Getting Your Data Set Up So You Can Make the Most Out of It (From 6/10/2025)

Mike Gibb hosted a webinar on setting up data for effective collection operations, sponsored by Prodigal. Experts Aaron Applebaum, Jeremy Jobson, Amruta Joshi, and Raj Khandelwal discussed the importance of data quality, governance, and strategic goals. They emphasized understanding customer behavior, segmentation, and the need for unified data. Key metrics include liquidation rates, cost to collect, and time to collect. They highlighted the significance of real-time data analysis, post-mortem evaluations, and AB testing. The panel also stressed the importance of compliance, engagement data, and the role of AI in improving data accuracy and efficiency.

https://vimeo.com/1092217556?share=copy

Transcript Mike Gibb 0:04 Gibb, Hey everybody, thanks for joining us today. My name is Mike Gibb. I run account recovery.net, and I want to welcome you to today's webinar where we're talking about how to get it get your data set up so that you can make the most out of it. Analyzing data is a lot like, I think, like trying to build something that you bought at IKEA with the instructions and done in the right order. It's a piece of cake without the instructions and trying to assemble it on your own. It's not going to end well for anybody. Data is incredibly valuable in today's collection world, but it needs to be organized and stored and maintained the right way, or it's just going to be a lot like losing the instructions. So today, in this webinar, which is being sponsored by prodigal, we've got a panel of experts here to break down how to set up and manage all the data inside your collection operations so that you can get the most out of it. As I mentioned, this webinar is being sponsored by prodigal. Prodigal is building the next generation of AI powered and agentic workforce for collections and servicing, leading the innovation in the consumer finance industry, with more than 100 companies served in the last seven years, prodigal is leading the way in transforming how collections and servicing are done with artificial intelligence. You can learn more by visiting them online at prodigal tech.com Let me introduce you now to our speakers. Aaron Applebaum has spent the bulk of his career working to help companies unlock the power of their data to solve complex business problems. He's a relative newcomer to the arm industry, joining blinton gains at the beginning of 2024 as their digital transformation manager. Since then, he's brought a data driven approach to help optimize business processes and drive innovation across the organization. Jeremy Jobson is a seasoned professional with an extensive background spending more than 10 years in the collections industry, his expertise in companies encompasses various roles within leading lending institutions, including banks, FinTech entities and medical institutions, currently holding the Vice President, position of Vice President risk collection, strategy and analytics at lending point, Jeremy orchestrates the implementation of cutting edge digital first strategies in a collections domain. Throughout his career, he has a definitely utilized diverse technology systems to execute a spectrum of strategic approaches, cementing his reputation as a versatile leader in the field. Jeremy's tenure underscores his profound insight into the industry's evolution, particularly in steering collections toward a progressive digital landscape. Amruta Joshi is a seasoned customer success and digital strategy leader with more than 18 years of experience in digital collection space and software as a service at Mercury financial she drives digital roadmap resulting into improved customer outcomes, operational efficiency in digital collections. Previously, she's led a $20 million SAS portfolio at fin by cat about and helped clients advance their customer success strategies. She holds a Masters in MIS and leadership certifications from Wharton and product management certification from the pragmatic pragmatic Institute. Raj Kendel wall leads digital strategy and operations at prodigal where he manages the digital communication strategy and execution for a diverse group of lenders, debt buyers and collection agencies. He plays a key role in overseeing $300 million in annual payments and facilitating more than 50 million in Portal payments. Prior to joining prodigal, Raj gained experience at Boston Consulting Group and Ernst and Young everybody who's watching us live today is absolutely invited to ask questions and submit comments during the course of our conversation. You can use the chat or the Q and A to submit your questions and comments. If you would prefer to participate anonymously, click on the anonymous button in the Q and A and anything you send to us, I promise, is sent anonymously. All right, jumping into our conversation now, let's start with a you know, what are the most valuable data assets for a company in the credit collection industry like we talk about data, every interaction, every attempted interaction, every conversation, everything that a collection agency or collection operation does generates data. How do I start to kind of understand what is the most important pieces of data in my operation, if I'm looking for driving data, you know, using data to drive solutions, what data should I be looking for? Aaron, I'm going to ask you to kick us off.

Aaron Appelbaum 4:19 What a broad question, thanks, Mike for having me today. First, the thing that jumps off to me with that question is, what type of organization Am I at? Am I at a law firm, and am I at a collection agency, and what am I trying to accomplish within, within my organization? Am I trying to execute an outbound communication campaign. Is it a digital communication campaign? Is it on the phone? Am I trying to drive more payments? Am I just trying to drive more touch points with the consumer? You know, there's just so many different things that you could try to be doing with your data. So before I even i. Try to attempt to answer the question of, what data points Am I trying to collect? I think the key thing to understand is, what are you trying to achieve within your organization with maybe it's within your specific role or or vertical within the company you're working for. So first and foremost, I think that's kind of the the key thing. And it'd be difficult for me to really answer, you know, really broadly. But if I were to try to put an answer on it, you know, I think, if you were trying to understand, like, what are the key drivers for your business, or what do you think the key initiatives you're trying to hit right now would be, you know, if it's an agent segmentation thing, you know, understanding or not agent, sorry. Consumer segmentation thing where you're trying to understand where your consumers are at or what type of consumers you have to inform some business decision you'd be looking at consumer demographical information or debt specific debt information, what type of debt Am I holding, whether it's balance or age, things like that. If you're trying to drive some kind of increase in, you know, consumer interaction or touch point, it's, you know, when did I last interact with the consumer? How do they want to be interacted with? Do I have, you know, good quality phone numbers or good quality emails? Those are the kind of things that you'd look at again. And I think the other panelists would agree with me, but it's quite, quite broad and depend, you know, depending on, you know, the type of company you're at and what you're trying to

Mike Gibb 6:32 achieve. Does everybody agree with him?

Jeremy Jobson 6:33 Absolutely, I agree. You know, it is a very broad question, and I almost kind of translated as what is like, the non negotiable data sets that you need, rather than the most valuable, because I think everything is valuable depending on how you're going to use it, or what you're trying to do. But there's certain things that are just non negotiable, like understanding your portfolio. So what is your your lending group look like, and then also bringing in what is your customers like, having that baseline to be able to understand, what are you trying to do, how is the portfolio behaving? So then you can start layering in all of those other data sets, where you bringing in additional consumer information, where you bringing in operational performance data, all of that is secondary to kind of getting that starting point of, what are your portfolio, what are delinquency levels, what are, you know, ages, all of those ages alone, all of those pieces, is kind of non negotiable, and then you can expand from there,

Amruta Joshi 7:34 yeah, and I'll agree, right? It's a fairly broad question. When I think about, I never start with the data, right? Because I don't know what data unless I know what is my goal, right? So I think the first step here is when I always have broken this down in the past, and I said, Okay, what is the strategic goal? What are we trying to accomplish? Right? Oh, I want to improve my you know. How about a collection rate by like, X percentage by going digital? But then you recognize that, you know, okay, so do you collect email addresses and phone numbers from the customers? Oh, maybe, do you have a good data? Oh, maybe, right. So I think it's, first thing is a strategy, right? What is your strategic goal that we are trying to accomplish here? Second thing is, whenever you think about it, I think there is a little bit of method to the madness. When you reach down to like, the raw data is basically your segmentation, right? As you understand your strategy, you need to then look at the segmentation, and how do you want to segment the population into a different buckets, right? So that can then brings it down to the treatments, right? And once you have to me like your strategy, your segmentation, your treatments, figure out, then you can see, how do you want to track the performance? Right? Whether you're going to be doing test versus control, whether you know you're going to be what are the important pieces of data and its collections, right? Let's be real, like, you always look at NCL rates and you look at the liquidation rates, right? You have certain basic things you're going to be looking at like, you're going to be looking at like, pay rates, things along those lines, right? But I just think, like, when I when people, when this conversation started with data Mike, I heard a little bit of chuckle, because I'm like, I don't know where to start unless I know what the goal and a strategy is, right? And how am I going to get into the My goal, right? So I think that's pretty much when I look at it. Now, if you want to, like, since we are talking about like, you know, audiences listening. Let's talk about like, broad set right? Again, going back to collections, basic 101, right? Customer behavior. You're going to be looking at their payment history. You're going to be looking at their delinquency history. You're going to look at like, demographic piece of data, right? You're going to be looking at account level data, like credit utilization. You're going to be looking at credit limit, right? So I think there is certain basics in collections that you will always look at certain matrix. But I think the key is to start with the goal. Question is, what's your goal? And then, you know, how do you take that down to segmentation and then treatments? That's how I have always looked at this question. To be very honest with you guys, if you guys agree, disagree,

Aaron Appelbaum 9:58 I think one. Point that I'm really just pointed out there, I think is an interesting one. And some of these data points that we collect are, you know, how did something perform, or what KPI are we trying to improve? Whether that's our liquidation rate that data is necessary to know is our strategy working or not. But then there's other pieces of data that you need in order to execute the strategy. So if you, if you, if you're focusing on, say, liquidation rate, you can watch how that moves up and down as you're executing your strategy. But in order to actually execute the strategy, maybe I need good quality email addresses for all of my consumers. And so it's kind of, where are you at in in your execution, and what are you measuring? And where is it? You know, there's measurable pieces of data and then obviously, fact based pieces of data that allow you to actually execute a strategy.

Unknown Speaker 10:49 I would agree with that

Raag Khandelwal 10:53 absolutely on board with everyone when they say that the data you need is decided by the outcomes that are trying to drive and what your goal is, right? I would broadly divide that into two buckets. One is compliance, and the second is effectiveness. Right? First and foremost, you need to make sure that you're being compliant. So you need data to make sure that everything that you're doing, all of your communications, your outreach, is compliant. And then the second part is driving effectiveness, where on the lending side you're trying to reduce roll rates, on the recovery side, you're trying to improve liquidation rates, and then you measure that performance by different segments, like Amrita mentioned, and yeah, so you essentially need to see what segmentation is really going to be helpful in understanding where performance is different and what kind of consumers are similar, and then use that to sort of improve effectiveness.

Mike Gibb 11:46 So we were doing a webinar a couple weeks ago, and one of the panelists was talking about Uber and mentioned sort of the, you know, this concept of a North Star. And Uber's north star at its beginning was rides per day. Everything it did was trying to drive and increase the number of rides per day. And so I've been kind of having this, you know, they've been throwing using this phrase a bit in conversation since then, trying to understand, sort of what, you know, what a collection operation, you know, their North Star is going to be. And obviously, I think at the highest level, it's collect as much, you know, collect as much money as possible, collect more than yesterday, or collections per day. But I'm wondering, sort of, listening to, to what I've heard so far, if that's, you know, if that's, you know, I asked a broad question, so is that, you know, but is it collections per day too broad in and of itself, in terms of, like, if I'm looking at sort of, you know, using data to drive, you know, operational improvements and drive more efficiency in make myself, make my operation better. Is right, you know, is collections per day at two symbol or, you know, like, if I'm looking at sort of, you know, I got to figure out what my goal is, to sort of figure out what data is most important. How do I start to figure I'm going to my The first thing I'm going to think of is collecting more money is obviously what's going to be most important. But is that the right is that the right metric to start with, is my question.

Raag Khandelwal 13:13 So if you're referring to third party collection agencies, right, because you brought up recovery rates, and I think there are three North Star metrics that come to mind and the priority between those three changes from company to company. But essentially, the three are liquidation rate, your cost to collect and your time to collect, right? And then on the lending side, of course, you're looking at reducing default rates out of the people who have defaulted. You're looking at reducing roll rates, and then it also tried to maximize recovery percentage.

Amruta Joshi 13:44 So Mike, I think Uber is an interesting analogy, right? And I want to stop there, right? I think in collections, you do need to look at the data every single day. What data do we look at? Is really, again, we've talked about that is really tried to what your strategy is, but I think at the ultimately, like, let's be real, right? You're in collections to collect so you want to make sure that you know you're reaching your goals, like collection goals, whether it is in terms of recovery rates, whether it is in terms of reducing losses, whether it's at a very high level, like, you know your NCL rate, or GCL rate, whatever you're going to be looking at, right? So I think, I don't think that you do look at the data daily, that's how you understand the trend, and you build the trends from like, you know, month over month, but collections per day concept and the Ubers, like, you know, right to a day concept. Are two different things, right? You do do daily analysis of a data is a different thing than, like, you know, just, you know, just collections per day. So I think there are two different things, Mike in my mind.

Raag Khandelwal 14:56 And just to add to that, I think your north star metric also changes. With time, right? Uber, when they originally started, they were trying to get consumers used to this new way of hailing cabs. They wanted to introduce this habit, and they also wanted to grab market share from competitors very quickly, and hence, increasing the number of rides per day was important. But as they looked to sort of go public, as they look, you know, to become a bigger, larger organization. I think profitability also starts becoming important for them, right at which point the North Star metrics that they track also change. So I think it's important to understand your business goals and use that to derive what your north star metrics

Aaron Appelbaum 15:42 are. Yeah, I think, just to touch on, I think it's a good call out to say, where are you in your in your journey? So to speak, collections today can be a little tricky, because there's a lot of things that can drive the pure dollars collected. You know, how many placements did I have this month? Did that change? If that number went up, and then naturally, you may get more dollars, but it may not be because of any, you know, process improvements you've done. So I think that's a that can be a tricky one, but say I'm really focused on digital strategy and my digital roadmap, I may be. It's possible I could be less concerned with how many actual dollars are collected, or even my collection rates, but maybe I'm just really trying to reduce my costs and I want to drive self service collections, then you're looking less at how many collection interactions that I have, but how many of them happened without needing an agent to pick up the phone. How much of that was completed, you know, by the consumer or the customer on their own on our website. If I'm really trying to drive interactions with, you know, digital communications, I'm okay maybe with having more agent interactions in the short term and just to see how my campaigns are performing in the in the near term. So again, it goes back to what's your goal, and you can drive your north star from from that starting point.

Jeremy Jobson 17:04 Yeah, but agree, I think the North Star can move, move around. I think sometimes you can over hype. You can hyper focus on that North Star and forget to kind of lay your guardrails around it. And so, for example, you know, if our North Star as cures and to prevent. You know, charge off well if we have programs, and then all sudden, we're using all of our programs to be able to get create those cures. And now we're over using those programs, or if you're if your goal is RPCs. And so then you end up over dialing because you're just trying to drive RPC. So you have to not just have a North Star, but you need to kind of lay out what are your boundaries for that North Star as well. How do you get there? In building that roadmap,

Mike Gibb 17:54 someone did put a comment in the in the Q and A as well about the compliance measures being also important. Throw that out there as well, but also use it as an opportunity to remind everybody, if you're just joining us, use the chat or the Q. A would love to see questions comments from everybody during the course of our conversation today. Um, Gary is the is the quant? Is the quality or the type or the quantity of data changing much say than where it was at the start of 2025 or where it was a year ago, or is, you know, the amount and type of data that's being analyzed the same? It's just maybe a more sophisticated level of analysis that's being done.

Amruta Joshi 18:36 I think, in my mind, I think it's the strategy shift, right? I think it's both. So it's a, if you think about it, and make, you know, there is more and more shift towards moving towards digital overall, right? There's more shift towards going AI and automation, right? I think that's driving part of like, you know what data people are collecting? So if you're moving more towards digital, you're, you're inclined to Aaron's previous point to collect more of an engagement data, right? You're going to be looking at like in an engagement data from a different digital channel standpoint, right? If you're going from an AI and automation standpoint, and rock can probably talk more efficiently about this, but with articles wise agent. But you know, you're collecting a different level of data. So I think the shift in a strategy is really driving some of the behavior in terms of what data folks are collecting, right? In my mind, it is also simple things, right? You have been using apps and you've been using like website for you know, for detail now, right? But now there is more emphasis on understanding customer behavior and tracking that behavior from an a website and app standpoint, to drive more strategies, right? Can you look at that behavior and then write a strategy so you can reach out to the customer as an example, right? So I think it's the way I'm looking at it. And I did publish an article on LinkedIn on this about few months back, where it's I think it's a shift, right? It's a. Shift in the channel utilization, how people are looking to reduce their costs, how people are effectively looking to collect more, and how digitally savvy the customers are, the fact that there is more phone usage, people respond to text messages more. I think it's driving, like, some of the shift in terms of the data that they collect and the level of analysis that is required into in terms of, like, you know what the data that you're collecting?

Jeremy Jobson 20:30 I think for me, there's been a little bit of, I don't know if you guys are in this boat as well, but sometimes I think there's just too much data, and so trying to be more strategic and more purposeful with what data do I want and be able to use, because we could do analysis for days, and it might not yield anything for us. But so I try to be more intentional on what I'm going to bring in and what I want to use with that. And to the earlier point, you know, asking the question, What am I trying to do? Rather than just grabbing everything and anything I can to put my database and then hope I'll use it at some point, been more intentional with what I'm doing. Do

Mike Gibb 21:13 you have an example of that, or is there something you can share about how you sort of, you know how you're approaching that process? Well, I think like

Jeremy Jobson 21:21 even using Bureau refreshes, you know, what's the frequency that I'm getting them? Am I getting value and getting it, you know, on every account, every month, or is there only certain points that I'm actually using it and getting value out of it? Do I really need a bureau refresh on 175 day past due loan. Or can I use what I got last month for it? And trying to scale back some of that, I think, is one example of, you know, asking the question, What am I using actively? What do I want to do in the future? And is that data going to support it or not? And if it's not, then maybe I don't ingest it, maybe I look elsewhere for another resource.

Mike Gibb 22:06 And then Amruta, to your comment about sort of, you know, the the amount of data, or the amount of analysis, excuse me, that needs to be done is changing and increasing. Can you give a sense of, sort of how, or, you know, some example of that,

Amruta Joshi 22:24 yeah, so I'll give you a simple example of that, right? Like, for example, how well, if you are saying basically that I want to improve the collection using digital channels, right? So the first thing that you would need to look at is basically, how engaged are your customers? That comes down to customers like open rate and click rate, and you know, things along those lines, right overall, then that leads you to the customers logging into digital website or your app, right? What do they do after they log into the app? Right now, that's going into little bit of deeper analysis. If you just look at like customer opened an email and logged in, but the customer really was on the website, and we're just looking at updating a contact information, your communications really didn't drive customers to take an action, right? So I think when I say like, detailed analysis, right? And initial stages, like, you know, you could have access to, like, maybe open rates and login reads now you have sophisticated softwares to track customer interactions, actually, when they're on app and the web, that helps you drive like, you know, some of those strategies you can drive a communication based on like, customer being on the website, doing something or not doing something, right? So that's when I say, like, you know, deeper level analysis is not taking place, but I would agree to Jeremy's point, right? Too much data is available, so I think selecting the right pieces of the data to drive that analysis is equally important. And that goes back to the previous point we all agreed upon, which is, basically, what's your goal, right? What's your business goal that you're trying to accomplish?

Jeremy Jobson 23:59 I like the comment about, you know, doing the engagement rates. In looking at that, I think telling the full story takes a lot more analysis than what people understand. You know, like, just because you have a good open rate doesn't mean that you have a good payer rate if you don't connect the full loop. And so, how do you, how do you do that and take the time. Don't just, don't just build the first piece and say, okay, we can measure open rate, but send a bunch of emails now, well, why did pay rate go up? Our open rate is up. It's like, well, because you didn't connect to say the open rate does correlate this way.

Amruta Joshi 24:34 No, exactly. And that was my point, right? Jovi, like you need to look at the whole customer journey. This comes down to truly understanding the customer journey that is driving, like your key KPIs, right? And then what alteration you need to do in the customer journey, which is, like, lot more deeper analysis than previously, people be like, Oh, it. People are opening the email, so let's send more emails. But if that's not driving an action, why are we sending emails? Right? Even if it cost you a few cents. And to send an email. So,

Raag Khandelwal 25:03 yeah, I completely agree with what Jeremy and Amrita said. I think adding to that, there are two other things that have changed drastically. One is the quality of data that you're able to collect today. A lot of data, especially in the 2010 was collected manually and then entered by a data entry assistant into the system, right into a CRM, but now a lot of that data is actually being captured by technology, and the quality of that data is significantly better. I think everybody here would agree that the notes that agents have been taking historically have not been to the quality that we would all appreciate, and then when AI is has started taking notes on behalf of agents, the quality of that data has significantly gone up, and probably the same for disposition codes right at the end of the call as well. We all agree that agents making stakes while entering in the disposition code, so having AI capture those for you just improves the accuracy and quality of that data. The other thing is actually how frequently you need to ingest data back to Amrita point on logging events from your web portal, right? If you can actually get that data in real time today, that's super, super helpful, right? For example, if a consumer tries to log in to your website, and they try and enter, let's say, their account number and date of birth, and that login fails, you could send them an email or a text following that up, saying, Hey, you had issues logging into your web portal. Why not try calling us in to get your query resolved right? So the frequency of that of logging that data has also significantly gone up over time.

Mike Gibb 26:45 Another question here from the audience, are collection apps reasonable as a channel? Why would a customer install an app? I think we're talking for third party collections. In that context of that question,

Aaron Appelbaum 27:01 this is a question that I wrestle with myself, and I don't know that I have an answer, but I think that's a good question to put out there is, like a lot of times, I feel that the industry looks at consumers or customers as just that customers doing, you Know, completing a purchasing transaction when that's really not that right? They're not buying something. They've already bought something. They're rectifying their situation. So a lot of times, I just as a comment, I think we look at consumers as, you know, as they as if they were purchasing something from Amazon or some some web store, when it's really not that way. And that's something I kind of think about, is, how do we look at these interactions, you know, through that lens, but also slightly differently? And maybe I'd pose that to the other panelists,

Amruta Joshi 27:54 in terms of, like apps, right? There they are being utilized to lead, right? You have app, I think it's also the consumer behavior, right? You open an account with a bank, you download the bank's app, right? So I think it's an actual consumer behavior that you're tapping into. And I think the, if I not mistaken, in terms of data, I think we do see the higher percentage of consumers downloading an app, right? It's just an actual I think it's also portion of it is you have access to the phone and the security of it, because app is directly on your phone, you can do like two factor authentication, things along those lines. So I don't remember exact numbers, but I think there are more consumers downloading the app compared to what we see in last 10 years, right? So I think part of it is you're just tapping into that natural consumer behavior. And so I don't know if there is really an hurdle in terms of consumers not downloading an app, but I don't know what job I mean, never run and run into that. But what are you other folks think on the panel?

Raag Khandelwal 28:58 So some interesting ideas that we've doubled with is essentially that if you want consumers to download an app, you need to provide something of value on it, right? There should be a reason for the consumer to download that app, right and open it. And one of the interesting theories I've heard is essentially to provide educational and informational content to consumers on how they can improve the financial situation, right? A lot of the people who are in collection are not not paying because they don't want to, or because they're trying to game the system. It's because they have made terrible financial decisions which have led them into that state. And what they're actually looking for is help to get out of that situation that they're in, right? And an app is an excellent way to sort of provide content for them to understand and get educational content on how they can improve the situation better. And then, of course, you can add other functionalities to the app. For example, you can have them pay you can have them set up payment plans as an auxiliary but just to do that, I think people would be more comfortable doing it on a website. Plan downloading an app to make a payment, so you need to provide something more than just making payments on the app to incentivize them to actually download

Jeremy Jobson 30:11 it? Yeah, I would agree it. Having an app that just processes payments isn't going to bring much value. I the education piece, I think is is a component for it, but I think having your payment plans in all your programs easily manageable in that app is critical for its success. If you can't do that, then it's not much different than a website. If there's that's kind of where I would see it. I work more on the lending side, so you have a lot more that's done with the app early on in the life cycle of the loan. Kind of end of life. That's where I would see the value.

Aaron Appelbaum 30:53 And that's why I posed the question initially. You know, this is the app in like, a third party debt collection scenario would be one that is designed to be deleted, so to speak. And usually with apps, you want to keep them coming back, but with collections, the goal is to make them go away and never come back. It's kind of like, think of it like a dating app. I mean, they have other motives, but the idea really is to download it, take care of what you're trying to do, and then never come back again. So the app is a little interesting to me, but I do, I do like, you know what rock which you pointed out about, you have to add something else to make it worthwhile. And maybe an educational component is interesting. But if you have a consumer who comes makes their payments and they're done, I don't know that an app does better than a website, but I really don't have any data or anything to back up that. That initial hunch,

Mike Gibb 31:46 just envisioning a consumer with an app swiping left or right to choose to pay debt,

Jeremy Jobson 31:52 hoping the payment plan gets better as it can swipe.

Mike Gibb 31:57 Love that. See, I love the brainstorming that comes out of this. So and Ruta and Jeremy were talking about, sort of the importance of the customer journey and understanding customer journey. And maybe it's not just about, you know, like a single data point along the way. So I'm curious, what are the data points that sort of, you know, that you know, what are milestone data points, or things that I should be pointing looking to on that customer journey, well beyond just whether they open the email or engage with the text message or something like that. What else so, you know, I look at sort of the log. Do they click? Do they log? Do they pay? Is that sort of the process that I should or the journey that I should be following. Well,

Aaron Appelbaum 32:42 I think, from my perspective, I come from blatant gains, is a law firm, so we have a little bit more in the arsenal, I guess, compared to just an agency, right? So one of the things we struggle with identifying is, what drove the consumer to do something? Was it that they got a really great email, or was it that they got served with a lawsuit. And you know, tying those different milestones and timelines together to build that full picture and help inform us of what to do next is something that we're really focused on. So I think, you know, understanding your organization and what are those big milestones that the consumer or customer has along the way helps you inform your decisions.

Amruta Joshi 33:27 So customer journey depends, right? But I think it's a fairly important concept, right? I think when I think about customer journey, I think about customer right, starting from when they were current customer, right? Customer doesn't become customer in collections. It becomes customer when you acquire a customer, right? So understanding the customer behavior starting from an acquisition, you will see certain trends, right? Their payment history would be an indication whether this customer pays or has a tendency to go delinquent, right? As an example, right? That's an important attribute, but QRS and comes back and becomes some current Do you does this risk scores from a customer standpoint, right? That tells us certain risks associated, you know, where does the risk profile of the customer reside, and would that mean that they have more tendency to fall into collections, right? So to me, customer journey starts at the acquisition, and then you look at it like as you progress from one bucket into another, right? How do you communicate with the customer? Differently in bucket one versus, you know, customer turns into a bucket two, and how your communication strategy changes. So I think it starts from an acquisition all the way till you decide now finally you're going to be selling off your portfolio. I think it journey really has, like, a different aspects of it. I think at every step of the journey, you do monitor the data that comes along with it, for example, like, you know, bucket one. One customer might be more inclined towards doing certain behavior, like in responding more to communication during early stage of the bucket versus the late stage of the bucket, right as you roll into bucket. Two, you need to make sure we have impacted your credit and how your communication changes, right? So I think journey, to me, is an end to end piece, from the time customer becomes a customer, till the time you decided to sell off your debt. And I think there are various aspects of data that you look through the journey, right?

Mike Gibb 35:38 So the topic for the webinar today is how to get your data sort of set up. So let's, let me point the conversation in that direction and see and see what each of you has to say. Like we're talking about all this different data. It's going to be coming in from different software, different pieces of technology, different things like that. So, you know, is this just everything gets dumped into a spreadsheet and you start building from there. You know, how what you know, maybe talk a little bit about where the data is cut, you know, where the data comes into and how you start to to build a process for analyzing it, or how you start to build a structure it so that it can be analyzed. I

Jeremy Jobson 36:23 I think it starts with what is your goal? You know, kind of going back to that. I think that's the first step, right? And so with some use, I want to understand my operational metrics. Okay? So then you need to build your data structure in a way that you're able to pull all of that together without, without creating too much noise. And so are you going to look at what are my agents doing? Or do I want to look at it at the loan level, and then understanding that, and then being able to break it apart? I think one of the mistakes people make is they try to do all their analysis from one place or one table or one spreadsheet or whatever. And agent measurements, agent performance is different than loan performance and loan measurements. And so I think understanding what do you want to look at, and then starting to structure your data that way, or how you want to look at it. That's how I've started with, you know, companies I've worked at and building that out is, what is that? What do you want to see?

Amruta Joshi 37:33 I think Jeremy in terms of goal, but I think there is also important thing, is unification of data, right? What I also see is like, you know, people work in silos. For example, you collect the data from like agent channel doesn't communicate with data that you can collect from like your digital channels. Right now, that's a problem. So I would agree. Like, you know, you want the data sets to be separate and in a manner where you can actually query the data and through the analysis and draw the insight. But at the same time, I think the key here is, like having this unified view of a customer truly to understand, and, you know, to understand how customer interacts with a lender, and how does that cost? What do they respond to? And how do they respond to, what's their preferred channel of communication is, you know, well, how do you best? What's your best chance of converting this customer and when understanding like, you know, different aspects, like the time of the day, etc? So I think that requires your unified view of the data, right? You really need to make sure all your data is unified so you can really understand a customer and not look at it like a channel based analysis, right?

Jeremy Jobson 38:44 I would agree. I think that's one of the pitfalls people have, is that your unique identifier for your digital strategy is going to be, you know, customer code, and then your unique identifier for your dialers, your dialer code, and then your unique identifier at your loan system. And so you have to join 15 tables just to unify it all. So I think making sure you're using a single unique identifier across to be able to connect it all is critical. Yeah,

Unknown Speaker 39:12 agreed

Mike Gibb 39:20 question from the audience, can you walk through the data structure around an intuitive such as texting? You have consumer data, what do you need back from your vendor? What other data sources do you need? Etc.

Amruta Joshi 39:37 So, so basically, when you when you're talking about texting channel, right? Think about first piece of data is engagement, where your text message is delivered, right? If you have a link within that text message, then the customer click on the link, right. Next piece of data, and that is not directly, which is an organization, the. You probably would have to collect, what action did customer do after receiving a text message, right? Which translates into, and we talked about some of that, which is the whether they ended up making a payment, Jeremy touched on like payment plans, right? You're talking about collections, right? Did they enroll into a payment plan or a settlement plan? Right? So I think when you look at the text messaging data, piece of it is an engagement data, which is always going to be a key, are people even looking are text messages getting delivered, and are people looking at the text? There is a technology out there where you can actually know, like people are on the phone, and then you know that you know you are actually sending them a text message that is out there, right? So I think there is a piece of inhibitor. Is also an important piece of it is it's great to send a text message, deliver it people to read it, but did they take any action using that? Or how effective is your text message? And you know, what is your CTA? Is the call to action within that text message that is driving customers to take an action. So I think that is a key, piece of it as well, which requires the data, not only from your vendor, but your own organization, and combining that data together to really understand the effectiveness of that channel.

Raag Khandelwal 41:12 Yeah, and then just adding to that, it is always helpful to have a group where you don't send text messages, and then look at your north star metrics across the two groups, the group where you send the text and the group where you did not send any text, and if the volume of consumers on both sides is significant enough, and if all else is equal that there is no other differential treatment apart from texting. A lot of the Delta in those North Star metrics can actually be attributed to the texting itself. So it's not just about the people who you texted, but also about looking at people who did not take that tells you a lot about the success of that campaign. I

Jeremy Jobson 41:49 think the only thing I would layer into that journey is also some of your compliance metrics, like you need to make sure that you're managing your stops correctly and getting those back and that loop is closed, and even, you know, early on scrubbing to make sure that the phone number hasn't been ported or some of those compliance pieces too at the beginning of your journey is helpful as well. And so I think we think about getting all your data set up. Those are two pieces I would, I would include in that I

Mike Gibb 42:23 uh, are should companies be setting up their this data analysis or this, you know, setting up their data so that they're making decisions in real time? Or is this more sort of a post mortem? You know, we did a texting campaign, we did an email campaign, or, you know, we were on, you know, got these customers on this journey. The next time we do something, we're going to do it differently.

Aaron Appelbaum 42:48 I think it goes back to what we said a bunch it's like, what are you trying to do? Right? So Rob gave the example of, you have some texting campaign, and then they got to your site, but they didn't make a payment. They logged in, they got to, like, the final step, but they didn't click, hey, like that could trigger something in real time. So, you know, you try and hit that consumer with a, you know, with a different piece of communication, hey, you left something in your cart. Come back and finish. You know, that kind of thing, once you get a little more sophisticated, can be really powerful to have real time data. If you're trying to analyze a campaign in aggregate, the whole campaign, then maybe you don't need to look at it in real time. You can let the campaign kind of run and then evaluate it afterwards and do a post mortem. So as with a lot of the things we've talked about, it really kind of depends what you're trying to do.

Jeremy Jobson 43:43 Do you think it's important that you do it in the right order? Though? I think sometimes people try to jump in and say, Hey, I'm going to start doing all this analysis, and I'm going to get real time and put and do all of it all at once. And I think they end up failing over and over because they're trying to do too much. I think the first step is you need to have the post mortem analysis down like that's just mandatory. You're going to need it whether you're making real time decisions or not. And then you move into real time decision, which I agree is incredibly powerful, but it's not powerful alone if you don't have the post mortem analysis up and running.

Amruta Joshi 44:20 Yeah, and I think I would agree with both of you. I think the only thing I would add to this would be a planning right, like, when you're looking at, like, your initiative, these are the type of things that you should already know, right, you know, okay, what does a postmortem analysis mean? Right? What are the metrics that you're going to be truly be looking at and have a plan in place, like, okay, am I going to be looking at it 30 days from the time of like, you know, when I started the new strategy, am I going to be looking at for like, 90 days to 120 I think there needs to be, like, a plan in place that kind of says, Okay, what am I looking at? What am I going to be analyzing, right? And then, What? What? What is that analysis? What? What are the expected outcomes? Like, what am I expecting or my out? Them to be and then, am I, like, you know, the meeting that? Am I not meeting that? And then make the adjustment, whether it's we need to go to, like, real time strategies to Aaron and rocks point, or whether we need to just take, you know, make changes and or do nothing, because it's working, right? So I think it's a planning is important piece of this too. Like, when you think about initiatives, it's like going back to basics, your strategy, your goals, and then what's your plan, and how you plan on executing and measuring that initiative, right? I think it's a key to aspect of this.

Raag Khandelwal 45:35 I guess planning is important, not just short term, but long term as well, right? Like Jeremy pointed out, it's important to start with baby steps and then build your way to the real time pieces. But then it's also important that when you start out, you make sure to invest in the right set of tools and technology that enable you to do that real time stuff down the road, right? So for example, if you're onboarding a portal vendor, right, who handles the payment portal? You need to ask them whether they do support giving you back portal events in real time, right? And then, once you get a vendor that enables you to do that, you start small, and then you maybe start doing it with a 24 hour lag, and then eventually drive your way to doing it real time. But then you need to plan in advance for it as well. So

Mike Gibb 46:24 so if the first step is to sort of figure out what the goal is, what the objective is, what are some of the questions that either you're asking today to kind of help you figure out what you're going you know what type of data you're looking for and how to analyze it, like, what are the questions you're asking, or what are the questions that you know someone just getting started on this journey should be asking? So to help them frame you know what they should be doing from a from a data aggregation and an analysis standpoint,

Amruta Joshi 47:02 I Oh, I can jump in. I can give you a little background. So I think one of the first questions will be like, you know, what at high level, is your goal to improve an NCL rate, right? Like, is it NCL rate? Like, how? What is your forecasted NCL rate? Like, target NCL rate versus how are you going to planning on accomplishing that right now, if you're talking about initiative base, you could be think about, okay, collections is about improving the collections rate. Are we going to be looking at, right? Does only improving collection rate? Are we trying to reduce losses? Are we trying to reduce, like, the bucket three, to charge off rate, right? Like, at very high level, there will be goals that and collections department by itself is trying to accomplish, right? So I think basically getting a better understanding of that, then aligning your initiative to, okay, so what is my initiative driving from, like, a high level goal perspective, right? And I'll give you a very simple example of that, right? So say my initiative is to improve collections using a new digital channel, right? Okay? And what does that mean that's going to help me collect more money through a digital channel. What does that drive? That drives an improved in the payment rate? But the equally important for me is I'm not I'm preventing also these customers from rolling forward, right? So I think there are different aspects of it that then leads to the questions of, okay, what type do we have, right communication from a customer standpoint, right? Do we have are our communications today reaching those goals? If they're not reaching those goals, then do I need a broader set of new communications, right, that I need to draft and take that goes back to Rock's point of like, test versus control strategy, right? Do I need to now build into the test versus control strategy? But I think where you need to every collections department is going to have a high level code, right? Your strategic initiative needs to be tied to the high level collections departmental goals that you're trying to accomplish. And then you then start asking, how am I going to do it? Right? How am I going to improve the payment rate? Okay, you want to improve the payment rate by doing having like, payment links embedded within your communication as an example, right? And then easy access for customers to maybe an easy way for to call consumers to make a payment without authenticating it as an example, right? So different ways of looking at it, but I think those are the type of questions, right, that you kind of start asking as you better understand how your initiative is tied to the higher level departmental goal.

Mike Gibb 49:51 If you were going to point to a small but what you think, or what could be deter, you know, could be an impactful change that a company can make today. A this month to start getting more value out of its existing data. What is that small change that you would advise or recommend or point to point to

Aaron Appelbaum 50:12 big one for me with any data driven initiative is data governance and data quality, and so I'm surprised we've gotten this far in a in a discussion about data without somebody saying, garbage in, garbage out, but, but I think that's a really important one. So understanding kind of where the quality is of your data, and can you even utilize it in an efficient way to drive your initiatives. I know that's a big one for us is, you know, the quality of data that we've received from clients isn't always the best, and so how do we, you know, prevent bad data from getting into the system or rectify it once it's already there? I think that's, that's a really big one. So making that part of your, strategic roadmap of understanding that Where does your data sit, and where is it at, and how do we make it better that will allow you or unlock the ability to use that data for whatever initiatives you come up with?

Jeremy Jobson 51:17 I would agree. I think for me, one of the big things that that I'm pretty passionate about is test into everything that you try, like, build your data in a way that you can have your testing controls and measure it and monitor it and compare. Don't assume that, hey, I'm going to do this digital strategy, and it's going to improve by X percent. So I'm just going to go, you know, all all in with it, like test into it prove it out. Like having that kind of rigor in your data structure and your methodology will will help you move smarter.

Amruta Joshi 51:52 And one thing we didn't talk about is AB testing, right? So test versus control comes into the play when you talk about collection strategy, right? AB testing comes into the play when you're talking about customer web experience, and how do they react to that web experience better, right? So that's another aspect of it that people could be, it should be, really be in your DNA, is to also think about AB testing like, you know, what do people respond to better? Right? When they log in, like, somebody asked a question, why an app? But once you're able to drive a customer to the app, whether you know, what, do they respond to better, right? What? What makes customer to take a positive action? So I think AB testing is another aspect of it that I think needs to be in a DNA along with test and control strategies.

Mike Gibb 52:42 Um, I'm curious now, having talked about this for almost an hour. And, you know, are there situations where data can't help an agency or a collection operation and someone's just has to make, you know, essentially make a gut decision. I

Amruta Joshi 53:06 Yeah. Well, you always hope, right, the data will lead to the outcome that you want, but that may not be the case, right? You may think that you have a right strategy, plan and you collect the data. So I think some of that goes down to like, you know, people's own experiences using a different channel. And, you know, I haven't seen that work. I think sometimes comes into the play. You always lead with the data, but and even the sophisticated analysis may throw, you know, surprises at you, right sometimes, and then you just stay so I think there is a piece of like experience that plays into this as well, that sometimes comes into the play along with the data, if data really doesn't give you the results. Mike, in my opinion, right.

Aaron Appelbaum 53:50 Going back to what Jeremy had just said about test out your hypotheses, I think when it comes to guess determining, what are the things that you're going to test out or try out or change. Those things can sometimes come from gut feelings or gut decisions, but you have to then use the data to validate whether or not it worked. So you may not always know, like, Hey, what should we do next? Sometimes that can come from experience or a gut feeling, but you should use the data that you collect along the way to validate that decision.

Jeremy Jobson 54:28 Absolutely. I mean, there's been plenty of times where we've proposed a new strategy, and maybe there's a point of, I don't know, is it? Is it SMS better at two, DBT or three DBD? You know, I don't know until we test it. Like I can't answer that question. I can tell you, you know, what my past experiences were, and I can tell you what my gut tells me, but I can't. I don't have data on this portfolio that tells me one way or the other right now, and that's where you do have to trust those instincts. Yeah.

Raag Khandelwal 55:01 I mean, if you don't have the data, then you've got to go out and create the data. Yes, so yeah. And then you basically test out different things. And then maybe what your gut says. You do that on a larger portion of inventory, but then you always have a hold out group, and then smaller portions of elementary where you're testing out the stuff that goes against, and then you keep monitoring the performance over time. So it's raining out here and the lights went off, but yeah,

Mike Gibb 55:35 well, good. All right, we are just about at the top of the hour. We've got just a couple minutes left, I'll ask each of you for to give the opportunity to make one last closing comment, and I'll give you three options, or you can just say anything you want. The options I'll give you are if you were going to assign homework to everybody who's watching this, if there's something they absolutely needed to do. As soon as this webinar is over, what is that homework that you would assign? If there's a most important takeaway that you want people to remember, you can share that. Or if there's something you didn't get a chance to say, just because the conversation didn't move in that direction during the during our webinar today, but there was something else on this topic that you wanted to share, you can do that. Or if you just have a if you just want to summarize, you can do that as well. So we'll start. We'll start with Jeremy.

Jeremy Jobson 56:31 I, you know, the biggest takeaway for me and this conversation, and what I hope people have, is that you know, anytime you're looking at data, you know, set your your target, your your figure out your KPI that you want to address prior to just starting to grab as much data as you can. And so I think that's that's probably one of the biggest things, is go after that set, set agreement with your business partners, and then you're able to move forward and building out your data

Mike Gibb 57:00 set. Excellent. Thank you very much. Raj, yeah,

Raag Khandelwal 57:03 I think I would like to go back to Amrita point on planning. I think that's one of the bigger takeaways. For me is don't think about what you want to do with your data today, but think about what technology is going to impact your business and what data is it going to generate, let's say two, three years from now, and invest in a CRM investment vendors. Invest in tech that helps you build a futuristic agent.

Mike Gibb 57:28 Love it. Thank you very much. Aaron,

Aaron Appelbaum 57:31 I like the point about testing. I think that's a good one to take away, whether it's control testing or AB testing, but also post mortem testing or validation, measuring how things performed and really looking at that after you've done something to inform whether or not what you tried worked and what you should do next.

Mike Gibb 57:58 Thank you. Amruta.

Amruta Joshi 58:00 For me, I think the important takeaway is data for no data for sake of data, right data for driving insights, understanding the value of the data, and then making the tweaks to your strategy based on the data, whether you do that based on your test and control, whether you do that based on your AB testing. But I think no data for sake of collecting the data, it's not building reports, it's building insights.

Mike Gibb 58:26 Excellent. Thank you. Well, I'm Ruda. Aaron Jeremy Raj. I want to thank each of you so much for being here today on behalf of everybody who's watching this live, or watching or recording this late, or recording of this later, and walking away with some much better under a much better understanding of how they should be approaching their data and what they should be doing with it. Thank you to each of you for being here. Really appreciate it. Raj, I want to thank you and everybody else at prodigal for sponsoring today's webinar. Please check them out. Prodigal tech.com is the website to those of you watching and listening. Thank you so much for being here and thanks for making us part of your day. We appreciate it if you're watching us live, we're back tomorrow on the webinar at one o'clock Eastern where we're talking about will versus skill, separating those who can from those who won't. If you haven't signed up yet, we've got a link in the chat or there you can find a link on the events calendar at account recovery.net got a great panel coming for a really interesting conversation. So if you haven't signed up yet. I recommend you do so, and we'd love to see you back for that tomorrow at one o'clock Eastern. Until then, enjoy the rest of your day, everybody. Thank you so much for being here, and we look forward to seeing everybody soon. Thanks guys.

Summary Getting Your Data Set Up So You Can Make the Most Out of It

Mike Gibb hosted a webinar on setting up data for effective collection operations, sponsored by Prodigal. Experts Aaron Applebaum, Jeremy Jobson, Amruta Joshi, and Raj Khandelwal discussed the importance of data quality, governance, and strategic goals. They emphasized understanding customer behavior, segmentation, and the need for unified data. Key metrics include liquidation rates, cost to collect, and time to collect. They highlighted the significance of real-time data analysis, post-mortem evaluations, and AB testing. The panel also stressed the importance of compliance, engagement data, and the role of AI in improving data accuracy and efficiency.

Action Items

[ ] Prioritize data governance and data quality as a foundational step.

[ ] Establish your high-level departmental goals and tie your data initiatives to those.

[ ] Focus on building data-driven insights, not just collecting data for the sake of it.

[ ] Implement a test-and-learn approach, using control groups and A/B testing.

[ ] Plan for future data and technology needs, not just current requirements.

Outline

Introduction and Overview of the Webinar

Mike Gibb introduces himself and the purpose of the webinar, emphasizing the importance of setting up data correctly to maximize its value.

The webinar is sponsored by Prodigal, a company specializing in AI-powered and agentic workforce solutions for collections and servicing.

Mike introduces the panel of experts: Aaron Applebaum, Jeremy Jobson, Amruta Joshi, and Raj Khandelwal, highlighting their backgrounds and roles.

Mike encourages live participants to ask questions and submit comments anonymously if preferred.

Identifying Valuable Data Assets

Mike asks Aaron Applebaum to discuss the most valuable data assets for a company in the credit collection industry.

Aaron emphasizes the importance of understanding the organization's goals and the type of data needed to achieve those goals.

Aaron suggests focusing on consumer demographics, debt specifics, and interaction history to inform business decisions.

Jeremy Jobson agrees, stressing the importance of understanding the portfolio and customer behavior to set a baseline for further data collection.

Strategic Goals and Data Segmentation

Amruta Joshi discusses the importance of defining strategic goals before collecting data, suggesting a step-by-step approach: defining goals, segmenting the population, and determining treatments.

Amruta highlights the need to track performance metrics like NCL rates, liquidation rates, and pay rates.

Aaron Appelbaum adds that data collection should also include KPIs to measure the effectiveness of strategies and execution.

Raag Khandelwal emphasizes the importance of compliance and effectiveness metrics, such as liquidation rates, cost to collect, and time to collect.

North Star Metrics and Operational Improvements

Mike introduces the concept of a "North Star" metric, using Uber's example of rides per day.

The panel discusses the importance of defining specific North Star metrics for collection operations, such as liquidation rate, cost to collect, and time to collect.

Amruta Joshi explains the difference between daily data analysis and the concept of collections per day, emphasizing the need for trend analysis over time.

Raag Khandelwal and Aaron Appelbaum discuss the evolving nature of North Star metrics based on business goals and the importance of setting boundaries for these metrics.

Data Quality and Compliance

Mike asks about the quality and quantity of data being collected and analyzed today compared to the past.

Amruta Joshi and Jeremy Jobson discuss the shift towards digital and AI-driven data collection, emphasizing the importance of data quality and compliance.

Raag Khandelwal highlights the improvement in data quality due to AI-driven note-taking and disposition codes.

The panel agrees on the need for real-time data ingestion and analysis to drive operational improvements.

Customer Journey and Data Unification

Mike asks about the importance of understanding the customer journey and the data points that should be monitored.

Aaron Appelbaum emphasizes the need to identify the milestones that drive consumer behavior and inform decision-making.

Amruta Joshi discusses the importance of a unified view of customer data to understand their interactions across different channels.

Jeremy Jobson and Raag Khandelwal highlight the need for a single unique identifier to connect data from various sources and ensure accurate analysis.

Setting Up Data Analysis Processes

Mike asks about the process of setting up data analysis and the importance of real-time decision-making.

Jeremy Jobson emphasizes the need for post-mortem analysis before moving to real-time decision-making.

Amruta Joshi discusses the importance of planning and setting clear goals and metrics for data analysis.

Raag Khandelwal advises investing in technology that supports real-time data analysis and planning for future data needs.

Key Takeaways and Closing Remarks

Amruta Joshi stresses the importance of using data to drive insights and making strategic adjustments based on analysis.

Jeremy Jobson emphasizes the importance of setting clear goals and KPIs before collecting data.

Raag Khandelwal advises planning for future data needs and investing in technology that supports real-time data analysis.

Aaron Appelbaum highlights the importance of testing and validating data-driven strategies.

Getting Your Data Set Up So You Can Make the Most Out of It (From 6/10/2025)

Daily Digest – June 17. Complaint Accuses Defendant of Not Going Far Enough to Investigate ID Theft Dispute; Nevada Governor Vetoes Medical Debt Collection Bill

23 Companies Seeking Collection Talent

mikegibb

23 Companies Seeking Collection Talent

Leave a ReplyCancel reply

Upcoming Events

News

Videos

Informational

Welcome Back!

Create New Account!

Retrieve your password

Are you sure want to unlock this post?

Are you sure want to cancel subscription?

Getting Your Data Set Up So You Can Make the Most Out of It (From 6/10/2025)

Support authors and subscribe to content

Subscribe

Buy Article

Related

Daily Digest – June 17. Complaint Accuses Defendant of Not Going Far Enough to Investigate ID Theft Dispute; Nevada Governor Vetoes Medical Debt Collection Bill

23 Companies Seeking Collection Talent

mikegibb

23 Companies Seeking Collection Talent

Leave a ReplyCancel reply

Upcoming Events

News

Videos

Informational

Welcome Back!

Create New Account!

Retrieve your password

Are you sure want to unlock this post?

Are you sure want to cancel subscription?

This recording is available for Premium Members.