddie Cliff, VP of Product Management at Salesforce. Join us as we chat about Salesforce Foundations and how it can give you access to even more capabilities within Sales, Service, and beyond. You should subscribe for the full episode, but here are a few takeaways from […]
Today on the Salesforce Admins Podcast, we talk to Lizz Hellinga, Consultant and Salesforce MVP. Join us as we chat about why clean data is more important than ever if you want to leverage the potential of generative AI. You should subscribe for the full episode, but here are a few takeaways from our conversation […]
Today on the Salesforce Admins Podcast, we talk to Skip Sauls, Senior Director of Product Management at Salesforce. Join us as we chat about why data is so important to reporting, what some big-data terms are, and some fun hobbies that Skip enjoys outside of building awesome products for Salesforce Admins. You should subscribe for […]
