Value-First Data - Mar 3, 2026

๐Ÿ“… March 3, 2026 โฑ๏ธ 43 min
Premium Video
42:39
Share: LinkedIn X

Recording from live stream on 3/3/2026

๐Ÿค–

AI-Generated Insights

Key Points

  • โ€ข Define Unified Customer View as integration goal.
  • โ€ข Include deals, services, tickets in customer view.
  • โ€ข Prioritize relevant data over 100% completeness.
  • โ€ข Automate/maintain critical properties manually.
  • โ€ข Train users on the value of the unified view.
  • โ€ข Send HubSpot notifications to user's workflow.
  • โ€ข Let users update properties at their convenience.
๐Ÿ“

Episode Transcript

Generated via AI Transcription (Gemini)โ€ข 90% confidence

[00:00] **Introduction** Chris Carolan: Good afternoon, good evening, LinkedIn friends, ValueFirst Nation. Welcome to another episode, ValueFirst Data here with Casey and Klemen on part six of our Unified Customer View series. I think this is the last part, folks.

[00:27] **Unified Customer View Series** Klemen Hrovat: Not the last part for sure, I guess, but the last part of that part. Casey Hawkins: The last part of one of four views. Chris Carolan: Yeah. George B. Thomas: Yeah. Chris Carolan: [inaudible] doing good otherwise. Casey Hawkins: It took us six, six hours, six weeks. Chris Carolan: Six parts. Oh man. But here's the why. Uh, we just had uh an email um from uh a client, a mutual client, me and Casey. Um, and listen to this uh Klemen. Uh they want to discuss the Unified Customer View and how they can make this a goal of their SFDC integration. Chris Carolan: Can you imagine such a goal related to integration of Salesforce with HubSpot? Klemen Hrovat: Nice. Chris Carolan: Right? That is in a nutshell the power of this activity that we've been going through for six weeks. When you can get, like it's a, it's simple language in terms of Unified customer and view. Why is it not a mythical thing anymore? Because we have AI to help us handle all of the things that get in the way or handle a lot of things that we've had to do so that now humans can handle the human stuff. Like bringing everybody in the room and having the conversations that, that we need to have. But, um, you know, this came from a session that we did yesterday where it was like the Unified Customer View and the need for it and that as a goal and inform something like a Salesforce integration in a way that avoids the two ways it usually happens, which is we want to simplify it. So we're just going to, you know, push one way and one system's going to be the system of record and we're not going to have to figure out dual syncing. Right? Or we digital twin them, destroying HubSpot's data model in the process, uh, because we don't or we can't figure out exactly which data that we need and we can't agree on it as multiple teams are involved and who owns which data where. Like those are the conversations that are super complex, super conflict heavy at times, right? But when you can meanwhile, the people doing the integration, they need to know which data you want to flow, right? So if you don't know, it's those two options. Right? And so what we're going to talk about today is more like like common pitfalls like that, the things that will usually get in the way of achieving, you know, Unified Customer View. And there's data stuff we're going to talk about. There's human stuff we're going to talk about. There's AI stuff we're going to talk about. But this is in a nutshell, like imagine if you could make a heavy complex integration easier because everybody agreed that Unified Customer View was one of the goals of the integration, right? Klemen Hrovat: Yeah. Chris Carolan: Yeah. Klemen Hrovat: You're 20% towards the goal already when you start. Chris Carolan: Exactly. Uh. George B. Thomas: Yeah. [08:20] **Common Pitfalls** Chris Carolan: So, common pitfalls and how to fix them. Uh even well-intentioned implementations can go wrong. Here are the most common pitfalls we see along with symptoms to watch for and specific fixes. If you're experiencing any of these, you're not alone, uh but they're all fixable. Um, speaking of minimizing the data flow, right? Only configuring contacts and companies. Uh, you've got a unified view, which maybe there's some engagement on there. Uh, maybe, there's definitely demographic data, like fit data. But when you are en-servicing deals and services and tickets and all the other context, right? People are forced to go into other systems or export data to to really create their unified view. Casey Hawkins: Yeah, I just actually just the call before this one was showing leads to a client. And I know leads isn't one of the ones listed here, but the I love when I get to show someone in something in HubSpot that like solves their exact pain point. Like the the exact pain point was we have these deals, but they're not really deals, but we want them to be deals. And I was like, leads. It's like just is leads, that's it. And um the specific person came from a Salesforce background and leads behaved differently there. And so even the word lead, they were like, no, that's not, I know how that works, but that's not how it work how I want. And I got to be like, no, this works exactly like what you're wanting to do. So I just echoing that. I think some of these get hidden, especially um you now in HubSpot have to activate a lot of the objects. Um, so for the specific client, they were like, I don't even see leads on my menu. And I was, but all I did was click two things or something and there it was. Klemen Hrovat: Coming back to, I guess, the initial discussions, you have to figure out the processes not limited by your understanding of the systems. Um, if you start with that, what you, what you're looking for and there's a place in HubSpot that can give you that. Um, yeah. Chris Carolan: And that's that was the thing that resonated yesterday. And why we just got this email. Like, without something like a unified customer view, you rely on the tech, the technology or the SaaS vendor to tell you how it should be done. And if you search right now, even with AI, AI is going to say, yeah, HubSpot, that's for contacts, companies and deals. Right? And it even though you know that's not going to give you a unified customer view. Like, you're going to, you're going to follow the, the best practices or the people that you're working with that are willing to do the work that you're paying to do the work, they're going to follow best practices as they know it, and they probably don't know a lot about HubSpot if they're integrating a different system. And, you know, here we are of impossible to get the view that you want, right? So we're just what we're doing is like the complexity of getting to your Unified Customer View shows up in all of the conversations that need to be had to create the slides that you need at the end of the month to create the Unified Customer View. Right? As opposed to figuring out the data model and all the records and object types involved that can just systematize it and make it, you know, create maintain, create maintained trusted data. Next on our list. Klemen. [15:41] **Properties Exist, But Nobody Uses Them** Chris Carolan: Properties exist, but nobody uses them. Casey Hawkins: Familiar? Klemen Hrovat: Oof! Klemen Hrovat: What's that? Never heard of that. Klemen Hrovat: Uh I I I think that came in each of the episodes in a way. There is data but if you don't make it available to the humans who should use it, uh no matter how good you are with bringing that data to that property. Uh if it's not available visually to the user who should see it. And then also all those um conditional properties that are visible based on where the contact is in the journey. Um, gets in in conversation to make that visible. If you have 30 properties all visible, right? Go and find the one you need. But, you know, if you if you can have only the the six relevant ones, then you'll immediately see those six relevant ones. If they are populated or not. And if they are populated, you see the data you need. Casey Hawkins: Yeah, I think this is a huge problem in every HubSpot account I've ever been in. Sorry. Chris Carolan: I think you have an interesting take here uh Klemen, like when you think about, how you've talked about it before, where it's like, forget about having truly clean data or truly complete data, right? Like can you share that here? Because it definitely applies. Klemen Hrovat: Um, what should I show? Chris Carolan: Like why is that a good policy as opposed to we got to get our data cleaned up and we got to make sure everything's filled in. Klemen Hrovat: Yeah. Um, for various reasons, you don't need to and and you shouldn't chase 100% clean or complete data. Um, one is sometimes there's just no data available and the there is no data is a valid information. Um, we we often have a discussion with with our clients when they're asking us, so can you help us, you know, have that property being 100% complete? yes, but tell me which random number I write you as number of employees if that information is not publicly available. I can give you a random number. Do you need 1,000? I can give you 1,000. Do you need 5,000? I can give you 5,000. If the data is not available, it's better to say not available data than do something which is completely useless. Um, the other one is if you're forcing users to fill out some data for the sake of just filling it out and not them not understanding the usefulness and the relevance of that data and how that is used further in the process for reporting for filtering, you will force people to fill out data which they don't wanted or they don't feel called in to properly fill it out. Which means you are forcing them to create a mess because they will. I would almost guarantee you if you will in that case go into the analysis, let's say, if it's a drop down menu of five values, I can guarantee you if you're forcing them to fill it out, 80% of the choices made will be the first one because this is the the easiest to click. Uh, so you get the complete data, but the completely useless data. Um, so it's it's better to have sparse data but relevant and and and useful than having a complete data. And then for the sake of of cleaning it up, the goal I I think it, you know, there is kind of um a plateau of how much time and resources and money it still makes to invest into something and let's say 80% of of the data cleaned up will get you somewhere to get to 85% you will need to, you know, spend the same amount as you spend for from zero to 80%. So then the question is where is the sweet spot of how much is is worth spending so that you get something back. And the other one is if we're talking about someone who just entered into the CRM, do you really need to have all the data completed and filled out as someone who is a week from closing a deal with? Most probably not. So again, don't chase and don't complicate too much to have, I know, job titles filled out for each and everyone that is entering your CRM so that you you ask too many questions on the forms just because you want to have a complete data. Make it simple. For them, when they go through the funnel, as you have in your mind, you will be able to to gather the data or ask them down the road. So it's the goal shouldn't be 100% complete, but the goal should be really relevant data should be filled out. Um, where I think it's a valid point to say, no contact should be assigned to a deal if they don't have a job title. That is a valid and I think a reasonable ask where you require someone to fill out a a job title. Um, but not every subscriber uh or audience stage person should need to have a job title filled out. There is no point in that. So there are different layers why I think strategically thinking of what's the sweet spot for every property in terms of completeness and relevancy. Um, that you can you can make and there are always of course tradeoffs. But um, there is a sweet spot. Chris Carolan: And again, this is the power of the view. Like the view is what usually helps you decide how relevant things actually are. Right? Like if you're to look at a deal view, one with one with job titles associated, one without, like you in that moment will get to decide like whether or not you actually need job titles, right? And it'll start to at least create the thought process of um, do we actually need the data or not. Uh, like can you click on the carrot uh next to high impact, right? So only create properties that will either be automated or are critical enough that someone will maintain them manually, right? And if the strategy is we always need to be talking to CTOs, the only way you can make sure you're always talking to a CTO is if the job title says CTO. Right? And this is where the process and relates to our first one, like when contacts and companies are the only things you're configuring, sales owns those, right? So if you're not on the sales team, like how much enablement do you actually have to help maintain the data and create the context needed, right? When you own, when you involve these other objects like that also have tons of properties on them, right? You get to own that and people start maintaining data together and that's when, you know, it actually works. Um, you hit on most of these issues I think. Uh again, it it does come back to the data. Um, But just the act of, right? Because I think that's where this is what makes this concept that you just shared. Documenting a unified customer view every organization in the world can do that without touching their CRM data. Like without cleaning it, without starting over, right? Like you can get sales in the room, marketing, whoever's customer facing, right? Maybe product too, the people building the product. Like, what do we know, people who buy from us? What do we need to know about them? Right? And again, from a marketing perspective, you might be saying, oh yeah, that's why we need to have the documentation of the ICP meeting. Like sales doesn't want to do that. Right? It's there's nuance, but documentation of an ICP is much less tangible like for a sales team to understand how that's valuable for them, especially if the thing that makes it valuable is marketing execution against that ICP, right? As opposed to Unified Customer View, right? Again, starts to streamline the process that guess what happens when you decide what your Unified Customer View needs to be. You start to understand what your ICP looks like. You start to understand what properties you need to do good scoring, right? Instead of the other way around. Um, so let's hit on I like this one about like so we just basically hammer data quality issues into the ground. Um. Uh. [31:34] **High Impact Configurations Complete, Behavior Unchanged** Chris Carolan: High impact also configuration complete, behavior unchanged. Right? This is where even when you get everybody in the room, sales is excited. We're at sales kickoff. We're going to start doing this thing. Everybody agrees, we're going to build these properties. We're going to fill in this data. And then people are still doing things outside the CRM. They're still exporting data to get their job done. They're not using the views. Right? Then when everybody asks, oh, what is the value of the Unified Customer View? Everybody's like, And then they're like, oh, well, Unified Customer View is obviously not the right thing for us to do. Right? Again, this should sound very similar to any CRM or HubSpot implementation you've ever done before. Right? But the simple act of just changing the frame of what we're trying to accomplish is turning out to mean a lot. Like I'm, I still get surprised at how quickly it tends to work in some of these conversations. But at the heart of it, like changing behavior, right? Takes a lot in terms of people needing to align and believing that the thing can be done and that it's going to be valuable for them. So including training, you know, and change management as a part of this from their perspective is, you know, still, still super important. Klemen Hrovat: Yeah. Casey Hawkins: I think that this is a really critical piece to maintaining the Unified Customer View. Um, one of my favorite parts of the Unified Customer View personally is the uh, we're putting it in the hands of the users in many instances. So we talked about job titles before. Uh, one thing I talk at nauseum about is, okay, so great, you have job titles for everyone in your CRM. That's awesome. When were they last updated? Are they accurate? Are they spelled correctly? Because if they're not spelled correctly, then the filtering and things um at this stage, is made me all frame it, uh get really problematic. So I think, yeah, getting that buying and training on not only what the view is and and how to use it, but I think even more importantly, like how it positively impacts the receiver. Klemen Hrovat: I think it's similar to data cleanup and the quality, it's not about data quality, it's about what you get because of good data quality and and data cleanup. Um, the same here, it's not about, I mean, setting up a Unified Customer View is the first step, but when people realize the value of that view, then they will use it. Um, so you should, yeah, definitely think how to get them to that point. And when they get it, I think they will get it and will use it. [36:47] **Notification Fatigue** Chris Carolan: As we go to the the medium impact issues, I think we hit on one of your favorites, Klemen. Uh notification fatigue. Um, and so often this is like, all right, we've all agreed to keep this data up. So if you don't put in the deal amount, you're going to get a task, it's going to be assigned to you and you're going to get eight emails about it. And then uh somehow they still manage to close the deal and then that's when the deal amount gets filled in, right? Because they still got a job to do. Um. Uh. How it's very so hard to disconnect these things, right? Like add more only when team asks for them, right? Uh, and current alerts are being acted on, right? So much of this can be solved for up front if you just have conversations of like, we agreed to do the view, you agreed it was valuable. Uh, you're getting these in your inbox, right? These notifications? Oh, you're not. Okay. Well, let's solve for that. Oh, why not? Because a year ago you set up a folder to push all HubSpot notifications out of your inbox. All right, good to know. Maybe we could do less notifications. Maybe it's training on how to come back to this view. Maybe we need to bookmark it because you're a bookmarks person and you would rather scroll through a thousand bookmarks than uh a thousand emails. Uh, like how how do you guys, like are you coming into HubSpot uh or after this and like starting from blank notifications and starting to work your way up from there. What's your approach here? Klemen Hrovat: Uh don't look at me, but I never open notifications at all. Uh, I don't even know where notifications are in HubSpot, sorry. Really. I always start somewhere else. Casey Hawkins: Yeah, I mean, I definitely don't feel like I start with notifications. Usually when I address notifications during an audit, usually it comes up in the um uh information gathering phase if you will, where at some point somebody complains that they are getting notified or they're not getting notified or something like that. Um, I think a little bit of that is just um going in there's a lot of places to look and notifications tend not to be my like first priority is what I want to frame it as because but some of that is the way people receive information is so different. Um, I work off of my email. Like that is how I work. Um, that is certainly not everyone. That's not everyone in a single organization and assuming that everyone has the same like mental organization pattern is just incorrect. Klemen Hrovat: Not to be the party breaker for notifications. Um, I I think these are important to leave where each person starts their day. If they start within notifications in HubSpot, those notifications in HubSpot should be only the relevant ones. Otherwise they waste, you know, too much time. If you are like Casey and and and and me, you would start in email. So I try to have, you know, zero inbox mantra. So I only have emails in my primary inbox that should do something with them, snooze them or react on them. And when I have clean inbox, I feel okay, I I did everything and then I look at my calendar. Some people start in Slack. If this is a place, maybe you should be just, you know, sending the same the relevant notifications from HubSpot to Slack because this is where they start. Um, so this is where you know, every user starts somewhere else. So bring the context and notifications and and and everything else as close as possible to where they want to have it or they they have it. Um, and that's how you you start pulling them. They will get to to the unified customer review, but the alerts should be where they leave, wherever that is. If it's WhatsApp, figure it out how to get it into WhatsApp. Chris Carolan: Right. Yeah. I I think at the this is one of those like accountability like situations, right? Like, we all agreed to do this. Like the point of the notifications is to make sure we do it. And are we putting people in a position to like if you're not sending it to where they are or if you're overloading them and then you try to hold them accountable. Like it's like Klemen saying, it's kind of on you because they have easy excuses to say like I can't possibly like go through all these notifications, these ones don't really matter. So I have a job to do again and it's not wait through all your notifications to make sure I'm filling out a property, right? And again, the difference between keeping data clean or maintaining data or filling in a property and saying, okay, Klemen, I get it. Uh, but we all agreed to maintain this Unified Customer View and you said this was going to help you uh close more deals. So what's the best way to help you keep the Unified Customer View like up to date? Right? Because you mentioned that a few different ways people could come from. For me, I'm opening up AntiGravity now, starting up a cloud code session, uh using the daily ops skill where he looks at HubSpot, he looks at my calendar, he looks at my email. And if there's uh irrelevant notifications or emails in there, I see it and I'm using tokens on it and it's a part of the report, right? And it's not relevant. So the output is is not as valuable, right? And can you have those kinds of conversations where you're like, okay, I get it. If we do this differently, like, can we get to this accountability like agreement to where if we do this, then you agree that it's a great outcome for everybody and you'll participate, right? In the process. Um. Go ahead. Just one one one one more thought on that. Um, you should let the user to complete that property at their own convenience. Because otherwise you will encourage them to just randomly type something or select a random value because at that point in time they're not ready mentally to choose the right, I know, close lost reason. Maybe they deserved half an hour end of Friday to go through those. So don't force them Monday morning to fill out those three closed lost reasons because you'll not get a useful and relevant data. You will get data, but not a useful one. Um. Yeah. Like creating relevance. This is where like relevance and timing and context, like just as important for your internal customers as it is your external customers as it is AI, right? And that's why the only way you can really get to that is by talking to each user, right? Um, and this next one, uh workflow workflow conflicts. A very specific use case, I think. Um, but properties changing unexpectedly. Uh, the thing that we're actually we need to address here is any uncertainty related to what's happening like in the CRM. Like why did this change? I don't know why this change. I don't know where this data came from, which has a way of ruining the rest of the data on the screen as well, because now if I don't trust this one, I might not be able to trust the other one. Or if I make this change, is that going to change like more data? Uh, what's happening here? And we can't be like, oh, it's okay. Just go to the activity feed and then check the workflows box so that you can see all the workflows and all the lists that everybody's a part of and you can get to the bottom of of why that, yeah, nobody wants to get to the bottom of that stuff. Except for the admins and the RevOps people. And how how do you help? Now, having said that, people do need to understand, I think, on some level, how the data does flow, why changes happen. How do you walk that that line of of avoiding the outcome where it's, oh, RevOps is responsible for for all the data, all the data maintenance and cleanliness. I'm just sales. I'm going to come in and it's supposed to be right and that's my job. Like how do you, how do you walk that line? We as celestial, more and more incline into let the user feel their efficient by letting them enter massive data with all the typos, for example. Don't force them to, you know, not do any typos at chief technology officer. Because this is a hurdle, a burden for them and they will be pissed off if they need to be responsible for for for the typo correctness. Uh, let them just write it down and have the system to clean it up. Now you have technology that can clean it up. Um, the same, you know, goes let's say for location. You from every source you always get a different format of location. Or if you type it in, you will have typos. And that's fine. I I think the user shouldn't be responsible for for that. There there are properties which are raw data where you expect mess to happen and RevOps should be then cleaning it up if you want to have it standardized. For location, if you want to use it as a as a standardized form as a drop down, let's say, at the state level. I now start, I I like a month ago I realized why HubSpot has the default country uh state property being a free text for that exact reason. Just let them type in or let the integration write whatever form, US, USA, US asterisk, whatever a everything is is is possible because then you can standardize so the users don't feel burdened by that quality. Um, and in my experience, even if you'll explain to them what what will happen through the workflows, tomorrow, I mean, workflows in HubSpot for sales people, I don't think they will ever care. And I don't think they need to. So even if you try to explain them all those workflows and and document and and walk them through, tomorrow they will forget. So I think as as RevOps and and and and admins the responsibility is to make it as effortless for them as possible and you take the burden off standardization cleaning up. Um that's how we see it. Chris Carolan: Umuted both of you. I was too long. Yeah. Uh. Now we hit this last one. Uh lower impact uh but so easy. And this is definitely, I think uh usually not a HubSpot thing, um, but very common in systems like Salesforce where people just can't they can't access data that they know is there, they might have inputted themselves. And again, um, it starts to like chip away at what the system can do for you, what you're seeing in any given moment. Um, and you only you don't try too hard to to troubleshoot that. Um, so again, understanding what each user needs and at the user level, right? It sounds like a huge burden to figure out what each unified customer review needs to look like. But when you understand that everybody's process has to have it and we want it to be in the system versus in their head or in their slide deck or in their spreadsheet, like it should be not be an option whether or not we've documented that the unified customer review or not. It has to exist because when it does, uh people see all the context and they know what action to take next, right? That's the ultimate goal here. Um. And it's. Go ahead. Just one one one one more thought on that. Um, you should let Casey Hawkins: It's pretty awesome.

[52:45] **Final Takeaways** Chris Carolan: do we get to 100%? Klemen Hrovat: Let's check. I didn't click anything particular today. I did quick part six. So what's next? Well done. Well done. And we are, no, 95. 95. So there's still something more to do. Uh. You know, we're not going to worry about complete complete data here. Um, and we are going to be happy with 95. 95 is good enough here. Right. It says complete, so. It's sigma 2. It's a it's a two sigma uh confidence interval. So I think I just said all of those words wrong. But we're good. We're good with 95%. We're doing a. Um. Yeah, if you guys want one more formative, yeah. Office hours every Tuesday, Wednesday, Thursday, 3 to 11 Central time. This is where we come and talk about this stuff. Happy to share any you know, use cases or or case studies that are happening in real time right now. It's usually the stuff we're talking about. So, uh we'd love to to have you there. Um, and I appreciate Casey and Klemen for for taking the time to really dig into this uh. This concept. I think we're on the right track here. Casey Hawkins: Thank you. Klemen Hrovat: Thank you. See you next week. Chris Carolan: Till next Tuesday. Everybody have a great week. Klemen Hrovat: Take care.

Enjoying the Show?

Subscribe to Value-First Data and never miss an episode.