The Hidden Role of Data Quality in Credit Union CRM Success

Jason: So we're here with Austin Wensloff from Nook. I unofficially the cr the Hubspot Q so. I've seen that all over LinkedIn and you know, Austin I'm sure is plenty of experience with with HubSpot, but we wanted to talk to Austin because he's got experience in in CRM in general. And it's an area that I find very interesting, because of some of the challenges that I've seen at my three credit unions that I've been at over the nearly three three decades in the movement. So I think Austin is our first panelist from the CRM series that we've started here at the podcast. So I wanted to welcome Austin as our Our Virgin Podcast CRM expert. So welcome, Austin. Why don't you tell everybody a little bit about you and Nook?
Austin Wentzlaff (Nook): Sure. Yeah. Thanks for having me. And I like how you said the unofficial HubSpot Cuso, because I I don't think we can make official claim to that, but that is that has been what a lot of people have been referring to us as because we are a CUSO and we do a lot with HubSpot. So you can make that you can make that conclusion for sure. but yeah, I think my my backstory, I guess, is I've I have spent a long, you know, long time, at least in terms of r relative to my lifespan in the crediting industry. It's little over a decade. I got my start right out of college, had no idea what a cred union was, nor did I aspire to go work in the industry, but I I got excited about entrepreneurship, which brought me to startups. And I joined a startup that did data warehousing and data analytics for credit unions. That was also a QSO. that was on approach back in the day. and had a good stint there. We got acquired. I left and and then started a consultancy called FinTech Accelerator. and then the whole point of FinTech Accelerator was I wanted to find the best up and coming solutions that I think should be in this industry and then help them crack into the into the industry because it can be pretty hard to sell to credit unions. You know, so getting to to bring those solutions was was a lot of fun. And I worked with Aluma, which was Aluma Labs time and Renify and you know both of those companies also became QSOs during my tenure there. and then now I'm at Nook, which I can go into deeper detail on there too. But like I said, Nook's also a QSO. So this is my fourth QSO that I've been a part of in the last twelve years.
Jason: That's great. I you know, I absolutely appreciate the the pedigree of QSOs and credit union movement. why don't you tell us a little bit about Nook and and and what Nook does and how Nook serves the credit union movement.
Austin Wentzlaff (Nook): Yeah, for sure. I think to do that, I want to take a s a quick step back. I'll explain kind of what brought me to this point of starting Nook. So, you know, like I said, I've been in the industry for a little over a decade. And during that decade, I've just been kind of observing what is happening to the industry. Where were we really strong in the past? Where are we maybe not that anymore? And where are we going to be that hopefully in the future to be able to position ourselves in in in a terms of you know position of strength. And what I came to the conclusion of was credit unions of the past had two really strong core differentiators. Differentiator number one was we were the best on rates. We were that because of our tax advantage status. We aren't that anymore because of FinTech competitors that are willing to lose money at all costs to acquire customers. so that's a differentiator we had, which we don't have anymore in most cases. And I don't know that we'll ever have that again, you know. Also not a problem. I was willing to try to solve. I can't get these guys to stop losing money intentionally. So I wasn't focused on that. But the other differentiator that I noticed was that we were the best on relationships. But I kind of came to the conclusion that cred unions got to become the best on relationships really as an accidental consequence of their design, meaning we were built for one group of people. And if you're built for one group of people, it's very easy to understand who they are, what they need.
Jason: Well.
Austin Wentzlaff (Nook): How we should talk to them. If they're the utilities employees credit union, I can put a guy on a utility pole with a hard hat, work in the lines, and that's going to resonate really well with that individual. but again, cred unions aren't that anymore. We've expanded way past our initial single seg, single field of membership type of environments that we once started with, and therefore we now have to, if we want to reclaim Our number one seed as the top relationship owners, which I think we sh we can and will do. We now have to be a lot more intentional this time around. And for us, that intentionality means better data, better tech, better strategy, better execution. And we, when we looked at this, we said better tech for us is gonna be Hubspot, which also, you know, obviously is a CRM and then some. but that's what brought us to this journey was how do we build deep relationships with our members so that Kredians can can win in the future and CRM is a component of of what's needed to help help that next leg of growth for us.
Jason: Mm-hmm. But Austin, all the core vendors have CRM products that are attached to their core. the core vendors have been very diligent over the last 40 years in providing, a vertical stack that they want you to stay in. without, speaking to specific features and functionalities, why do you think There are still to this day a lot of credit unions that are still locked into whatever core CRM relationship tool is available from your core vendor. Nope.
Austin Wentzlaff (Nook): Yeah, I think you're trying to get me in trouble here. I see. But yeah. No, I think the I think the the core provided solutions, you know, this goes for CRMs, data warehouses, you know, all of the different kind of out-of-the-box capabilities the cores have provided. A lot of them do work decently well, but they work for a very specific use case, which is you're on Pfizer DNA or you're on
Jason: Right.
Austin Wentzlaff (Nook): Scitar Episys or you're on whatever core it is, well, CRM, data warehouse, all of those things obviously need more data from different sources that aren't that provider, that speak different languages. and it really starts to become limited when you start thinking about it more from an enterprise scale. So I think some of those tools work and they serve a good purpose. but once you start saying no CRM and data, that's actually an enterprise function, that's where they kind of they kind of fall apart. So we see HubSpot or solutions like HubSpot, you know, kind of solving that problem. There's still work to be done to get them to be an enterprise, you know, platform where they can integrate data from loan origination and core and payments and all that stuff. but they are architected in a better manner to accommodate that versus the core specific ones were architected specifically for a core, you know? Thank you.
Jason: I that was very diplomatic. Well done. so you know, when whenever there's the CRM conversation comes up, I I hear a lot of people instantly recoil because of past histories with having to deal with, you know, pipelines and manual input of leads and and and a lot of focus on the sales side of CRM and and relationship management in general. W where do you fall specifically, you and Nook and you know, over the last, 10 years that you've spent sort of in the movement, where do you see CRM sitting in the credit union ops stack for lack of a better word.
Austin Wentzlaff (Nook): Yeah, I think where I look at it is kind of to take a step back on my own journey here is the I started my career on data warehousing, like I said. And you know, kind of the promise of data warehousing is once you have this single source of truth, the holy grail is unlocked and all things will will, you know, be available to you. The problem is it's still missing that activation layer. What are you gonna do next with that, with that data and how are you gonna automate that? and I think that's been the gap of where CRMs have failed in the past too, is that they are that activation layer. You can do those things with it, but they also need that data first. So we are kind of the bridge in between those, where it's like we almost need to build a data warehouse, if you will. That's our Nook data platform. You know, we don't we don't tell people it's a data warehouse. That's not our claim to fame or what we're trying to do. But essentially you need an enterprise in integration engine. That can get all the data cleaned up and readily available. And then you need to put that into in our case, HubSpot. And then HubSpot's able to now build the automations, the workflows, the segments, you know, and fire off real-time things with trigger events. so that's kind of I think where the promise has fallen short on both of those things. We're trying to come in and and connect those dots. and I think a lot of it is one. Not a lot of people have thought about it exactly that way. And but two, I think it's advancements in technology that allow us to do this. You know, we're able to build stuff a lot more efficient efficiently than you used to be able to build stuff. Like core system integrations is a great example of that. Those are a beast. They're difficult, they're complicated, but it's getting easier, you know, and it because it's getting easier, we're able to build a more end-to-end comprehensive solution. that doesn't cost twenty thousand dollars a month right out the out right out the gate. I think that's been the the hindrance of like, yeah, I have a data warehouse product and I want a CRM, but both of them out of the box are eight grand a pop. It's like that that's where it gets it gets difficult. Yeah.
Jason: Mm-hmm. So ha have have you sort of cracked the code and and come up with a a framework or an infrastructure that basically a credit union sits their CRM product, whatever that is, in your case, it'd be HubSpot on top of to provide next best product, actions, things like that. you're you're essentially building out an orchestration layer that lets lets you do things with data coming from disparate data sources. Would that be correct?
Austin Wentzlaff (Nook): Yeah, that's exactly it. So it's, you know, we always say it's it's an industry agnostic, industry standardized data model. So we don't really care what the source system is. We want the HubSpot architecture to look the same from our clients so that we can start to, you know, we're trying to do what QSAs are supposed to do. Let's bring the resources of the industry together. Let's work on common problems and let's and let's tackle that together. You can't do that if you're on your own independent. system, your own independent journey of Hubspot. So someone has to come in and normalize things, standardize it and do all of that heavy lifting first. and then once you do that, yeah, then there's a lot more, you know, opportunity. and I think, you know, you you mentioned a could a couple of good examples like NextBex Next Bus product, attrition modeling, those types of things. You know, those are definitely things you can start doing now that you have this. And what we love about You know, like I said earlier, the data plus the activation layer is we can automate a lot of this now and we can set it up to where the moment Jason's next best product is an auto loan, we fire off that trigger and no human needs to get involved in that process. We know the data, we know the analysis, we know the template that's gonna go out about an auto loan. And the moment that you get added to that list, you get that email, you know.
Jason: So y you had you you said there one of the one of the the advantages, one of the differentiators we had was was rate, which I would agree with you. That's those days are long gone. We, you know, even just from an efficiency perspective, fintechs can do better, faster, you know, smarter, and they've got the you know, the capital to make sure that, you know, their products and services get delivered way quicker, so their margins are much smaller than ours. so we're definitely not going to compete there. On the relationship side, things like next best product, 360 view degree view, all of those sort of you know, marketing catchphrases around building the relationship or at least storing the relationship digitally in a platform that is then actionable for the colleagues. we call our employees colleagues. So whenever I I say colleague, I mean employee at of the credit union. so Is that the new def differentiator for credit unions? And is there is there a sort of an an asset size that's an entry point? You know, I don't think a $30 million credit union is going to have an orchestration layer, is gonna have you know like a you know a a a big hub spot CRM deployment. Where do you see sort of the sweet spot of when credit unions should be talking about operationalizing? some of this data that's available so that they can become more actionable and they use that data.
Austin Wentzlaff (Nook): Yeah, I actually think the the bar is becoming a lot lower. and I think a lot of it comes to what I was saying earlier about our ability to build this technology a lot more cost effectively than it once was. You know, building a data integration to a core 10 years ago was a person or a group of people sitting down and writing SQL and or whatever, you know, whatever programming language you're using, and it's lines and lines and lines of code and hours upon hours of development, whereas obviously there's advancements with. the AI technology there. So I think where a $100 million credit in five years ago could never afford a data layer to then populate their CRM slash marketing automation platform. We're changing that. We actually have $100 million clients now that have legit robust data infrastructure that was once unattainable for them. And we're doing it on an asset-based pricing model. So it's affordable to them because, you know, that's just what the industry's used to. So we're, we're, we're building our pricing model, you know, to accommodate that. So so yeah, I think it's actually it's not a one billion dollar plus luxury anymore, as it as it once was, you know. And I think too pricing from from a HubSpot perspective is another thing that I've been actually surprised myself about when I've come into these hundred million dollar credit unions.
Jason: Mm-hmm.
Austin Wentzlaff (Nook): I have them kind of do an analysis with me of okay, what are you paying for now? You have an you have a email marketing tool, you have a social tool, you have a survey tool, you have all these things that I can potentially consolidate. Like let's add it up. In two of the last instances where we we signed up with a hundred million dollar credit union, HubSpot was cheaper. It was cheaper to go to a better platform for them. It was kind of
Jason: Wow. And you can throw a chat out on their website. Like, you know, there's a whole bunch of things that that are kind of bolt-ons to HubSpot specifically, but all of the platforms have similar, you know, value propositions. So I the it's great to hear that hundred million dollar credit unions are getting into the game because you know, to your original point, competing on rate and you know, competing with some of these fintechs is impossible. But y it's hard to keep up. And you mentioned before you mentioned attrition and tracking attrition. I think that's more important today than even, pipelining and and having a, you know, an eyeball fixed on, adding new memberships. You you gotta keep an eye on the existing members. One, it's way more expensive to acquire a new member. And if you can just, you know, get some insight into what's happening with your members when they're on the way out the door. This isn't an instant thing. Some it takes time. You know, you see direct deposits slow down. You see, you know, debit card activity decrease. And if you can get ahead of that stuff, you know, it's it's more cost effective to just hang on to them, you know, even if it just it it's just a matter of generating an email or two or you know, an outbound phone call, anything. So I love that. you you mentioned repeatable, and I'm curious from a a core perspective, one of the I think the challenges that we have in the the movement is there are a small subset of core providers that have a disproportionately large share of the market. does that make it easier or more difficult for organizations like Nook to exist? Is it have you have you found that it is a repeatable process for any premiere client, any DNA client, any you know, scimitar client, is it rinse and repeat in a lot of cases or are are are the credit unions shooting themselves in the foot with over customization and and just being completely difficult with their data?
Austin Wentzlaff (Nook): I think I think there's definitely some of the the latter scenario there. But I think, you know, we know everybody uses Scimitar Episys differently, you know, and Pfizer DNA and all the core systems. They're not all identical in their use case. Like in Scimitar, you know, in particular, I know it to be 70% standard, 30% custom fields. So those 30% custom for sure the Wild West. But even the 70% standard, we still use those kind of differently. You know, we have different ways of populating the data in there. So that is those are nuances that we have to deal with. But the the good news is we have been able to find an ability to standardize those and to make it to where we can get it into our data model in a in a nice clean standardized way. And then also have that that CRM or the HubSpot architecture match that as well. So our scimitar to Nook data model doesn't look all that different than Pfizer DNA to Nook data model. I might be mistaken here. I'm pretty sure Pfizer DNA is a a member centric data model. Scimitar Episys is an account centric data model. I know that one for sure. those are two different ways of looking at things, you know. So we as you know, the the partner that's saying we're focused on RevOps, which RevOps is sales service and marketing. we're focused on those use cases, those are all member centric use cases. Account centric
Jason: Mm-hmm.
Austin Wentzlaff (Nook): doesn't do you any good there because the marketers care about what does this member need from me? How am I gonna talk to this member? So we have to take Scimitar and make it more marketing friendly, you know, and we do that with every system that we work with, but that all goes through the standardized Nook data model that we've that we've developed.
Jason: How much of the engagement that you have with credit unions is trying to find a nice way to say this, unwinding in some cases, decades worth of probably not the right word, misuse of the core and you know having multiple persons in, you know, let's just say in Episys. having multiple person records and set episodes that are all slightly different. Some have social, some don't, some have first name, some don't, some have middle initials, some don't. How much time are you spending, you know, in at a typical engagement with sanitizing the data and cleaning it up and doing a lot of that normalization?
Austin Wentzlaff (Nook): Yeah, we've made a pretty important business decision that is that we are not here to fix all of the data problems that have been generated in the last, you know, two decades. but we are aware of them and we will make we will make you aware of them. And we also are starting to put processes in place so that you can work around them. So, like a good example is we use SSN as our unique identifier. We don't send SSN into the CRM that stays behind. That's another important part of our our architecture. it but we can use it to derive a new unique identifier in there. But in some cases, even at even though SSN is a really, really good unique identifier, it's not perfect. So we're starting to put in indicate indicators that say whether or not we're a hundred percent certain that this is the right, the right person, like a verified check mark, you know, kind of to way to think about it. like your Twitter verified or an unverified. So that when a when a
Jason: Yep.
Austin Wentzlaff (Nook): When a person's using the data, they can go, yep, this data is with absolute certainty. We know it's we know it's correct. and there's a lot more, you know, examples of that, or where we come across data and we say, Hey, just so you know, we're not very confident of this. You shouldn't be either. ideally you would fix it in the core, but where you're probably not going to. But that is actually a thing that we are now seeing an uptick from when our clients really appreciate that about us, is like we are calling out. Hey, there's some, there's some crap going on in your core that if you fixed, it would make our next load easier. It would make your experience easier. We can verify this data better. And some of them are starting to think about how do we use Nook as a data quality tool, essentially. You know, not a not an add-on service or anything, but we're gonna take the the output and we'll go manually go into the core system. You know, I think the dream state is that we would be able to do a bi-directional sync and that you could fix it in HubSpot and then sync it back to the core, which I know you can do. we just haven't got that far yet. We don't have any any client pushing us pushing us too hard on that exact use case yet.
Jason: So w would it be fair to say you you tend to focus on accuracy over precision? And for the most part, accuracy is good enough. If you've got some, you know, some not quite straight data, you can live with that because, you know, it's an outlier. You don't have, you know, sixty thousand data quality issues. You have a handful and it's not really gonna be a material problem. it might be for that one particular member who may get you know, two pieces of marketing collateral sent to them via email. But at the end of the day, you know, you're still serving fifty-nine thousand nine hundred and ninety-eight members really, really well. It's just that one member that has, you know, maybe a couple of emails that they shouldn't get and you fix it and you move on.
Austin Wentzlaff (Nook): Yeah, exactly. You're never going to get a hundred percent. And if you try to, you are going to forever live in data quality resolution, you know, which I've seen I've seen happen in the past, and you just never move forward. So you kind of have to get to a point where it's like, and also I think the other thing too is it's nothing we're doing is creating a new problem. It's just it's just saying, hey, this problem exists in your core. So Nook found it, but had we not found it, you would have still acted on it.
Jason: Yeah. It's still a problem. Yeah.
Austin Wentzlaff (Nook): And I think so another good example of that too that we do solve for. So we don't we this is a data quality resolution thing that Nook provides is say like the type case. So if you put in JSON in all caps in Scimitar, that's gonna come over to HubSpot or it it would in our Nook data model. We've created logic that says, nope, only the first, you know, only the first initial should be
Jason: Mm-hmm.
Austin Wentzlaff (Nook): you know, and change that. But let's say that person has like a, you know, like an O'Brien type of name where it's first and second are both capitalized. Well, you your logic can only be so intelligent because how often is the full the full caps name a name like that, you know?
Jason: Mm-hmm. You could probably do it with a a re a regular expression, but again, accuracy versus precision. What's is it worth the extra compute resources to go through and say, hey, first letter, but if you see you know, an apostrophe in that situation, the next one also is capital. And then some cores like apostrophes and some don't. So some have it and some don't. Some are sometimes there are a space. So you could solve for it all if you really wanted to, but What's the, you know, what is the point at the end of the day? You've got one or two names that may look slightly different. And if they're in all caps everywhere else except for the email campaign that you happen to send the member, I think that's probably okay. so you had mentioned, some of the marketing marketing automation pieces. I want to give love to all of the departments. so for a CRM, where do you see like what percentage? I don't know that that's necessarily a fair way to frame it. But there's service, there's sales and business development, and then there's marketing automation. W for a credit union, and and this is gonna be different for different credit unions, but generally for credit unions, where do you see how do you slice up the big CRM Pi between those three groups? And did I miss anybody?
Austin Wentzlaff (Nook): No, I think you hit it pretty well on the head. I think my my world might be a little skewed in the sense that I'm dealing with HubSpot and HubSpot tends to be a marketing platform first, you know. So much so that a lot of times when I'm talking to clients, they're like, Well, when do we do the CRM project? And I'm like, You already did it. It's it's the underpinnings of HubSpot is the CRM. So there is no extra CRM project, you know, but it's But it's because that kind of old school logic of CRM equals sales. You know, it's a sales tool. So like when the marketing team gets this HubSpot platform, they're not thinking about it from a CRM standpoint. So I think in our case, CRM is owned by marketing to start. and I think it does make sense that they own it because a lot of the inputs or the data that's in the CRM is data you would want to be able to. Build automations and real-time trigger events and those things that would go out from a marketing perspective, emails, SMS, you know, that type of stuff. but there's other criterions that are adopting Salesforce. And Salesforce is much more you know, it's much more brought in by salespeople first. And then marketing, you know, there's an add-on called Salesforce Marketing Cloud, but a lot of times. Those clients don't have marketing cloud. They also have HubSpot, you know, or they have a different marketing tool, you know. So I think it's an evolving, you know, who's who who who owns this is definitely a changing. but I definitely think everybody should be involved in it. And that's why we talk about Nook as a RevOps company and not a marketing company. Revops is the interconnectivity of sales, service, and marketing through technology and data. So we want all of those players at the table owning their their piece of it, not HubSpot being a marketing tool for as an example.
Jason: Uh-huh. So you didn't give me a percentage. So I'm gonna push I'm gonna push back because one of the first points you made was one of the differentiators is the relationship. So take yourself out of being the CEO of Nook, but also having 10 years, you know, of time served in the credit union movement. Where do you see how do you see the breakout of those three different business units using the CRM product, as that differentiator that you started off the podcast with, as our go forward. Like I think we're all agr in agreement. Rate, good luck with that. But relationship building, I saw a survey fairly recently that for the first time in I think forever, credit unions no longer have an advantage at at trust and customer service that they once did. So how do we reclaim that crown and is it shifting some of the percentage from service and business development away from marketing? Or is it are we 33 and a third percent across the board and, you know, there might be some differentiation from credit union to credit union. give me what give me your thoughts on that.
Austin Wentzlaff (Nook): Yeah, I think it's it's probably like a sixty to seventy percent marketing right now. And then like a twenty percent sales, business development, and then a zero to ten percent service right now, really. I think the numbers should be thirty, thirty, thirty or thirty-three. I think, you know, maybe one group should have a bigger percentage because someone should always own the thing and you can never do that with you know complete democratic democratic you know, type of situation, I don't think, but
Jason: Wow. Okay.
Austin Wentzlaff (Nook): But yeah, I think like I'll give you a good example of where I think these three groups should come together in a real-time automation. So credit unions today will send out surveys, like an MPS survey, and a member will give you a two out of ten. I am not happy right now. Your service is terrible. The last branch visit I had was was not great. What happens usually? Usually that surveys
Jason: goes in box and it stays there, it shows up in an Excel spreadsheet and once a year somebody will look at it and go, Huh, we're a three point five and we used to be a three point six. What happened? I don't know. Move on.
Austin Wentzlaff (Nook): Right, exactly. It's usually an endpoint. It's we did a survey and it dies, you know. But what should happen where where I'm hoping the industry will go is sales service and marketing are all sharing ownership in this CRM. So when a two out of ten happens, multiple things happen. One is an automated email from marketing goes out saying, We're sorry you feel that way. We we can do better. This is not the standard that we want for our credit union.
Jason: Yep.
Austin Wentzlaff (Nook): All that warm and fuzzy kind of apology type of stuff. Automated trigger. At the same time, you can also do an automated trigger to update that CRM record. Put an unhappy member flag in there. So anybody that logs into it knows this person's not feeling good about us. You can also build a trigger that says, do not send this person any marketing, you know, related information. So these are mostly marketing things first. Secondarily is They might be in a sales process right now. They might be going through a loan origination process. Sales should know about that, you know, or sales should know to not try to enter them into a new sales process because that'd be like, you know, hey, I'm really unhappy. Okay, would you like to buy another thing from me? It's like you didn't hear me at all. You're not listening to me. You know, so you want to tell that sales group, do not sell. And that can be done with that sales flag. It can be done with automated internal notifications, all of that. And then the same thing on service, somebody on the service team should be notified. We should probably pick up the phone. If it's that bad of a member experience, they just gave us a two out of 10. We should call them and say, hey, I know you got an email from my marketing team kind of apologizing, but I personally wanted to reach out to. So that's like equal ownership of that one thing, or sales, service, and marketing all should and can be involved involved. The problem is in the past, that's impossible to do because of what you just said.
Jason: Mm-hmm.
Austin Wentzlaff (Nook): That's a survey that now turns into an Excel spreadsheet. In order to take that Excel spreadsheet, I now need to go to three different groups and tell them manually what to do next. And like that's where all falls apart. You're not gonna do that. And I think, you know, kind of same thing with some some data analytics where it's like, hey, we have report, but someone has someone has still the grabs that report, do the thing with it, you know. And I think the new era that we're in is all of what I just described is not science fiction at all. We can easily set that process in workflow up.
Jason: Well, your example I think is is spot on because, you know, I over the years at multiple credit unions, whether it's Radden or Baker Hill or, you know, any number of third party data analytics groups that go through and survey members and look at act transaction history and all that sort of stuff to give us a a sneak peek or of a a view of our membership and satisfaction and things like that, you know, those are single event, you know, activities. We get a report, we look at report, we, you know, pat ourselves on the back or, you know, wring our hands and say, We need to do better here. And then twelve months goes by and we rinse and repeat. We do it over and over and over again. I think CRM and some of these tools, certainly looking at OKRs and KPIs and and having some sort of data layer data layer that gives us insight into our memberships actual day-to-day. I think helps surface those things that right now for most credit unions are annual reviews and and I I would say at maybe high performing credit unions, they're quarterly at the most. They're certainly not day-to-day. They're certainly not instantly reactive to a negative service, you know, incident. so so the all the stuff you mentioned, I think, you know, you you made the point of saying it's not science fiction. So let's talk a little bit. science fiction. Let's talk about five years from now. Or, you know, if you're a progressive credit union that has, you know, maybe some resources and some talent, maybe is, you know, a little bit more near future, tomorrow, next year, next month. let's talk about some of the agentic tools that we're starting to see that to one of the points that you made earlier, traditionally the CRMs have been a a one way tool. You know, it's a it's a It's a window into other systems of records data so that, you know, we have, you know, that you could always you just call it single pane of glass, whatever you want to call it, to take a look at data from disparate data sources. but with some of the agentic tools and certainly with some of the advances in low code, no code and all those sorts of things and drag and drop and you know, clawed code and MCP and all of that, you know. acronym SOP, what does the future look like? are we going to start to see maybe some of those percentage points shift away from business dev and move to service? I I think marketing automation is probably here to stay because we still need acquisition. We still need to be able to reach out to the members, but are we looking at the the renaissance of the service CRM or or CRM as a platform? For service for credit unions.
Austin Wentzlaff (Nook): Yeah, I think so. And I think the reason is because again, like I said earlier, the problem with the promise of data warehousing in the past and CRM of the past is somebody still needs to act on that. And in order to sit down, to make sense of it, to take the action, it's a lot, it's a lot to do, you know, and it's a lot for someone to think about where is an agent that you can set up to to do that that activity. they can just do it when it comes up and and you know make these things happen. So I think the the future, you know, is we're gonna first go to a hybrid model, which is our CRM, our data analytics are designed today to be consumed by humans and acted on by humans. So we're not in hybrid yet. Hybrid means we're gonna start to introduce agents into the mix. So humans and agents will work side by side within this CRM data environment. And then more and more as we go into the future, I think the humans will start to become less and less involved. Like I actually envision a situation where people stop logging into things like HubSpot. Like there's not a need to. I can interface through it with Claude as an example. And that's, you know, or I just set up an agent and I have enough automations built that the human doesn't need to. I think that's what you're seeing obviously Wall Street respond to if you look at the the SaaS apocalypse that they're calling it, if you look at the stock price of Salesforce or HubSpot or any of these companies that have had seat based models, that's what they're saying. They're saying you built your business around human butts sitting in those seats. And we don't know if there's going to be a human sitting in those seats anymore. So what happens to those seats? Do they go away? Can you sell a seat to an agent at the same price?
Jason: Uh-huh.
Austin Wentzlaff (Nook): Or does an agency cost five bucks where it used to be 150? Like, I do think that's where we're that's where we're heading, is that you're gonna same thing again. I'll I I have a obviously a deep data, you know, analytics background. So I think the same thing is true with like the data warehousing realm is what most data warehouses have today is a lot of reports and dashboards and things that you're supposed to go into to make sense of, to then take the next action on. I think reports and dashboards go away. I don't think you need them anymore. I think you say, hey, Claude, or whatever tool you're going to be using for you know your your agent tool, you go in there and you say, you make sense of the data, you'd report back to me. And we don't need to have this library of 500 reports sitting there that 2% of them get used.
Jason: Yeah. Yeah, your your your butt-in-seat license is gonna pivot over to a a resource consumption model. So, you know, now you're gonna start paying for compute and storage and transfer and I I I'm not a fan. It's just it's so much harder to forecast and you know, especially when you're in an environment where the butt-in-seat model didn't necessarily get the same or the expected level of use. So That at least is a very fixed relative cost. You know, I know it costs twenty dollars for a Power BI Pro license. I know what that cost is per month. And whether, you know, Jane uses it or doesn't, it's twenty dollars a month. Now, if Jane, you know, we have this, big movement, we decide we're gonna do a lot more with data, Jane's g all of a sudden just gonna be data crazy and my snowflake bill's gonna go through the roof because now we've got a a ridiculous amount of consumption because Jane wants every piece of data and probably doesn't need it. But yeah, it's it's gonna be it's gonna be interesting for the next few years. You see a lot of this anecdotal stuff and some of it not anecdotal about token usage going through the y roof and organizations getting half a million dollar, bill for token consumption and Claude just because they gave it to everybody. I I that scares me a little bit. so we We definitely need to make sure we've got the right guardrails in place when we move forward with some of these things. So
Austin Wentzlaff (Nook): Yeah, my my hope is that we'll get to more of a success-based pricing model where you actually pay for real action that that gets created, you know. And HubSpot's actually experimenting with that right now on their prospecting agent. Like if the agent doesn't deliver and doesn't doesn't I don't know exactly what the criteria is that that justifies success, but basically you don't pay unless it unless it completes this action successfully. I think we'll be able to get to a state more like that where it's not just
Jason: Interesting
Austin Wentzlaff (Nook): How much are you consuming? But it's how much did we actually produce in value for you? I don't know. That might be a dream state too. You know, I don't know that we can do that, especially for most use cases. There's I think prospecting is an easy one to say, yep, it was successful because this prospect came into our pipeline and the agent was responsible for doing that. But a lot of use cases are not that easy to define, you know, and you can't just be like, that was successful. It's like, no, there's 10 other things that need to happen before
Jason: Mm-hmm. That's no.
Austin Wentzlaff (Nook): wait before success happens, you know? And you should pay compute power for those while we're doing Yeah.
Jason: Absolutely. So we're we're we're touching on the the business side of CRM and and living in this world. Let let's talk a little bit about specifically your interactions. You know, I'm I'm gonna have the same question of of all of my participants on this series. what what what challenges do you run into on this sales cycle trying to Help credit union that's already expressed some interest in moving forward with something. They may not know even know what they want. I guess my first question is, is that a challenge? And how do you feel credit unions are prepared to ask the right questions in order to get the right product from, in your case, Nook?
Austin Wentzlaff (Nook): Yeah, I think right now we're experiencing not a lot of friction in the sales process. Like we're actually kind of going gangbusters right now. And I think the industry has finally kind of turned, you know, turned and realized that this is a high priority. We do need to do this. I would say the challenge that I see though is more of a longer term expansion of the software and usage of it. And what I mean is I think A lot of these implementations are starting off with maybe five seats, but that credit unit has 500 people. So when do you get this to be a a true enterprise CRM tool where the whole company is logging into that thing? most credit unions that I'm working with today are ta taking that as a multi-year journey. And I do think it should be a multi-phase journey. I don't think you should try to do this all at once. I don't think you're ever gonna get the culture adoption to like be like, hey, you know, you and the branch, you're now using. HubSpot tomorrow. Like you need to first get comfortable with it, get comfortable with the data. The enterprise needs to get bought in, and then you start to roll it out. So I do think that makes sense. But I think, you know, as with most things credit unions do, we're taking it a little slower than we probably should. You know, we should probably be looking at let's do this sprint for the first three months, then let's get that next, you know, tranche of seats in here for the next three months. whereas in a lot of cases it's like that next set of of
Jason: Mm-hmm.
Austin Wentzlaff (Nook): new seats or you new expansion of the software doesn't come anytime soon, you know, and it probably takes a little too long. so I think people know they want it, they know they need it, but the a lot of because a lot of times this is what happens in my my sales conversations with HubSpot involved is the cred union will say we are we're a cred union of 250 employees and our aspiration is to get all of these people in the CRM platform.
Jason: Uh-huh.
Austin Wentzlaff (Nook): and in HubSpot no to no fault of their own, they go, that's a lot of I'm excited about this. Yeah. And I and I have to explain to both, you know, both groups, yeah, like, hey, that's a great aspirational goal, but when are we actually gonna get there? And what's the real starting point? And more often than not, the initial we thought it was gonna be this many seats gets cut down to ten percent. So it starts at ten percent of where we actually want the organization to be. and I just don't see that extra.
Jason: Two fifty. Awesome. Whoa. Mm-hmm.
Austin Wentzlaff (Nook): Twenty percent, thirty percent, you know, I don't see that happening at the speed in which they said it was gonna happen when we first met with them.
Jason: What what do high performing credit unions who've who've gone who's who have engaged with Nook, if you could if there's one thing you could tell your next prospect, you know, you've got a a gold star perfect client, implemented CRM, is now, you know, it has 500 seats, is firing on all cylinders. What did they do? You know, what is one part of that implementation with with that credit union that you would if you could make repeatable for every one of your engagements with other credit unions.
Austin Wentzlaff (Nook): I think it it would go back to what I said, which was the the phased rollout approach. You know, I think they need to be cognizant of what they can actually handle, you know, and then get and then figure out what those goals are and milestones for them to feel like we now have confidence in our data, confidence in the platform. We as a, you know, initial group of users feel like we can train the trainer and help the next group, you know, and then you roll it out to the next people. And those people are responsible for helping them get onboarded. And that way it's just kind of more of a domino effect than than trying to boil the ocean. Cause I think boiling the ocean is just gonna result in a disaster, you know. And I think we s we do see that happen sometimes where they do think they're ready for the 200 seats right away. And it's like you might be, I might be wrong, but what I what I think is gonna happen is Nobody's gonna take a lot of ownership. No one's gonna be able to get that full buy-in and it's just gonna kinda flaunder there and you're gonna be paying for two hundred seats when you should be paying for twenty, you know.
Jason: Absolutely. What a a lot of credit unions now are in the sales cycle for renewing or conversion to a different core. what are some things that credit unions should go into those negotiations? You know, knowing that they have aspirations for down the road, looking at a CRM. let's Just assume they're not going to stick with whatever their core is providing. Let's start there. what conversations should they be having with their core vendor about things like SIM exchange, about core API for DNA, you know, about access to the data? what's some advice that you would give credit unions to have those conversations so that they've got the tools that they're going to need when it comes time to start looking at that integration with HubSpot?
Austin Wentzlaff (Nook): Yeah. Well, for the first thing I was gonna say is yeah, the contract negotiation tell them, no, we don't need that CRM. I I get that it's discounted heavily for my 10 year contract with you, but you still don't need it. but no, I think the the data access thing is actually a it's a pretty serious problem in this industry. It's something that I am still really surprised happens today. Like we're some cores are better than others for sure, but there's some cores that we work with that They are beholden. The credition is beholden to them to get them anything. And we're working with one right now where the journey is going to be two months, two months before you can get the extract files to us. It's like, I can get those immediately from Scimitar. They're there every single day. They show up every morning and I know exactly where they live. So like that's that would be a demand, you know, that I would have as a credit union to say, how am I going to get the data? What access to I have? How do how beholden am I to you? You know, what does that look like? Because if you can't extract your own data, that's that's a problem. You shouldn't be on that that platform. and I do hope anyway, and this is where I'm gonna get in trouble now for saying this, but I do hope that those core systems that are doing that are not gonna be here much longer, or at least they'll change their ways because it's their data, it's the credit union's data, and they're saying, No, no, no, no, you can't extract it. We have to extract it for you. Which
Jason: Yeah.
Austin Wentzlaff (Nook): I wouldn't have as big of a problem with if they could do it in a in a day or two. Two two months, you know, like this is where I get frustrated and I'm like, I bet I could I bet I could go into that core and I could throw clot on it right now and I could get that data tomorrow.
Jason: Yes. Yep. You think you're frustrated. I'm a client. I feel the same way. I I'm you know, it's easy enough to find out who my core provider is, but w we've had similar issues, even in scenarios where there are, let's just say, canned products available that will extract the information from said core in a format that's recognizable and supported and standardized, you know, to take eight weeks, nine weeks just to get that delivered.
Austin Wentzlaff (Nook): Yeah.
Jason: installed, tested is insane. So I agree. There's got to be better ways to do it. What what are your thoughts on, I know CRMs tend to be a little bit more batch, you you tend to go back 24 hours, 30 days, you know, depends on what you're doing with the data. I would say, the sooner the better. But what are your thoughts about real time access using things like, core API and some exchange and stuff like that from some of the more modern core providers or partnering with you know folks like Portex that have some pre built, you know, API connectivity to some of the cores. What are your thoughts there? Is API overkill?
Austin Wentzlaff (Nook): Yeah. I think it's overkill right now, but I don't think it will be in the future. So I do think that we're we're moving in that direction. I think the most clients ask me about it, you know, before we get on. They say, What's your what's your data load frequency? Cause we want real time. Cause as you you said earlier in this, there's a lot of marketing buzzwords and real time's absolutely one of them. Real time, real time, real time. And I'm like, Okay, for what? You know, so we so I kind of go through and like, if you can give me a real concrete answer, I'll in you know, a Janesey or a Portex or one of those companies that have it, because we don't, you know, and that's not what we're building our model around. We're building on the foundation data that you need. And nightly batch is usually pretty good for almost all of that. but if the moment that a credit union says, no, I absolutely need it and here's why I need it, then we'll go down that path. And I think my view of it is going to be it's still going to be mostly batch. So it's not, it's not an all or nothing thing. It's not gonna be all real time or all batch. It's gonna be mostly batch because most use cases do not need to be real time. And then you'll work with a Portex or a GNC and say, these five data elements, I need these called in real time. So give me those things. And then we'll build that in in parallel to our model. so yeah, that's kind of where my my thing is at. The same thing is I mentioned it earlier, but bi-directional syncing is another thing we get asked about. I think that's
Jason: Mm-hmm.
Austin Wentzlaff (Nook): should come sooner than later too. I do think it's a real good goal, you know, and I would love for us, like like I said, we're identifying a lot of data quality issues. It'd be cool if you could use because HubSpot has some data quality tools as well. So if you use their button that says fix this data, it doesn't fix the core. So my data load's gonna override that tomorrow because it didn't get fixed in the core. So
Jason: Mm-hmm.
Austin Wentzlaff (Nook): those tools are kind of rendered useless. Whereas if we had a bi-directional sync, you could say, yep, fix all of the, you know, the state short codes because we're using I N N and N and Minnesota fully spelt out and like we're doing everything, you know? So you can fix that. And HubSpot recognizes it. But again, you need something to tell the core, hey, this was fixed, you know. So so it's coming. It's, you know, that's not a that's not a a sci-fi thing. It's just I haven't had a
Jason: Absolutely. Mm-hmm.
Austin Wentzlaff (Nook): a credit union and tell me with certainty like this is this is the use case we're spending more money on. 'Cause that's really the what's that's really what it boils down to. The technology exists to do real time and bi directional. It's just it's more expensive. It's quite a bit more expensive actually to do those. So
Jason: Mm-hmm. Yep. Absolutely. So if if I'm a credit union and I'm looking for information on let's just say Nook in gener and HubSpot, you know, where will I go? Where should I go to get more information and and consider what my first step in this journey is gonna look like? Yeah.
Austin Wentzlaff (Nook): Yeah, I would say, I mean, obviously this is a biased response to my own benefit here, but I would say come talk to us first. You know, I think the the big thing is, you know, we love we love HubSpot. They're a phenomenal partner, but they're not built for credit unions at the of the day. You know, and and what ends up happening is a lot of times the sales process will start with HubSpot directly because the credition will say, We want HubSpot. They don't know that Nook exists. They go to HubSpot as as you would, and then HubSpot ultimately pulls us in and says this is our.
Jason: Mm-hmm.
Austin Wentzlaff (Nook): union experts, we're gonna bring them into the conversation. But a lot of times I'm five steps removed from where the conversation started. And we would have been able to have them ask some of some questions differently and start the process differently, scope the, you know, scope what the total you know software package is going to be a little bit differently. And we do help with that before any contract execution gets, you know, done and everything. But I just think having that partner first goes a long way. and I think another thing is just it's an education thing. I have to figure out how I educate the market more on this because a lot of people perceive it as well, if I bring Nook in, there's an extra cost. Cause obviously Nook's gotta make their cut, you know. And the the fact of the matter is that's not how it works. HubSpot compensates us, but but the credit union has no extra fees. In fact, the HubSpot contract actually goes down if you use a partner. They waive their their onboarding fees, which then you know get replaced by us doing a
Jason: Mm-hmm. Sure. Yeah. Yeah. Yeah.
Austin Wentzlaff (Nook): specialized, you know, onboarding and implementation. yeah. So it's no it's no extra cost. We're here to be your partner. We we're the ones that come in and say, Hey, yeah, two hundred seats sounds great, but I don't think you're ready. Are you ready? No. Okay, it's twenty seats then, you know.
Jason: Mm-hmm. Yep. Well, and and the benefit of getting out in front of it, you don't have to unwind, a a first swing at at integration or building out what's in HubSpot. You can start, in in you could start on better footing with someone that's got a little bit more knowledge about credit union data than than HubSpot by themselves would or s a marketing third party may have. So the I I definitely see the value in someone that's got ten years worth of experience specifically dealing with this and data warehouse and and credit union data, which is somewhat unique. I would definitely agree with that when you start talking about members and persons and accounts and orgs and joints and, you know, authorized signers and trusts and et cetera, et cetera, et cetera. And, you know, unwinding all of that is a project in and of itself. So it's great to do that right from day one instead of having to go back and redo it. So
Austin Wentzlaff (Nook): Yeah, you I think the biggest thing here is, well, we're again, this is another education thing for me, but you just hit the nail on the head, so I have to double down on it here. It's like they don't realize how important the partner is in this decision. The credit union thinks that the sole decision they're making is about should I buy HubSpot or should I buy Salesforce or Creatio or whatever they're looking at. And the reality is those are only going to be as good as the data foundation that was built on that on top. That they were built on top of, you know, and I think that's where actually, in fact, the bulk of our clients today, Nook's growth has mostly been creditions that were set up on HubSpot a year or two ago that were not set up on a good data foundation because the partner didn't have the same level of data expertise and and history that we do. And then they get a then they hit a scalability wall. They say, I want to integrate my next data source. Well, you can't. You don't have a good strong identity resolution logic or unique identifier. And they want to
Jason: Sure. Right.
Austin Wentzlaff (Nook): expand into sales and service because it should be a RevOps platform and not a marketing platform, but it wasn't architected that way, you know? So it's the decision should be which platform are we going to go with? Yes, that's a really important one. but then equally important, it's which partner are we going to pick and who has that data, you know, expertise and who's truly building this for credit unions. You know, we're the only QSO that does this and we're the only ones that have our own proprietary data platform too. So that's kind of what sets us apart is we know we know how to do this and we know the nuances and the crediting in data is the nuance. It's the problem, you know. Yeah.
Jason: It is. Yep. Absolutely. Well, this was awesome, Austin. I appreciate, your ten years of service in the movement and everything that Nook is doing, you know, to support the movement. I absolutely appreciate your time here.
Austin Wentzlaff (Nook): likewise. I think that panel should be fun. You don't you don't get a lot of your quote unquote competitors together to chat, you know, and talk shop about what we're building the industry. So that should be fun.
Jason: Mm-hmm. I Austin, I very much appreciate it. And you know, I look forward to our next session together. So thanks again.
Austin Wentzlaff (Nook): Yeah, thank you.