Steve Groccia on Keys to Effective Cohort Analysis

[00:00:00] Steve Groccia: If you want to trust your historical behavior for a customer, you need to trust your historical data, otherwise, you know, it’s probably not gonna help you, right?

[00:00:09] So it doesn’t mean you can’t do cohorting, but it means you may have to start fresh. Gathering the data in the right structure, in the right format and start feeling confident with it, to then maybe reap the benefits two, three months later, when you actually start getting, you know, that data in a place where you feel good about cohorting it. 

Steve Groccia Introduction

[00:00:24] Joe Michalowski: Hello and welcome to another episode of The Role Forward podcast. My name is Joe Michalowski, and this episode is brought to you by Mosaic, a strategic finance platform that transforms the way business gets done. And today, we’ve got a bit of a different episode. We talked to a lot of external

[00:01:00] finance leaders, and today we’re talking to someone from Mosaic. We’re chatting with Steve Groccia, the Head of Customer Success here. But he is a former Head of Finance himself before he made the shift to CS. So Steve, thanks so much for joining us.

[00:01:12] Steve Groccia: Thanks for having me, Joe. 

[00:01:13] Joe Michalowski: Awesome. Yeah, this is gonna be great. Steve, I’ve gotten the privilege of chatting with Steve about finance topics in the past.

[00:01:19] I’m really excited about this episode. Before we kind of dive into our main topic, do you wanna just give a little background on yourself? What were you doing 

[00:01:26] before Mosaic? How you made the shift to CS? 

[00:01:29] Steve Groccia: For sure. For sure. Yeah. So started my career in audit at Deloitte in, in Connecticut. Started the accounting route from college and spent a few years in audit doing banking, financial services type clients. My wife and I kind of got to get out of the Northeast,

[00:01:45] and so we moved to San Francisco where I joined the M&A practice, worked a lot with SaaS companies, worked a lot with private equity firms to buying and selling companies really on the, the financial diligence side. Learned a ton there. And then being in San Francisco, I think everyone gets the itch to join a startup.

[00:02:02] And so I tried to leverage a lot of my learnings from finance and was the first in finance at a FinTech company called Fundbox. Joined there when they just raised series B first in finance. And spent four and a half years there, really building out the function, seeing the growth, learning a ton about just the business from the inside, and then got approached by the co-founders at Mosaic.

[00:02:23] And saw it as a, as an interesting opportunity to kind of take what I’ve learned across my career. You know, both, both on the professional services side but trusted advisor, if you will. And then it happened finance and leverage kind of both of those together in a customer-facing role to really help solve,

[00:02:38] you know, a lot of the problems that I faced, and I know my peers face on a day-to-day basis. So it was a, a unique opportunity, but definitely, I’ve been with Mosaic now for two years. Definitely has been an awesome ride. I’ve, you know, learned a, a ton. But yeah,

[00:02:50] Steve Groccia: a little bit of a unique route to get to where I am. 

[00:02:54] Joe Michalowski: Yeah, no doubt. I think that’s, you know, I say it a lot either on this podcast, I say it a lot internally. As a content person, I’m super fortunate to have people like you who come in and, in all the roles we have in Mosaic, really just a ton of finance roles that are no longer like sitting in a finance position at this company.

[00:03:11] And I think it makes everybody better at, or trying build. And so, when we’re doing this podcast episode, try to come up with a topic that is relevant to whoever’s coming on. And so your dual experience in CS and finance makes for a good episode about a really critical topic for everybody which is retention.

[00:03:28] And you and I have talked about it in the past. But specifically gonna try to dig into cohort analysis, mostly because we did this research report at the beginning of 2022, really about like benchmarking, not just SaaS metrics, but like the process of getting those metrics. And one of the questions we asked was, “Do you use cohort analysis for

[00:03:46] things like net revenue retention or gross revenue retention?” It was like 14% of finance leaders responded saying that they use cohort analysis to look at retention, which felt really low to me as somebody just like outside looking in. So I’m curious why, you know, as somebody with this dual experience, why do you think that number is that low?

[00:04:06] Is that surprising to you or, or no? I don’t think it’s surprising. I mean, I guess if I think about it, what are the challenges, I think sometimes just the fundamentals of having time, right, having resources to do what I would call like these deeper analyses. Right? And, you know, I remember my first couple years at my last company just doing everything, right?

Why Teams Avoid Cohort Analysis

[00:04:23] Steve Groccia: You’re doing everything from closing the books to payroll to anything that touches a number you’re kind of responsible for. And you’re the only one responsible for it, right? And so there’s always these big burning fires, and you never really have the time to do this deeper level analysis.

[00:04:37] And I think, I bet if you asked that group how many want to do the cohort, I bet that number’s really high. Right? But you know, time becomes a challenge, and even just like getting data in a structured format to be able to do it, like, even if you had the time, it takes a lot of time to just set up the data in a way that would actually make it usable and make it work.

[00:04:57] We had to leverage, like our BI team to build structured data and then get all that data into like a Tableau, BI tool, and then I had to learn the BI tool and then figure out what I’m like, ” What does the data look like?” And then figure out, “Well, what do I actually want to do,

[00:05:10] Steve Groccia: and how do I want to interpret this data?” So all that takes time, all that takes a lot of resources and energy, which is a challenge. And then, I’d say for some companies, it may not be beneficial until you get to a certain scale. Right? And so when you’re especially when you’re thinking about cohorting, you think about like the number of customers you have, the number of transactions you have, that’s where it becomes super valuable.

[00:05:29] If you have these 

[00:05:30] Joe Michalowski: Hmm.really large contracts, bespoke contracts, things like that, it’s hard to actually bucket customers into segments and actually leverage what the power of cohorting is. 

[00:05:41] Steve Groccia: At Fundbox, it was…we’re dealing with in the SMB space. So 

[00:05:44] Joe Michalowski: Hmm.

[00:05:44] Steve Groccia: tons and tons of homogenous customers, you know, we had no choice, but we had to cohort. If you’re trying to get into the weeds of how, what’s this customer doing versus this customer, you’re never gonna actually find insights. 

[00:05:55] So I mean, in, in a word, cohort analysis huge pain. I think that’s honestly kind of the crux of a lot of what we’re doing here with our content with Mosaic is that finance has a lot of like highly manual tasks that just take a lot of time. And so, not surprising to me based on all the conversations we’ve had is that it just comes down to that time.

[00:06:14] Joe Michalowski: And I liked what you said about how many of those 86% that said they don’t do it. How many of them want to be able to do cohort analysis. So I really liked that. I, I think that kind of leads to my next question is, when cohort analysis is necessary, when isn’t it. So can you just talk a little bit about the difference between like a, a basic churn analysis, maybe like the sort of basics of looking at your retention versus what you’re talking about, where you’re going deeper with cohort analysis? Is there anything you miss out on, by not going the deeper level or basic churn analysis fine,

[00:06:47] in many cases?

Churn Analysis vs. Cohort Analysis

[00:06:48] Steve Groccia: Yeah, a good question. I think the basic churn analysis and, you know, I kind of think about it like, you know, you’re looking at your churn over time, but you’re, you’re doing it at, at a super, super high level. Right? So that’s usually your starting point. And it’s a number that you need to know inside and out from a finance perspective. It’s a number I need to know inside and out now at Mosaic from a CS perspective because it’s the number I’m responsible for.

[00:07:09] And it’s obviously a number your investors care about if you’re in a SaaS business. That 120% net dollar retention is like that quote-unquote magic number. And so that’s your starting point, right? And so it’s table stakes just to know where do you land, relative to other companies of your size and what have you, but then to really understand the dynamics of it. Like that’s where the cohorting becomes very powerful because you could actually start slicing and dicing this data in so many different ways to really

[00:07:36] understand why. Right? That becomes very powerful from a finance perspective and, you know, sometimes you need the structure, you need the tools to be able to do it, but understanding the why is really how you could find the insights and really drive business change which, you know, that is the gold standard for anyone in finance, right?

[00:07:52] Is you do all this work, you wanna share it with the rest of the organization, and you want it to drive more revenue. Right? You wanna become revenue-generating in finance if you can. So, that’s where it’d be helpful. 

[00:08:02] So I hear you on, you know, net revenue retention, kind of hitting that 120, but if you’re talking about the high level, like when you’re going a level deeper, like, what are some of the most common use cases where you decide like, “Okay, we’re gonna go a level deeper.

[00:08:12] Joe Michalowski: We’re gonna do cohort analysis.” What are the most impactful ways that you can use that, become that net revenue generator 

[00:08:17] in the company? 

[00:08:19] Steve Groccia: Yeah.

[00:08:19] I think there’s a few different ways. I mean, I could probably talk cohorts for hours upon hours. You know, I kind of think about two different flavors of cohorting. You know, let’s call it like segment-based cohorting and time-based cohorting. And so, segment-based, we had one investor when we were going through a debt financing at Fundbox that really wanted what they called a layer cake chart.

[00:08:38] Right? And so, okay, that sounds cool. No idea what it means. So we started to look into it, and now,now I love layer cake charts. And it’s essentially, you know, you have your dates across without getting too technical, the dates across your X-axis and think about like, you wanna look at your ARR over time.

[00:08:54] Right? And so, if business is going well, then ARR over time is going up into the right really fast. But then actually segmenting that ARR by your different cohorts, and you know, each layer of that ARR stack is a different, let’s call it, when you signed on customers 2017, 2018, 2019, 2020, and onward.

[00:09:12] And so it actually, it looks like a colorful layer cake that you would, you know, you would have for dessert. And actually, it’s very insightful on, when you dive into it, it’s insightful because it tells a lot about how much of your revenue is from existing customers, how old your customers are, right?

[00:09:28] Which from 

[00:09:29] a SaaS perspective, like the stickier you can make the product, the better for just the long-term viability of the business. And you did learn a lot by like segmenting your different data by cohorts. And then, you know, you’re trying to find that balance with, you want a lot of layers because that means you have a ton of existing customers using your product, but you still want that top layer to be really high because you wanna keep growing, and keep growing fast.

[00:09:51] Right? And so, so that’s one way. Does that make sense? 

[00:09:54] When I explain that to people like it’s hard without a visualization, but wanna make sure that makes sense. 

[00:09:59] Joe Michalowski: No, absolutely. I can, I’m picking like an area chart where it’s just like, uh, multicolored. I don’t know how many segments you’d have kind of up into the right, but I can picture it, and I don’t work in finance. So I’m guessing somebody else that does who’s listening will be just fine kind of visualizing it. And we can throw in some, I’ll put something like the show notes with like a graphic bringing still something from Mosaic and 

[00:10:19] maybe give people a visual. For sure, for sure. And it’s funny, I actually thought about like the layer cake, and threw on a timer, and tried to see how long it would take me to build a layer cake in Mosaic. Right? And this is 

[00:10:29] Yeah. 

[00:10:29] Steve Groccia: free Mosaic plug, I guess. But it took me 40 seconds to build a layer cake in the way that I wanted to see it. God. 

[00:10:34] That means I’m probably losing a step in the app. I’m sure my CSMs could do it in, in 25 seconds. But… 

[00:10:39] Joe Michalowski: Yep. 

[00:10:39] Steve Groccia: Yeah.

[00:10:39] It was cool to kind of see it, the visuals with the colors and, what have you. Yeah, that, that’s a super, super powerful way. I think, other way too, we spend a lot more time cohorting at a deeper level is,

[00:10:50] what we call like time-based. 

[00:10:51] And that’s really to understand like your behavior, like a behavior of something over time. 

[00:10:56] Joe Michalowski: Sure. 

[00:10:57] Steve Groccia: And so, you know, it’s the classic. You, could you look this up, you’ll see like the classic waterfalls where along the top, it’s 0 1, 2, 3, 4, 5, 6, and then, along like the Y-axis is some sort of segmentation that you’re doing and net dollar retention over time, lifetime value over time.

[00:11:12] And so it’s really meant as a way to, to analyze some sort of behavior to find insights and then see what behavior works, what does not work. 

[00:11:22] Joe Michalowski: Gotcha. Well, I mean, so when I’ve spoken to you about cohort now in the past, it’s usually in the context of retention because you’re the CS guy, and like that, that’s what we talk about. And so it sounds like you can apply it to all kinds of things. Like I know, lifetime value is retention metric but doesn’t necessarily have to be, you know, how many customers are retaining period over period. And so, if you can apply cohort analysis to so many different metrics with like the 

[00:11:47] time-based approach, 

[00:11:49] is there any scenario where it’s kind of overkill? Like where you’re syncing so much time into it when really like, you’re not gonna get that much value, or there times where you shouldn’t use cohort analysis? 

[00:12:00] Steve Groccia: That’s a good, it’s a good question. I mean, I’m a, I’m very pro cohort, so I like to at least have the option to investigate the data with that view on it. I think where you could get in trouble is like trying to find insights specifically in a cohort, and if it doesn’t exist, so you’re almost spinning your wheels trying to find something that’s just not there.

[00:12:20] So you have to find the balance between, all right… Taking two steps back and, “What am I trying to solve here?” And, you know, sometimes cohort is not the answer, but I mean, there’s still many, you know, we of cohort everything at my last company. And even now, you know, I think about,

[00:12:33] Steve Groccia: you know, how we leverage cohorts, how I, you know, as a CS leader, leverage cohorts day to day. I mean, I could paint a picture for you. Like I mentioned before, if my team is responsible for net dollar retention, right, I want to know, I wanna understand my customer behavior around when renewals happen, you know, when upsells happen, what retention looks like, but then being able to slice that behavior

[00:12:54] by different segments. Right? So I wanna see maybe how each CSM is doing and the behaviors of their customers to maybe the insights on CSMs performing better or worse than others. Right? Or we have different size customers, right? Or different industry customers, customers using different products.

[00:13:11] That’s when becomes, you know, really powerful is getting that time-based cohort, but also layering the segment-based cohort, to really understand and slice and dice your data in so many different ways to find insights. So that’s more on like the finance side, so that’s kind of still 

[00:13:24] Joe Michalowski: like 

[00:13:25] Steve Groccia: Yeah, that finance guy in me wanting to do that as well as, know, numbers that, me and my team, are ultimately responsible for. 

[00:13:30] Joe Michalowski: Sure. 

[00:13:31] Steve Groccia: Then it’s also like the product aspect 

[00:13:33] of it,

[00:13:33] right? And that actually is very unique when you wanna understand specific customer behavior, right, to maybe be proactive about when usage is going down. Right? At what point do our customers, like at what point is Mosaic sticky for our customers, what we could actually look at weeks since the customer started onboarding, and what their usage looks like.

[00:13:53] Right? And then, you have that ability to see these customers are getting the most value out 

[00:13:59] of a product, 

[00:14:00] and this is that inflection point when it happened. And, okay.

[00:14:02] Well, if we could get a customer to bring on X amount of people to collaborate with, and X amount of users and have, you know, Y amount of, of time on the platform over, over first four weeks for six weeks, that 

[00:14:13] Joe Michalowski: is 

[00:14:13] Yeah. 

[00:14:14] Steve Groccia: a step to success.

[00:14:14] Right? And 

[00:14:15] Joe Michalowski: Yeah. 

[00:14:15] Steve Groccia: it becomes very powerful when, you know, and all you’re manipulating really is like that timeframe, and sometimes like, just like what the underlying data is that you care about. 

[00:14:24] Gotcha. Really interesting to hear you kind of take it out of the finance perspective. So, obviously, when I come up with these topics, we’re a finance audience, it’s really finance-focused. And this one, in particular, you start looking up articles about it, there’s a real mix of like heavy numbers

[00:14:38] what are our financials look like and heavy, like product manager audience kind of like trying to figure out when, as you said, when the product is sticky and when it kind of loses that stickiness. Again, like really interesting to hear your perspective since you kind of straddle the line now between the two sides. I want to get really nuts and bolts for people. I try to make things as actionable as possible when we do these topics. And I’m wondering if you’d be willing to like, take a step back to your Fundbox days and walk me through like a step by step of cohort analysis? Like, I, like it’s we finally closed the books,

[00:15:12] I wanna look back at our numbers, some of the use cases you mentioned, and what step one of, “I’m about to do cohort analyses, what do I need to do?” And I know Mosaic plug, like, “It’ll be easier in Mosaic. We got it.” But, anyone who doesn’t use the platform, like, what does that look like for them when they have to do go end to end? 

[00:15:29] Steve Groccia: Yeah. Yeah.

How Software Makes Cohort Analysis Easier

[00:15:30] That’s a good question. Let me go back into to my memory bank, and I tried to put some of this out of, my memory when I joined Mosaic, but, yeah.

[00:15:37] Happy to, happy to, to give some insight here. And, even some, some additional context, like cohorting was actually like that was the underlying methodology for us to plan the business. And so, without getting into too much information on, in detail on, on what Fundbox does, they essentially automated underwriting to give small businesses loans. And so there’s a com, there’s a usage component,

[00:16:02] there’s kind of a recurring component of, “I’m giving you, Joe, a, a line of credit of 25 K to go use in your small business. You could draw on that line of credit and payback.” Right? And so a lot of complexity around, “Well, how often are you withdrawing? How big is your credit line?

[00:16:17] How often are you paying back? What does that look like over time? Are you drawing more and more over time or less and less? Are we giving you more credit increases? Are you stopping to pay?” And so, there’s a lot to really digest and take in. And so the complexity there was, you know, even before diving into the numbers, we’d have to cohort everything. We have 6, 7, 8 different types of cohorts. So that, that hopefully paints a picture of how wide this was. And then the actual fundamental of doing it, I mean, know, you’re taking, I’ll do the super manual way first, and then we actually upgraded to have like a BI tool and hired a team of four people to structure our data.

[00:16:48] Right? But before that, it was taking raw data, I have it in a massive spreadsheet, looking at the key fields, like the key date fields that, that I care about. Right? And so, if you’re comparing cohorts over time, what is that starting and ending point? Usually, the starting point is like, when does a customer first become a customer.

[00:17:05] And then time is kind of relative to every time they, they drew out a new loan. And so, I’m adding in additional columns and rows to my Excel sheet. I’m doing all these formulas. I’m making sure everything works appropriately. And that’s like, step one to like take some sort of data and kind of, of restructure it,

[00:17:22] so it’s usable. And then step two is, “All right, now I’m gonna have another sheet. Now I actually have to set up what the cohort looks like and how it works, just to get the data into a more cohort-like form.” And then step three is depending on if you want to take an average if you want to take a percentage of the initial balance, if you want to take a running sum, I mean, that stuff’s not easy to do even with like pivot tables and power pivots. So then replicating all of what I just did then doing additional for formulas on top of that.

[00:17:47] Right? And so, I mean, it’s taking me a couple minutes just to explain it. You could imagine when you’re dealing with large data sets, how frustrating it’ll get. 

[00:17:53] Excel is just not big enough sometimes, so we’re dealing with big data sets, and oh, we realized something was wrong.

[00:17:59] We pulled in the wrong, we didn’t pull in enough of the data. So now I gotta go redo all of that. And that was just to kind of get your starting point. And then the second point was like, “All right, well, what’s actually going on with these cohorts? It’s a new month, it’s a new quarter.

[00:18:11] What’s changed?” And like our assumptions are driven off of these like master cohorts. And so we think like we’re bucketing half of our potential new users into what we think is gonna be their behavior over time. And then, you know, now we need to dive in to be like, “Well, is that behavior curve,

[00:18:27] if you will? Is that the most accurate curve?” And you know, now if you’re tweaking the curve a little bit, the numbers could change dramatically. Right? And so it gets a little scary when you’re forecasting, but it’s also very insightful of, “Well, hey, marketing team. Hey, retention team. This is what we’re seeing.

[00:18:41] If we could, could move the needle a little bit on behavior, it’s actually gonna have a massive impact on our customer base.” Right? And so, that’s where things become, I mean, that’s like the fun part of finance, right, is like, you kind of find some of that and 

[00:18:52] present it back to the rest of the organization to find actionable insights.

[00:18:56] But yeah, that stuff takes time. Right? And, I, I wish I had more time to do what I described on a monthly basis and update that data monthly, and get folks in the loop. The reality is it’s just painful, is kind of like, all right, every quarter we’re gonna refract it, make changes and what have you, but, yeah. 

[00:19:12] Joe Michalowski: Is that the ideal cadence would be monthly? So now say you put something like Mosaic in place like you can look at it in real-time, is it still… I’d imagine you don’t want to get bogged down just constantly looking at this data every day. Like you could just drive yourself crazy, probably looking at minute changes.

[00:19:27] Is it just like we close the books and every month we’re gonna look at those, see where we’re at, and adjust, or is there a different cadence that makes more sense? 

[00:19:34] Steve Groccia: So I think it depends. You know, I think the monthly cadence is like the natural best cadence for finance, right? Because you want the books to close. You want everything to come in. You’re usually doing everything on a monthly basis. Sometimes if you don’t have time, it’s probably quarterly when you’re doing your refracting.

[00:19:48] But other times, you actually may want to look at it, maybe not daily, maybe that’s overkill, but maybe weekly. Right? I mean, in, at Fundbox, customers would repay weekly. And so we would actually leverage cohorts as a way to get the initial insights on, “Well, oh, this customer actually just, you know, missed a payment or this cohort is trending pretty bad what’s going on?”

[00:20:07] Right? And so, we were able to get more proactive by looking at it on a weekly basis. I mean, probably something I’d want to do on the CS side too of, you know, weekly logins and weekly usage across our customers to be as proactive as possible in terms of getting in front of potential issues. 

[00:20:23] Joe Michalowski: Gotcha. No, it made, all makes total sense. It’s kind of the main answer to a lot of the questions I ask on this podcast. Is like, “Should you do X?” And it’s like, “Well, it depends.” And like, “Here’s like the three different scenarios, where it does or does not make sense.” And it’s always nice to hear guests, people who have done this before for so long, dig into to what those nuances are. That’s why, honestly, it’s the only reason why we have a whole podcast for us ’cause if everything was so easy, we wouldn’t need to talk to anyone about it. 

[00:20:48] I want to get into some of your CS experience, and now that you’re on that side, is there anything you’re hearing from customers about cohort analysis that is surprising to you?

Common Use Cases for Cohorting

[00:20:59] Maybe use cases you’re finding matter the most to people that we are selling to, you know, we, we sell mostly to VC backed SaaS companies like in growth stage. Are there use cases that you’re surprised to hear are most important to people? Yeah, I mean, I don’t know if I’m necessarily surprised, but there’s definitely use cases that are more important to the customers we work with. A lot of hyper-growth company is, right, they’ve raised a ton of money and now looking to really deploy that cash. Right? And so,

[00:21:29] Steve Groccia: where a lot of them deploying cash is, to really, build out a sales function, hire a sales team to bring on more revenue. Right? And so, a big thing that comes up is sales rep ramp, right? Especially 

[00:21:40] Joe Michalowski: Yep. 

[00:21:40] Steve Groccia: in a context of planning and the context of,… “Okay. I wanna understand, I wanna understand like how fast a new sales reps can start producing revenue and see what that looks like over time.” 

[00:21:52] Joe Michalowski: Hmm. 

[00:21:53] Steve Groccia: So I know if I hire X amount of sales reps now, here’s what the output may be over the next 3, 6, 9, 12 months. Right? And so it, you can’t expect to hire 10 sales reps tomorrow and get full output tomorrow, right? Like there’s a whole process. You have to learn the product, you have to follow the playbooks and start building your pipeline.

[00:22:13] Right? And so it’s really important for customers, especially when it comes to planning. And, if you work backwards, I wanna double revenue at the end of the year. “We have five salespeople, we know we can’t do it with just five salespeople. When do I have to start hiring salespeople?”

[00:22:26] Right? And based on the production, we’re expecting them to get overtime, well, now I’m just doing math to see, “All right. What’s the optimal time for me to hire a salesperson?” Right? And I mean, you could take that a step further. And historically, how long has it taken me from posting a sales job to hiring a salesperson, how long does that take? So now I could almost like cohort upon cohort to say, “All right, well, you know, I need to have five new salespeople on the phone, you know, in June. All right. I gotta talk to my people team because we have to start thinking about hiring now.” Right?

[00:22:56] And so, that’s a just a couple ways that cohorts could work in the context of growth. I think even from like a sales pipeline perspective, if you know, our marketing team is doing an awesome job of generating new opportunities, well, how long, long does it take those opportunities from when like a deal becomes a potential opportunity? How long does it take to actually close?

[00:23:15] Steve Groccia: Right? And so how many now working backwards, how many opportunities do I think I need now because I want to close these deals in three months, six months? Is that, does that make sense? 

[00:23:24] Joe Michalowski: Yeah, it absolutely does. I think, to me, you know, it’s, again, outside looking into this job, it just seems like my first instinct would be to cohort everything. You mentioned cohorts on cohorts, we just have like cohort inception going on. It’s just, it, its sounds like you can get really tricky to manage and obviously, just to roll it back to where we started with the report, and it was like 14% of people are actually doing this. And that was just for one metric. Like if it’s 14% of the people can only manage to do it for like net revenue retention, how many are really doing it to the extent that you had to do it at Fundbox or that our customers, if you’re, this is what they want to do?

[00:23:57] Like, it’s just really hard. Like, it’s just, it’s like you want to do it and just don’t have the time. And so, I know we talked a little bit about challenges, but just to wrap up the idea of you being more customer-facing now. Are there challenges you’ve heard from customers outside of the ones that we’ve mentioned that maybe are on like a deeper level?

[00:24:14] Like, is there like a, I dunno one that stands out of someone who came in? It’s like I have to do X, Y, and Z, and it’s impossible. Like, can Mosaic help? 

[00:24:22] Steve Groccia: Yeah, I think there’s probably two. I think the first, the bigger challenge is the cohorting generally places, a, a heavy reliance on historical data, right? Especially if you want to trust your historical behavior for a customer, you need to trust your historical data otherwise, you know, it’s probably not gonna help you, right?

[00:24:41] So it doesn’t mean you can’t do cohorting, but it means you may have to start fresh gathering the data in the right structure, in the right format and start feeling confident with it, to then maybe reap the benefits two, three months later, when you actually start getting, you know, that data in a place where you feel good about cohorting it.

[00:24:57] So a big one. And, you know, even if you get past that point, I think sometimes even where I struggled in the past because I loved using cohorts is how do you actually visualize it in a way where you could explain it to someone who has no idea what a cohort is, what they’re looking at and why it’s important.

[00:25:13] Right? And so, when I think of cohorts, I think about it in these massive data tables, but you’re not gonna share a data table with a hundred lines in all these random numbers because folks are not gonna like understand that or digest it. So even things our customers look for now is,

[00:25:27] sometimes they do wanna show all the numbers, but having things like heat maps, right? Where you could see, “Okay. Just focus on the green things because that’s good. Focus on the red ones, that’s bad. Now we’ll dive into what those numbers mean.” Right? Or putting it in chart form, so you could kind of see what the curve looks like over time. 

[00:25:43] Joe Michalowski: Yep. 

[00:25:44] Steve Groccia: Pretty evident, you still have to explain what the chart is, but it’s evident to where things break. 

[00:25:47] Or 

[00:25:48] Joe Michalowski: Yeah. 

[00:25:48] Steve Groccia: where there’s an inflection point Hugely important theme for Mosaic for all the content we do, for everything we talk about here, we’ve mentioned it on multiple podcast episodes with people that aren’t even from here. So this idea of like finance as a customer service organization, like for the business, and it’s funny, ’cause like, you are the customer success lead for this company,

[00:26:09] Joe Michalowski: so you epitomize that idea, that’s really what it comes down to when you’re like, you know, you need to visualize this in a way that people can actually understand it. I’m wondering if now that you’ve been on the CS side for as long as you have, is there any like newfound appreciation for CS challenges that may be like you didn’t realize on the finance side? If you were on like that side looking into CS when you had to collaborate and you just like couldn’t understand why you were up against a certain barrier, maybe some advice finance leader interacting with CS when their CS leader does not have a finance background? 

[00:26:46] Steve Groccia: Yeah. I think when you’re in finance, you really work for the rest of the organization, right? If you’re effective in finance, everyone at the org wants to work with you, and they want to get your insights, and they trust you with the numbers, and they come with you to a problem, they come to you with a problem, and they want you to do your magic data and manipulation and calculations to help give them answers. 

[00:27:08] Joe Michalowski: Yep. 

[00:27:08] Steve Groccia: Over CS, it’s almost like the rest of the organization works for you. 

[00:27:12] Joe Michalowski: Hm. 

[00:27:12] Steve Groccia: You’re working directly customer, so the customer feedback loop and product and engineering and support when there’s bugs and working with sales to make sure that we have a nice, smooth handoff. 

[00:27:23] And so, working with the marketing team, working with the sales team around and new content and material, so it’s definitely flipped on the CS side. And, you know, I I definitely have more of an appreciation for that. And, you know, there’s somebody who started at, at Mosaic and, you know, they asked me what, what I did at the organization, and I said customer success,

[00:27:40] and they were in the engineering, or they say, “Well, I’m in customer success too.” Right? Which is like very resonated, very well. It’s like everyone at the organization is, ultimately like doing this for the success of our customers. Right? And so, you know, our team is kind of a conduit to our customers, but we’re kind of relying on everyone in the organization to really make it happen. 

[00:27:59] Joe Michalowski: Yeah, I love that. I’m gonna have to ask you offline who the engineer was, ’cause I’m gonna have to chat with him or her as well. So, yeah, it’s a really unique perspective. And it’s true. I mean that, that is what we’re all driving toward. I know we’ve been going for a little bit. I want, I want to of get to our most common question

[00:28:13] we ask it for everybody that comes on. And it’s zooming out of everything. No, no cohorts, no nitty-gritty what you do on a daily basis, but would love to know something that you know now that maybe you wish you knew at the start of your finance and now CS career. Anything at all for advice for people.Yeah.

Finance Career Advice from Steve Groccia

[00:28:31] Steve Groccia: It’s a good question. I think one thing is assume you know nothing, right? And so what that means is, you know, I came from audit where you just, you have to ask a lot of questions. You have to do so much more listening than talking. 

[00:28:44] Assuming thing and

[00:28:46] sometimes, asking the basic questions actually become very powerful to fundamentally understand the concepts. And then you could actually dig in and learn more and ask better questions. So, really trying to simplify things, processes, numbers, behave, you know, a lot of the stuff we talked about. Try to put it in its simplest form and ask them basic questions to get a lot of context.

[00:29:08] Joe Michalowski: Hm. 

[00:29:09] Steve Groccia: And just to make sure you understand the fundamentals of it, is gonna go such a long way into learning as much as you can about it. And actually just slowly becoming smarter in that space. 

[00:29:21] Joe Michalowski: Yeah, that really resonates with me. I, I love this question mostly because it always ends up applying to much more than just finance. So you ask finance leader for advice, like what they wish they knew. And it’s never like, “I wish I knew more about building cohort tables, like in Excel.” It’s always something that zooms out a little bit for, like the longevity of their career. You know, for me, like just doing this podcast, just getting to learn from people like you and the other guests on the podcast. Just coming in and being curious about what you all do on a daily basis. So, I couldn’t agree more with the idea of just assuming you know nothing and going in and trying to learn as much as you can.

[00:29:57] I think it’s super important. I think everyone will appreciate that. 

[00:30:00] Steve Groccia: Totally. Totally. And that’s one of the cool things about being in CS now is Hm. 

[00:30:03] we’re talking to customers every day, different size customers, different personas. Like one day talking to a co-founder or CFO, next day talking to 

[00:30:13] VFP&. So, you know, you just, just asking questions and, you know, I always tell my team like context, context, context and just start getting all these dots and trying to connect them because it’s, it’s gonna make it easier for you to help the customers, ultimately what they care about. Right? 

[00:30:27] Joe Michalowski: Yeah.

[00:30:27] Steve Groccia: And so, that’s one of the coolest things is half of our job is just asking questions and listening. 

[00:30:32] Joe Michalowski: Yeah. And then trying to figure out where we can help based on what we hear. 

[00:30:35] I love it. It’s a great way to, to kind of wrap up here, ’cause I know we’re, we’re at time. Normally I give people a spotlight to say like what company they’re from and what they do, but like, go to mosaic.tech and learn about what we do. But, I’ll open it up to you, just where can people go to connect with you, learn more about all the work are doing here, maybe reach out about working with you as a potential customer? 

[00:30:55] Steve Groccia: Yeah. Yeah, totally. I am at LinkedIn, for sure. You know, if you have any connection to Mosaic, it’s pretty easy to get ahold of me, I think. 

[00:31:02] Joe Michalowski: Yep. 

[00:31:02] Steve Groccia: ‘Cause of 

[00:31:02] Steve Groccia: the, the running joke. I think most of, of my customers, probably have my phone number. So you could probably get that if you, you dig hard enough. 

[00:31:07] Joe Michalowski: Yeah, let’s not throw that out on the podcast. We don’t need to be giving that out. 

[00:31:10] Steve Groccia: For sure. For sure. But yeah, I’m, I’m accessible, for sure. 

[00:31:13] Joe Michalowski: Cool. We’ll see. Thanks so much for joining. Really great to learn from you, hear more about your background. And all, all the insight about cohort analysis was really great. So just appreciate you spending time with us, and, uh, it’s great having you on The Role Forward. 

[00:31:24] Steve Groccia: Yeah. Awesome. Well, thanks for having me, Joe. 

[00:31:26] Joe Michalowski: All right. See you soon. 

[00:31:26] Steve Groccia: See you.