Name: How to Stop Duplicate Data Building Up in Salesforce
Uploaded: 2026-09-29T17:30:46.811Z
Duration: 1 h 3 min 28 s
Description: How to Stop Duplicate Data Building Up in Salesforce

Transcript for "How to Stop Duplicate Data Building Up in Salesforce": Alrighty. Good day. Good day. Welcome to the session. Good morning to everyone who it's morning to and afternoon to probably the majority of everybody else joining. I'm always curious. Always always curious. If you can jump in the chat and post where you're joining in from today, I'd love to see if there's any other Australians that, happen to be up at this time in the morning. If you jump over into the chat, you can post, say, g'day. I'm joining from wherever you're joining from. There we go. We got Oklahoma, East Texas, Toronto, UK. Yes. That makes sense. All the other Australians are asleep. Thank you so much everybody for joining us today. We've got a fantastic session. Just so you know as well, so you're in the chat right now posting where you're joining in from. That's fantastic. I love to see it. I can see all those different locations, flowing in now. However, throughout the session, you've you will have some questions. Right? And we want you to record and to publish those questions. But if you can jump over to the q and a tab. So there's a chat tab and a Q and A tab. Chat, throughout the session, just post your thoughts and comments about what you're seeing. In the Q and A, what we'll do at the end of the session, there'll be a dedicated time for question and answers with the Cloudingo team. So make sure you post those in the Q and A tab. And that way, they'll be curated, for the end of the session when it comes time for Q and A. You'll also see that you're able to sort of give a thumbs up to any other questions that you see. So if you jump in there and you're going to ask a question, you see somebody else has already asked it. Rather than asking it again, if you give it a thumbs up, that way we'll be able to see, which are the sort of the most, you know, the the question most people are looking to have answered if we have a significant amount and we can't get to them all. Hopefully, that makes sense. Look, I'm gonna stop talking, because, you know, nobody's here to hear my voice. We're here to hear from their lovely Cloudango team. So today, we're gonna be talking about how to stop duplicate data building up in Salesforce. We've all experienced duplicate data in Salesforce over the years. We're gonna see also why manual cleanup, is oops. I'll I'll, why manual cleanup can't keep pace, and we're gonna learn a better approach and a better way of handling it. For those of you who are new to Salesforce Ben, or aren't necessarily aware of everything that we do, you may have read some of the articles on the site. We also have our newsletter, so you can use the newsletter to stay up to date on all the latest and greatest information that we are talking about in the Salesforce realm. We also do Events. So there's these, Virtual Events and Webinars like this. We also have in person Events. So, you know, at Trailhead DX and at the recent Dreamforce, we are there, we we are present. We like to have a lot of fun at those events as well and meet our audience. There's the YouTube channel as well. If you're like me and you like seeing things in action, not just reading about it, you can jump on and subscribe to our YouTube channel, and stay up to date with anything that we post there. And we also have the SF Bend Insider Podcast. And look, without further ado, again, like I said, you're not here to hear from me. You're here to hear from the Claddingo team. So, I would love to welcome to the stage, Beth and Steven from Claddingo. Guys, do you wanna jump on and introduce yourselves? Thanks, Tim. We're we're glad to be with you. Thank you to you, and thank you to the Salesforce Ben team. So I'm, Steve Harding, and I, am one of the cofounders of Cloudingo. Been working with Salesforce and Salesforce apps for, oh, probably twenty six, twenty seven years now and started this company when we became a Salesforce partner about sixteen years ago. Sort of just a little bit about my background, and I'm here with my esteemed colleague, Beth Turicki. Thank you. As Steve said, my name is Beth. I'm the VP of customer solutions here at CloudDingo. I've actually been had the privilege of working with Steve and the whole CloudDingo team since 2013. So I tell customers I came on as it was a baby dingo, and, it's been wonderful to see it grow through the years. And, they're a great team to work with. So it's a pleasure to be here today. It is. fantastic to have you both on as well. thank you. So let's get started and, just talk about, get started with the session, and we'll talk a little bit about what we're gonna cover today. So here's our agenda. And the first thing we're gonna do with this just look at what we're dealing with today and sort of what seems to be on the horizon when it comes to Salesforce and data management. All of the changes that we've seen, we just recently had Dreamforce and lots of announcements there. So we can we're gonna begin seeing those things filtering into our actual interaction with Salesforce and and what that means for how we manage that data. Then we're gonna look at why the issue of data quality and in particular duplicates, which is a seemingly very simple and clear cut issue, continues to linger. So why do we still have this problem? Then we'll talk a little bit about data quality. We talk about it all the time, but what actually does data quality mean, and how do we get to the point of clean data? So what Beth and I are gonna try to give you today is a step by step approach that you can begin applying to your data management, to your Salesforce instance and to help you start thinking about how you can think about the data that you hold within your Salesforce better so that it can serve you better in all of the ways that it needs to now and all of the ways that it will be in the very near future. And that leads me into the next agenda item, and that's we're gonna touch just a little bit on the issue of AI. We've talked about this before. AI is all over the news. We hear about it all of the time. There's maybe a little bit of anxiety around AI, and we want to connect that with data quality and in particular Salesforce, particularly after you've seen the the announcements that were made at Dreamforce just a couple of weeks ago. Now we'll wrap things up a little bit with a very brief demo of our solution, Cloudingo. I just wanna introduce you to what a third party solution can do for you when it comes to data quality and deduplication. It's not my intention to create a sales pitch here for you, but I wanna see I want you to see what's possible, and how an application like CloudEnd can facilitate that process. And then we'll wrap it up. We're gonna give you some time for for questions. You can feel free to put questions in the the q and a tab, within the the webcast application, and we'll hold those to the end though. But I also want to welcome you to reach out to us directly either to Beth or to myself. You can find us on LinkedIn. You can find us through the Clodingo website. We're we're always available for folks. So let's move on. Love it. Well, look, this is gonna be a very exciting session, Steven. I appreciate you introducing it. And I know you've got some numbers to show us in a second. But before we dive into those numbers, I got a question. What's your gut sense on how bad the duplicate data problem is today in 2026 from what you've seen? So that's that's actually a really interesting and a really good question because what I tend to see in customer accounts is that the problem is much bigger and more widespread than customers typically imagine. And that goes across industries and across org sizes. Beth, because you're at the forefront of our customer interaction, I mean, would you agree with that, or or do you have other perspectives on Absolutely. how bad the problem is today? I think it it surprises customers. If they come in and they run some sort of a a comprehensive report on their data quality and their duplicates, usually, one of the first things we see is, gosh. I had no idea it was this big, or there were so many of these. So I think people are aware that there's a problem, and it's kind of a fuzzy notion of, yeah, we have terrible data, but really aren't aware what's going on or how big the problem is until they get in and see a report. Yeah. I I think it's really easy sometimes to find some duplicate records or just some bad data in your Salesforce instance. And most of the time, you're just finding that because you're doing a search of some sort and you come across it. Or you might even be able to use some of the native tools or run a few reports, do some sorting, and that gives you an idea. But I think what what I am surprised by and what our customers are surprised by is when they do have some tool that enables them to get that comprehensive view. It's quite surprising when, they can go deeper than, say, just what what emails match. And it also surfaces other sorts of of data problems that maybe go beyond duplication. It could just be, you know, unverified data or inconsistent formatting, all of those sorts of things that could cause some issues for how your Salesforce instance actually functions and operates. And I think that that data quality at large and and specifically duplicate data seems like such a simple problem, and yet it persists. And though that problem may be simple, the solution can be much more complex. And and and maybe complex isn't the right word. Maybe it's, nuanced and deep. So let's let's just dive in a little bit and and see what what we're seeing, what we're hearing in some of the surveys that, in this case, Salesforce Ben has taken. So, look, looking at those stats, this is this is quite interesting to me. Right? So so 34.3% is is is the full set there, which is over a third of Salesforce admins say that deep duping eats more time than any other task that they have to deal with. And only 2.4%, of responders in that survey consider their ought to be clean. Why is this problem getting worse and not better even though we've got better tooling than we've ever had? Yeah. So I think there's a lot of factors here, and I I wanna bring in a couple of I don't wanna overwhelm folks with with statistics, but I think, we can bring in a few other points from other research, to to get a sense for why a third of admins say that they're dealing with this problem and spending a lot of time doing it. And yet such a small percentage actually believe that their their datasets are clean. So Salesforce's own reporting and analysis, and and I can if if anybody's interested in the citations for these statistics, I do have them. But Salesforce's own reporting and analysis estimates that nearly 25% of of due of data is either duplicate or bad data. So that seems like a really high number, 25%, one quarter. But remember that when Salesforce is doing its analysis, it's looking across hundreds of thousands of orgs and billions of records. But I think laying that over, what you as individual admins are having to do and and tasked with and managing your own data is enlightening. Then we'll take another step, and we'll look at g two. So g two is the the software marketplace and reviewing sites, and, they also do a lot of research and and surveys. Their most recent survey showed that data grows at an annual compounded rate of 26%. So we've got 25% Salesforce estimate of bad data, and we've got a growth rate annually of 26%. So you can see how those things together, exponentially expand this problem. And then the last thing I'll bring up is that the the global management and consulting firm McKinsey, again, they're a big research agency, they just put out a new report, in August on AI and data, and they estimate that 20% of productivity is lost due to bad data. So we're sort of stuck with, well, why does all of this matter? Well, if if productivity is being lost, what does that mean? Well, we're losing efficiency. We're losing cost effectiveness. We don't have those consolidated views that sort of Salesforce three sixty view of a customer if it's dispersed across bad data and duplicate data. We don't have reliable reporting and forecasting. And it certainly causes problems for things like marketing and sales and service, all of those, tasks that rely on Salesforce to do their jobs. And so I think in the end, when we look at 34% of of our of Salesforce users are struggling to manage the cleanliness of their data and still so few think that their orgs are clean. It's it highlights the fact that data is just incredibly dynamic, and businesses today are moving at lightning speed. And most of us are just trying to keep up. And data quality is kind of a a hidden problem, and, you know, who wants to go looking for problems? And I don't know. I'd be interested if if any of the participants in the in the webinar, if you have a reaction to that, there's a little there should be a little react button in your your webinar controller or maybe put something in the in the, chat. And then are you aware of the problem, and do you, like, try to ignore it, or, or is it just one more thing that you don't have time for? But while you're doing that, I'm just gonna continue and move forward. We also have a a poll that I want to, have you address before we, before we continue because I wanna get a little bit of feedback on where you're at with duplicate cleanup. So we just put that on the screen. It's just a quick click on where you're at with your duplicate cleanup. Have you started? Are you just manually doing things on an as needed basis? Are you using the native tools in Salesforce? Are you looking at any third party tools, or are you already using a tool? I find it really interesting. I in my experience, and, of course, we work mostly with customers who already have identified a problem, and they've come to us. And they come to us after just trying to tackle the manual aspect of it. And so the middle steps are often skipped. And the interesting thing is when they actually run some of those initial analysis and get a sense of their data quality, you know, we talked a bit a little bit about this earlier in the presentation that they're often surprised because they didn't realize how expansive the problem potentially is. So how are we looking? Do we have results? Yeah. That that what you were just saying sort of lines up with what we're seeing on screen. There's a significant, jump in those who are sort of manually handling their handling their cleanup, of duplicate records there. If you haven't submitted your response yet, make sure you get your your result your, selection in the poll and then hit submit to lock in an answer and send it through to us, by the way. Anybody who's sort of, still answering that poll? And and I think those poll results do in many ways align with my expectations. I think that the the manual approach is is one of those things that we just do because we have to. We're doing a search. We find a duplicate record. We deal with it, but we don't go any further to to identify whether or not there's a larger issue or that we can systematically address some of those problems in the datasets. And not know. what we hear a lot from customers is that they feel like manual is really the only way their own set of data can be handled. A lot of people don't realize that by really digging in and evaluating what your rules are and what makes a duplicate and what creates the bad data, you really do have other options. There's there's kind of a narrow, blinder point of view that, no, my data is too complex to to handle some other way. Yeah. That's a really important point. Thank you. And and we're going to, as we get a little bit further into presentation and we give you what I'm sort of referring to as a step by step plan on how to begin thinking about your data quality, we'll address some of those concerns because they are very real concerns that we hear at Klaviyo every single day. Alright. But let's continue moving forward. Oh, apologies. I have to reshare the screen. There we go. Alright. Alright. So let's take a moment to think about why duplicates compound. We've talked about how it seems like a fairly simple, easy to understand issue, but why does it get worse and worse? Well, the first thing to think about is where is your data coming from? So often, most Salesforce instances have various kinds of integrations. Those might be forms on your website. They might be, marketing automation or Marketo platform or or accounting gauge, which used to be Pardot or HubSpot, might be, ServiceNow or ERP, any of those things that integrate with Salesforce. And particularly when you're using Salesforce as the system of record, all of that data is coming in. And if you're not checking it, well, it's just exacerbating the problem. And to further, if you're using data enrichment providers, so if you're you're looking to enhance your data with refreshed emails or or mailing addresses or, phone numbers, any of those things that is that are changing your data, making your data dynamic, it's also a a potential weak point for expounding this problem of duplicates. And then, I'd like to jump. in. Get the forms. People were registering for different events and downloads, but they all use slightly different data and different requirements for required fields or formatting for that. And so the system is not set up to check as you're having the new information come in, which, of course, everyone wants to get all of their leads that they can. But it leads to a variety of different data problems. Another one we hear all the time is the the manual creation of records. So you've got sales and marketing and service people who are creating records based on email, communications, or or whatever that might be. And I hear this complaint almost every single day from cell from systems admin, Salesforce admins that their users don't stop to check for duplicate records before they insert the data. And that's a human nature kind of thing, and I don't think we're ever going to solve it. As much as you may want your users to be checking for records before they create a new one, they're probably not going to. And, you know, there is a there is a compelling reason to avoid creating any kind of friction with your users. Most of the time, you just want them to be able to get that data in and start working with it. And there are ways that we can show you that you can manage it on the backside. So you don't create that friction, and yet you're still maintaining a clean set of data. Also, imports. They're another large way of, creating duplicates. Most people who are in charge of doing the imports, they're just tasked with getting the imports in very quickly. And so there may be improperly formatted data, things in the wrong columns, things that if you spend time you can clean up. But the reality is the person who's just trying to get it in very quickly just puts the data in, and it often creates, a mess that needs to be cleaned up on the back end. Yeah. And most importing processes don't have any data quality checks as the data is flowing in. Same with your integrations. And and I also want to also state that it's not just a matter of inserting new data. Of course, we're always interested in new data. Your sales and marketing teams always want new data. But it's also a matter of updating existing data. So if you do mass updates, if you have different kind of integrations that update existing records, every time a record changes, that's a potential for it being identified as a duplicate to some other record that you have in Salesforce. And so you need a system that's in that's going to enable you to check all of these data changes and this sort of, dynamism that you have in your data. And it's so important to try to stay ahead, and yet it is still so very challenging. Something I noticed as well, if I can jump in, I saw that pretty much half of the respondents to that poll before was saying that they were handling their their manual cleanup sorry, their their their deduplication cleanup manually. Why does manual cleanup consistently fail to keep pace even in orgs that dedicate real time to it? Yeah. Well, I mean, at its surface, man by its very definition, manual is manual. It is tedious. It is slow. It is difficult. You know, we're a small company, and we use Salesforce, and we have one and a half million records in our Salesforce instance. I can't imagine going through my Salesforce instance and trying to find those duplicate matches, much less than merging them in a particular way. So manual, I mean, even just a one time data cleanup is something that just won't scale. They creep back in. And we've talked about all these ways, your integrations, your forms, your imports. Those are processes that are dramatically and drastically changing your data on a regular basis. And so you need a plan to be able to maintain that data. And, again, it's just it's not anything that scales over time. And we work with customers routinely who have come in and someone has done a nice job. They'll say, well, you know, a year and a half ago, we had a big cleanup project and got everything clean. And then now look where it is today. So if you're trying to do just one off manual cleanups on a periodic base basis, it's waiting for a huge mountain or for a snowball effect to take place before you go in and, clean it up in a systematic method that you can keep running. Thank you, Beth. Alright. So let's move forward. Yeah. Look. That that doesn't sound ideal or good at all. But look, as as we sort of move into what good does look like, what's the biggest, misconception that admins have about fixing deduplications in their org? Well, there's so many so many things we hear from from the people we work with on a daily basis. I mean, one is that they just don't have time, or or that they feel that it's too complicated. Or, you know, it's it's not important enough. There there are other higher priorities, and they're busy putting out fires. And I think, Beth, you you were just talking to me the other day about a client who was just afraid to make any changes. Yes. There's a lot. of fear that, everyone knows it needs to be done, but if you do it wrong, the cost of doing it wrong and hearing from people come back and say, you got rid of this account, or I can't find this record, or this belong to to a certain person. So the fear is kind of a a big stopping block for someone wanting to step up and say, I can take a hold of this, and, yes, let's get it done. Yeah. I think in the end, Salesforce admins tend to be data stewards and not data consumers. And that may seem a little counterintuitive, but as a steward, the the focus is more on how the Salesforce data model functions properly, and less so about how the actual data serves business operations. So you've got the operations folks who need to work with the data, and they're complaining all the time about its quality. And you've got the stewards who are just making sure that Salesforce is working the way it's supposed to be working. So there's a little bit of a disconnect there about who should be responsible for the actual quality of the data. But the good thing is that we have a plan for how to get you started. Right? So there there are just four steps, here, and you can see the little icons that Beth and I are gonna go through. They are simple, but they're also detailed. And we will follow-up at the end of the webinar with a checklist. So you'll you'll have this available to you. But the first thing that anyone should do when they're gonna think about their data quality and and how data is functioning within their Salesforce instance is to audit the data. And what does it mean to audit the data? Well, there's several things there. The first thing is what actually counts as a duplicate in your dataset? It's not the same across all industries or across all Salesforce orgs. And let me give you just one very simple example. Most of the time, people will think that an email address needs to be a unique, value. And then if an email address if there's an email address is shared between multiple records in Salesforce, those must be duplicates. But in the nonprofit space that use the household model, oftentimes, email addresses are shared among members of those households, and those are records you would not wanna merge. So that definition of a duplicate for that particular nonprofit kind of operation would be very different, than the duplicate definition for a b to b sort of corporate business oriented, use of their data. Alright. So the first thing is what counts? How do you define a duplicate? The next thing you wanna do is try to understand how many you actually have. Now you can run some reports as I've told you in Salesforce, but I don't think that that's gonna give you the kind of comprehensive view you want. And you're probably gonna need to use some kind of tool, some sort of third party tool. Next thing you wanna do as part of that audit is also understanding where your data is coming from. So we have dynamic Salesforce orgs. Do you know all of the entry points? Do you know all of the forms that submit data to Salesforce? Do you know all of your integrations? Do you know which users have permission sets to, enter and edit data? And then there can also be a lot of loose ends in your data model. Do you have standardization rules? Do you use things like, validation, rules to make sure that you maintain your data in a particular way as your data grows. Then I'm gonna let Beth take the next point. So the next step is to plan and actually take action. And so we wanna think about things like, what are you going to address that was found with your audit? There will be duplicates, of course, but there are also standardization issues, which can help improve your data drastically. We wanna talk about data model gaps. Do you would it help to have pick list values instead of just free text for fields so that you have a standardized way of looking at important information for you. And the second thing, once you've kind of identified what needs to be done from standardization to detuning to addressing your data model, you need to find and apply the right tool. And it's amazing, if you really look into once you once you, put in place a tool, it may be a suite of tools in one where you can do standardization. You can, update pick lists. Lots of these things that all need to be done that then make deduping and keeping the data clean much easier on an ongoing basis. And also make it ready to be automated. You'll need to talk to stakeholders and decision makers. Makers. Because one thing we see is a lot of times when I'm helping a customer, we've done an analysis, we're ready to start. building out rules for what they want to do with their data, and they say, well, we need to talk with marketing about the leads. They want things handled in a particular way. Or we need to talk to rev ops because we can only handle data, or accounts that have financial, assets in a certain manner. And so you need to take into account who are the stakeholders and have a conversation with them to find out what's important in terms of how we handle things. So you can put a meaningful but comprehensive plan in place. And starting from that sort of ground up is gonna make it much easier to move forward. It also needs to have flexibility, and for variant use cases. Because we're going to be able to put in place a meaningful approach, and you'll be able to handle it on an ongoing basis, but there are always the outliers. And and a plan that's flexible enough to handle the bulk of it, sort of the old 8020 rule. Most of it in a meaningful but, easy manner. But then make allowances for the fringe cases which need to be handled differently and may take more periodic monitoring than other things. And finally, a good application is going to also service issues that you might not have thought of otherwise. As you get digging into all of this, you may find things that never occurred to you, but that will smooth the past path moving forward. And the next thing you want to to make note of is that you need to be able to repeat this process. So you need to be able to run this process on a consistent cycle on meaningful intervals and iterate. And it's not just a a function of achieving a baseline of clean in your data. Again, we we are constantly advocating about the importance of maintaining your data. And if you're going to maintain it, the process, the plan that you put into place needs to be one that can be repeated. And to that end, you want to start automating, get things running on their own, but also come up with some sort of a meaningful cadence when you're gonna monitor and maintain. Things periodically change over time. You may need to handle something a little bit differently down the road. There might be tweaks that need to be made in six months, or even just have someone who's willing to take a look at things. For instance, in the early phases, you might have a few more errors for for items that didn't occur to you to to fix in the beginning. And all of that by monitoring and and automating as much as possible Yeah. will smooth things and make a much more easy process down the road. So so to restate, these are the best practices. You're gonna audit. You're gonna come up with a plan. You're gonna analyze your data. So that's the audit. Analyze your data, understand what your problems are, what you need to do. You're gonna come up with a plan for how to tackle those problems, how to address them, how to fix them. You wanna make that plan a repeatable plan. And then as much as possible, you wanna automate it so that it can continue to run continuously in the background. Now we'll talk a little bit more about when you wanna automate and when you might not wanna automate. There are different circumstances around when you can and when you shouldn't automate, but we'll cover that a little bit later. So okay. So if I got this straight, you got audit, then you plan to take action, then you make it repeatable, and then you try and automate it. Those are the key four, steps to that recommended process. Is there one of those parts of that recommended process that gets skipped the most from what you've seen? And what are the consequences that you've that you also have seen of that, being skipped? Yes. There that most definitely is. I'll start, and then I'll let Beth jump in too because she works so much directly with our customers. The first step, the audit step, is the one that that is typically shortcut. And I'll and I'll tell you why. So so customers often come to us because they recognize they have a problem. They connect to our application, but our application is not so different than a lot of other applications. And they get this analysis in their data, and they immediately start working because they see how easy it is, and they see what something like Clodango has surfaced for them. But they haven't paused a moment to think about how the data actually needs to be tackled. Now Cloudingo provides a very easy, very powerful way of identifying and analyzing your data, but you also have to customize that approach. And you have to tell Cloudingo that you want it to look at the data in a particular way. And once it finds those results, you want it to take particular types of actions on that data. So I I can't emphasize enough the importance of the auditing and planning phases instead of just jumping into the application and starting to click buttons. Would you agree, Beth? That's very, very true. The auditing, and going very carefully about what you're gonna do so that so that when you start automating, it's it's a very deliberate plan with actions that you expect, and you're actually testing your process along the way rather than jumping in and doing everything at once. You're executing part of it and looking at your results and and and, deciding where you need to go from there. Beautiful. And if we're about to, obviously, we're gonna talk about, AI and the impact on data quality. You guys have been deeply involved in this conversation for a long time now. How have you seen the conversation change particularly over the last few years? So AI, is both exciting and a little frightening. And when when we're when we talk to our Salesforce admins, our customers who, are working on their datasets, they use AI, but there seems to be a little bit of a disconnect between how data quality matters when it comes to AI. And I can see why. So I use Cloud or JetGPT all the time to tell me how to do things, in Salesforce. I just created a new flow last week, and it would have taken me probably two and a half hours to create that flow. I asked Claude how to do it, and I was done in in in thirty minutes. And that's an amazing use of AI, and that made me more productive. You remember those statistics about how 80% of folks believe they're more productive using AI, and yet, the value of those to to operations is not exactly clear. So I think admins are are using LOMs, large language models outside of Salesforce. But what is changing now is that the AI is being integrated into Salesforce. So just, last week or two weeks ago at Dreamforce, AI Force was announced. So AI Force is going to enable you to work with your Salesforce data outside of the Salesforce UI. So think about that for a moment. You're gonna be able to do things with Salesforce data, but you don't have to be sitting at your computer logged into Salesforce to do that. That's available now. The problem with that, though, is that if you're relying on AI for us to to, present you with the data that you need to take action on, and maybe not you as admins, but your your data users, your data consumers, that data needs to be able to support, those AI queries so that they're getting good results. Mhmm. Similarly, Cloudforce is available. That's in beta right now, but you can download it. And it's just like using Cloud outside of Salesforce, but instead, it's actually analyzing your data in Salesforce. So if your data is a mess, the the results you get from Cloudforce are gonna be a mess as well. And I think it's gonna become more and more prevalent using those sorts of tools, instead of using Salesforce's reporting engines or analysis engines that you're gonna be using Cloudforce to do that. And so I think that that data stewards need to be proactive when it comes to how AI is gonna be relying on your Salesforce data and then making sure that your data supports it. And I wanna bring up a point from that McKinsey report that I referred to, previously. Again, it's a report that was published in August of this year about AI. And the one thing that stands out is that there still tends to be a high anxiety, about the fear of job loss because of AI, that AI agents are gonna be taking over our jobs. And yet McKinsey reports that a very, very small number of jobs are actually being lost to AI applications. And I think that's interesting. And I think that what that tells us is that AI is not taking our jobs, but we're gonna need to be able to do our jobs better and support that data better so that all of the different kinds of AI applications that are supporting our businesses can function well and reliably. I think we have a few statistics on the next slide. Alright. Oh, so I alluded to this. I I jumped the gun a little bit when I when I alluded to this. That 80% of people say that AI has made them personally more productive. But if you look at the next statistic, only 37% of organizations are seeing that AI is affecting their bottom line. So what's this gap? Well, that gap is your technical debt. So if if AI is making you more productive because you can quickly create flows or you can quickly do some analysis, that you need to do for in Salesforce. But but those efforts aren't supporting the bottom line in terms of revenue and profitability, all those sorts of thing. That means that your data is not actually supporting the goals and outputs of AI queries. And this is the opening for us as Salesforce admins. It's not about a fear of losing our jobs, but I think there is gonna be some pressure that we do our jobs better because AI is gonna start surfacing some of these problems or maybe to the point AI the results that we get from our AI analysis are far less reliable. It's a little bit of an unknown area right now. Things are moving so fast, and we don't exactly know what the next year will look like. But I do think that this is an opportunity for us, as data stewards, as Salesforce administrators. And, Alright. Steve, one thing I would add to that. There's. Please. a piece of the AI, puzzle. We're we're finding that with some of our customers, someone outside of that admin, someone maybe higher up has gotten excited about AI and put together a list of what they think are duplicates. But then it's not in a meaningful form that you can take action on with your tool. So whoever is working with AI also needs to work with the tool and find make sure that the information you're getting, the valuable information, takes into account all the things you need to take action. Like, what are protected records? How are we gonna merge if there's a hierarchy? So that it's really working synergistically instead of two pieces that are sort of dissonant. Yeah. That's a really great point, Beth. Thanks for bringing it up. I can tell you the number of times just in the last, few months where customers have come to us and said, well, I ran this list of my records through, Claude or through ChatGPT or whatever the the, the LOM is. And it's put all these matches together, but now I don't know what to do. Well, we don't know what to do either. But there's. another way and perhaps a better way that you can achieve that end goal. And so that's what I wanna show you now, and that's Cloudingo. So let me stop the presentation, and I'm gonna jump over to our application real quick. Bear with me. There we go. Alright. So this is Cloudingo. Now I don't wanna spend tons and tons of time going through Cloudingo. It's overwhelming. Nobody wants to sit through a webinar where you see an hour long demo of every detail and every nuance of an application. But I'm gonna hit on the highlights. I want you to understand this conceptually. So there are three things that I'm gonna talk about. I'm gonna talk about dashboards. I'm gonna talk about filters, and I'm gonna talk about rules. Just about everything that you do and can do with the data quality application like Cloudingo is based on those three things. So let's start with dashboard. So what you're seeing right here, this is, just a dedupe dashboard. You can see all of these different options that you have. Each one of these functions, import, converting, mass updates, mass deletes are all done through dashboards. Very quick and easy concept to understand. The next thing is filters. Everything in CloudDingo is based on a filter, and think of a filter as just a report. When you build a filter in Cloudingo, it's pulling together lists of data and based on your criteria. And in the case of the dedupe dashboard, it's pulling together lists of duplicate records. Now at this point, nothing's changing in your data. This is just an analysis. It's a way for you to understand how big the problem may or may not be. So I'm gonna drill into just a pretty simple filter here. I'm gonna look at a lead filter to show you what a filter does. So here you can see the just the beginnings of the filter, easy to build in the UI. Right here is the rule. I'm gonna talk about the rule in just a few minutes. So hold on about the rule, but the rule is very important. That's the one of the real powers of Plantinga. But down below, in this particular filter, I have defined this filter to pull together leads, lead records, that match on the email address, that match on the last name, and match on the first name. But I want you to see that you can use any of the fields on the object that are available to you. And, again, that's why we talked about the importance of defining what counts as a duplicate in your dataset. The next drop down is the matching algorithm. So Cloudingo comes with a whole variety of different kinds of algorithms. The algorithm defines the degree or style of match, and so you can do very tight and very reliable matches. Those are the kinds of things you might feel comfortable automating. But you can also do a whole range of looser or fuzzy matches. And as you get more towards the side of the world that's fuzzy and less clearly defined, those are things that might require some manual review and wanna double check. Alright? As we scroll down, some of these other sections on a filter just create subsets of data. Alright? So you can just for example, if you only wanna look at a particular record type or records that are owned by users, you can, again, define how you want to work what you define what data you wanna work with. So filter defines the data that you wanna work with. So let's talk about the rule for just a moment, and I'm actually gonna drill into this filter. I'm just gonna open up a set of records here. I'm gonna click this button to show you the rule. So the filter defines the records you wanna take action on. The rule defines how that action actually takes place. So it defines what your master record is. You can control things all the way down to the individual field level. So, for example, if you want the email address from the most recently created record, assuming there that if it's the most recently created record and it probably has the most current data, you can pull that into your master record. If you have blank fields on your master record but the duplicate records have data, you can pull that data in as well. So then in the end, the goal is to always build the best version of that record that you possibly can. So you can see how I've built just a few rules here, things like email opt out. In my duplicate set, if any one of those records is true for email opt out, I want the final master record to be merged with email opt out as well. Again, you have all of the ability to use any of the fields that are on your records and set rules for every single field. And, again, this is that sort of auditing stage, that planning stage that you should go through slowly to fully define. But once you have it there, you should be in great shape. So let's go back here to this quick example, and then we're gonna move on. I've got three lead records. I can quickly open them in Salesforce. There's all my lead records. I've got a just an overview list right here, but I'm just gonna go ahead and merge those records. The moment I click that record, Klaviyo, it's sending a call over to Salesforce and telling Salesforce what to do with those three records and how to merge them. And when I go back here and refresh this page, when I refresh, there is now a single record. Now I wanted to just touch on that conceptually. Again, anything you can do in Cloudingo is done with a dashboard, a filter, and a rule. But the last thing before we jump off and open the the session up to questions, then I wanna show you the data quality dashboard. So CloudEndure has a data quality dashboard, and that's what you're looking at right now. And the data quality dashboard is a health check. It's an analysis of your the state of your Salesforce instance, and it surfaces all sorts of problems that you may have in your Salesforce that you're either not aware of or you have difficulty finding and locating, things like inactive pick list values, records owned by inactive users, variety of duplicate accounts, all sorts of things. This is an incredibly powerful and useful analysis of your dataset, And I would welcome anybody who's interested to just run this health check on your data. There's certainly no obligation for that. Alright? Okay. So I think that's enough because I don't again, I don't want to overwhelm you with the details of Clubdingo. The point of this session was to get you thinking about your data quality and coming up with a plan and being aware of it because you're going to need to with the changes that are continuing to come and with the dynamism we have in Salesforce and in our jobs. So I want to jump back over to the presentation, and I think we're ready to open it up for some questions. If I can ask a quick question too, just before we jump into the the questions that others others have asked. So you showed a lot there. That was fantastic to see. For those admins who feel a little bit intimidated by the rules and the filters, what's the simplest way for them to get started? So I think the simplest way to get started is, running that health check. Just running a health check highlights so many issues, and you can start the the beauty of Klaviyo is that you don't have to commit any changes, that you can run an analysis of your data without actually, making any changes. And it gives you time to use the UI to examine what the results of mergers would be, to examine what the results of imports or updates would be before you actually commit them. Mhmm. And so I think that that's the easiest part, just spending time orienting yourself to what the application is showing you. And that goes for any application before you start clicking buttons. Does that make And, Tim, also, just to let anyone who's listening know, this comes with with preconfigured filters. sense? So the queries or reports Steve was showing you, if you start a trial, you won't have to create everything from scratch. It will come with some very common useful, think of them as starter filters so that you don't have to build everything on your own to begin with. And that's another nice nice way to get started. Yeah. And Oh, I appreciate me state that, though, to to Beth's point. Yeah. We do get you started. I think any data quality application will probably get you started in some way. I do wanna encourage anybody to get started in that way, though, to slow down a little bit and take a pause and look at what those filters are doing and how they're constructed and the rules behind them. Because I think I I said earlier in the presentation that this is where I see things kind of fall down a little bit because customers get excited. They see how quickly they can make changes, but they haven't taken the requisite time to really audit their data and think about what that process should be to support their operational needs. Makes sense. And let's go back to your framework too. Step one is audit. Really understand sort of, you know, what's what's required, where you're at before you take action. Right? I like that. Beautiful. We did have a couple of questions in look. Before we jump into questions, I just wanna thank you both very much. That was really it was good to see, Cloudingo in action as well as understanding that framework as well that you that you've displayed, earlier on in the session. This has been really valuable already so far before we get to to q and a as well. So before we, yeah, continue on, I did wanna thank you both for, for for being here and for showing us that. We did have. a few questions have to take. that? No. Absolutely. But the first question we'll go through, is from Regina, which is how is Cloud and Go better than using Salesforce native duplicate tools? So that's a great question. So there's a couple of things about Salesforce data Salesforce's native tools. One the first one is that most of the time, there is a report you can run, but Salesforce native tools will present a duplicate when you're actually on the record. So it's almost as if, it's it functions in the same way as searching for a record. If you look up a record by email address and you get three hits or 10 hits or however many, when you're when you're on a record in in Salesforce, if you have enabled their duplicate management system, over on the right side, and this is in lightning, only not in classic, you'll see a list of duplicates. But, again, that's on a record by record basis. Or if you run the report that you can run, it doesn't give you a way to mass correct those problems, and it also doesn't give you a way to create a system of maintenance. Now if you like the that, sort of alert that Salesforce will give you, And by by using that alert, it does, draw your Salesforce users into the process of data maintenance, and there's value in that when it's a team effort and not just entirely on the shoulders of an admin. CloudDingle also has a lightning web component, which functions the same way, but with more flexibility because it draws on the filters you've built in CloudDingle and allows you to, dig into those records in a deeper, more systematic way so that when you see that component within the Salesforce UI in that upper right corner, it's it's just a little bit richer results. And then, again, you can get your users on board with, with participating and helping to maintain your data. Also, you. have the ability to build out the merge rules that Steve was telling you about where you can handle different fields in different ways. You can create notes for alternative emails and phone numbers. You can, audit as many fields as you want in an exportable report that you could then run for a year. So there there are a lot of benefits in addition to working with the native Salesforce that that take it to another level. Yeah. Yeah. Salesforce doesn't give you the same level of flexibility in defining how those records are actually gonna get merged. That's fair. No. I like it. Beautiful. So there's another question here from May saying, is there an option to auto archive data points that don't match? For example, patient ID mismatch, keep most recent archive, alternate ID, etcetera. Yes. There are multiple ways. And this is, again, one of those things that you would define. And and I'm speaking of Cloudingo here. Multiple ways to define how you wanna preserve that data. The first way is in, what we refer to within Cloudingo, an audit report. So we build into the rules that you wanna preserve certain values, and they're written to a report. And that report can be exported and archived. But most of the time, people want the data in Salesforce. So there's a couple of options for how you do that in Salesforce. There's the function within Cloudingo that takes those values and writes them to a note that is attached to the record itself in Salesforce, and you can preserve any or all or selected values that you want. And then the third option, if you need that data to be searchable so remember notes in Salesforce aren't searchable. You need those values to be searchable. You can actually write them those values to a custom field. So let's just say, for example, you have three records at the same address, same first and last name, but they all have different email addresses. And they look like those email addresses are all valid. Maybe two of them are personal email addresses and one is a commercial email address. But the Salesforce email field has a validation rule on it saying that you can only hold one email, a single email address in that field. When using a custom field, you can write as many additional email addresses, to preserve those values on the record themselves so that your users can see those email addresses, and then you can also search and run reports on those fields. So, again, three different ways that you can store that data, but with absolutely, there are various ways that you can preserve any historical data that you would need to to preserve. That's straightforward. I like that approach. I have another question here from Peggy Sue Marlin, and that is, what about cross object dupes like, leads that match to a contact, for example? Yes. So, again, very easily done. And and I didn't I didn't wanna overwhelm, folks with the specifics of of Cloudingo. But remember I told you there was those three things. There's filters, the the rules no. I'm sorry. The dashboards, the filters, and the rules. There is a convert dashboard as well. So on the convert dashboard, you have multiple options. You can match existing leads to existing contacts and convert that lead into that existing content. You can match existing leads to existing accounts and more and convert those leads into new contacts on those existing accounts. And the third option is if you have a need, you can also mass convert leads. So if you have a list of leads that you just know they those need to be contacts and accounts and not lead records, you can mass convert them using ClubDingo as well. And it does a lot of checks in that process, to to make that conversion happen. Gee, that's handy having that, mass conversion tool there. It's a nice nice clean way of, of handling it. Yeah. I'm conscious we have a few seconds left. There's two more questions I wanted to get to. So, how would it work for nonexact matches? So they're given an example saying, I know that company a and company b in my Salesforce org are the same company, but they have different names because the legal name changed. I wanna force a merge of those two accounts even though they don't match the rules. Okay. So you have multiple options to handle this particular scenario too. So let me just I'm gonna start with the beginning of that question and speak generally, and then I'll speak specifically to to this question. All of those algorithms that you can build in your filters will pull together your data in various ways based on those algorithms. Some are looser than others. And that's where you need to slow down and check your results to see how reliable they are if you are using very loose or fuzzy matches. On things like accounts, we have some really powerful algorithms that allow you to to match, records that have inconsistent prefixes and suffixes. And the example for that would be the Acme Company, Acme, Acme Co, Acme Inc, Acme Incorporated. Using a very specific algorithm, you can pull all of those examples together, and they will be merged. And it's a very reliable, high confidence kind of algorithm. Now the specific question is if a company has changed names, and you want to merge those records, that would be sort of a one off use. And within the Cloudingo, there's a quick search function. We call it quick search. And all you gotta do is type in the values that you want. CloudDingo will find those records and present them to you in the UI, and then you can choose to merge or not merge. You can select which records you wanna merge and apply a particular rule for that merge to happen. So that would be the use case where, you know the records, the the account records have different names, but they are now one company and you want them to be represented as such in data. You could just look up those records and merge them via Klondinio and apply that same level of rule and filtering so that you get the best version of that final record. And if you need to preserve any of the historical data because that company existed as two separate contingencies at one point. I like it. So you still got that granular control at the end of there, which is good. Exactly. Beautiful. And, look, the the final question, which I think I might be able to answer, actually, and that was I think this was related to when you you talked about Dreamforce and AIforce. There's a question saying, where can we download Cloudforce? The quick answer to that is on the agent exchange, which also you will see that there is a Cloudingo, listing on the agent exchange. But I know the team are about to point you to that QR code, which is where you can follow-up and get your Salesforce health check and get in touch with the Cloudingo team for any other questions that you have. Was there anything else that you guys wanted to, to to wrap with there? Because we are a little bit over time. Again, thank you so much. for for being here today and for presenting for. us. If you are interested, please do the health check. There's no obligation. I I think the information it provides is hugely valuable. So so take a look at that, and we will follow-up with a checklist that can sort of guide you in this process of of figuring out how to raise the level of quality in your data. Good stuff. Good stuff. Steve and Beth, thank you again so much. This has been extremely valuable. We really appreciate you presenting for us today. And everybody who joined, yeah, definitely scan that QR code or go to cloudingo.com/salesforce hey, salesforce health check. And, yeah, get your health check, get in touch with the team if you've got any further questions. And with that said, I will leave you to it. Thank you again. Appreciate your time, and we'll see you next time. Thank you so much.