Name: CFO Collaborative: The explosion of data and how to navigate it
Uploaded: 2021-12-23
Duration: 19 min 36 s
Description: Senior Editor Lee Clifford discusses with CFO's across the business world their ideas and best practices.

Senior Editor Lee Clifford discusses with CFO's across the business world their ideas and best practices.

CFO Collaborative: The explosion of data and how to navigate it

Leadership

The explosion of data and how to navigate it

we're going to dive into one of the very specific challenges facing Cfos today. The explosion of data and how to navigate it? Uh, we're joined by our discussion leaders, Dylan CFO, Alan Parker and AMC CFO, Christina Spade, who will kick us off with their thoughts, but we're eager to hear from all of you. So as Alan said, please try men in the chat and um we will get to as many of you as possible. Hi, Allan and Kristina. Hi, good afternoon everybody. Thanks for joining us. Um I wanted to start out, you guys are both in businesses where obviously data is absolutely essential to engaging customers. Talk a little bit about what you guys can do now that wouldn't have been possible a decade ago, Christina, you want to kick us off? Sure. So the number one thing being an entertainment media company with AMC Networks is we'd love for data to be able to predict all the hit programming that's going to be out there before we invest in it. But unfortunately we have not cracked the code for that yet, but I keep trying. Um but 10 years ago we didn't really have a lot of direct detail on consumer viewership information. Certainly we had Nielsen information other formal metrics, but now in the world of streaming, um we actually learned quite a bit more about our viewership and what people watch and uh, so it's opened up a whole new important world to us that we really are benefiting from. You guys are getting pretty good at predicting what I might like based on what I've watched in the past. Alan, what about what about you guys at Zillow? Yeah. Thanks. So you know when Zillow was founded, it was really about empowering the consumer, our customers uh, and democratizing the information available on the housing industry um for the benefit of those customers. I mean this is a big their largest financial transaction they'll make. Um, and so it was, it was really opening up information to the customer. And and the best example of that was obviously the testament where we actually put, and we have now a market value on 100 million rooftops across the country. And that market value is based on publicly available data, customer submitted data along with market macro trends and location specific. But this gave the customer or consumer uh, the opportunity to to go and shop and dream and look at their own property and and get a market value. Um, if you think about what we can do now, you know what I really say is that the scale of data that we have access to is just grown exponentially. Um we just recently joined uh, multiple listing services. So mls to those who have bought or sold the house recently. Uh, and so we now get feeds from IVF which is the interchange that provides that data. Um and that allows us versus using some third parties to get that you can imagine if there's a house that's for sale, having that information real time Um can be the difference between one of our customers being able to make an offer on that house or seeing it a day late and missing it in the current market. Um but another big area, you know, like I said on scale is we have 2.5 billion visitors a quarter now. Um And so with the advent of machine learning and technology, we can take what those customers do, what they look at, where they live and are they looking at other cities? There's all this information that we're now able to use that enhances that ability to value that house or value that location or to improve the customer experience when they're on our site uh to where we can help them get to their next next best action based on what they, we know about them, what they're looking at, where they live now and and where they're perusing. So, you know, it's this ability to scale and iterate very quickly with machine learning. Um I think the next stage is gonna be computer vision as more and more three D home tours populate our site where we're actually able to use computer vision to even be more accurate about the value of these various houses. And you know, we we announced that we're actually, you know, taking the estimate to the ultimate, it's not quite, I'll call it an ultimate right now level where and for about 1.6 million homes and 23 of our markets where we do Zillow offers where we actually buy houses from our customer sellers. Uh, we make our estimate the initial offer, uh, just the start of one click. So, so it's, it's really this fast iteration and scalability. And I'll say, you know, with the advent of the iphone and other media, we just get a lot more data points which allows us to be a lot smarter. That's amazing. And um, Alan Murray has a good question in the chat. He wondered how much of what you've just described as in the purview of the CFO? Well, yeah, uh, well I am very, very interested and very engaged in all of our algorithms. Um, you know, we're making decisions on this testament as an initial offer to actually purchase a property that we have to, to resell. Um, and to be honest, you know, the, the top of funnel metrics that I see where customers are coming and engaging and shopping and dreaming is super important to the long term sustainability for our business as we move to actually helping our customers get through the transaction. So it's super important that we have top of funnel that we understand the leaky bucket. And so I'm constantly working with our data analytics and machine learning teams to understand how quickly we're iterating, um, at what confidence interval are we willing to accept making estimate is a one click offer versus having someone intervene at least on the initial offer. Um it's it's something, you know, again, I I spent 13 years at amazon. I'd say prime videos also working on trying to figure out what to invest in and what not to make money Christina, but um you know, so I'm very familiar with and really believe that there are a lot of challenges and a lot of problems we face in the business that are knowable with data and our job is to figure out how to make that knowable that there's there's always going to be a distribution curve. And so then you also figure out where on that distribution curve you're willing to take the risk, but I'm very engaged. Uh and it's actually very interesting to me, especially as we get into the computer vision part. Well, Helen, I will say that I am very engaged with my estimate. I'm not obsessed with it, but it was second all the time. So um Christina, I'd love to go back to you, we all do, we all do Christina. I'd love to hear a little bit more about forecasting. Obviously that's one major way to use data, but this year it seems like everything going back to year is no longer relevant and we don't know what's coming. So how do you do the job of forecasting? I think it's a great question and it's a very relevant question for where we all are now in the finance world, in the sense that, you know, I always like to say whenever you're done with the budget, when you put the pencil down, it's already wrong because so much has changed. So to me the budget give you a set of guidelines to work with that you've thought through strategically, but it's not something that you follow, like a bible. I think the power of the forecasting really comes in when a collaborative group across the company is able to access how you're doing real time against the budget and then what new measures you're seeing that you didn't anticipate in the budget and how to adjust accordingly. I think as a CFO it's really important that you look at your job as really optimizing and not leaving any revenue on the table and making the smartest investments you can with the information that you have to drive the business. And for AMC networks, we actually have been through a bit of a transformation from last year into this year um aided I think by the acceleration of viewership in the streaming platforms, but we've launched in a significant way, targeted streaming platforms and c plus, we launched in october 2020 and we're seeing a lot of success with it and it's really where If I would have looked a year back, you can't really forecast and say we're going to be at X in 12 months, you can have goals and a vision for it, but it's all about the execution. So relative to forecasting, it's important to be nimble, but it's more about the communication of what you're seeing against the forecast. The other thing I can say is when you think about Covid and what it did to our world in the financial space, it wasn't really about how am I going to do against the budget? Because this time last year your budget was very different. It didn't predict Covid. So then it was really all about balance sheet strength, liquidity and then also supply chain, you know, and and with the Suez Canal and now with what happened with the gas prices etcetera, you know the supply chain concern goes even into past Covid, I think it's gonna be something that's going to continue to be important. Absolutely. Well getting back to the kind of top line theme here, how do you both um keep from being overwhelmed by the amount of data that's coming at you, How do you edit out what's noise and what's really important to your business. Alan do you want to tackle that? Yeah, yeah. And I'll just kind of play off, you know what Christina said? Um you know, for me it's it's and it it was important during the pandemic. I think it's important at any time but um we focus a lot on the input. So I think you need to have a good strategic plan, a long term plan where you're looking far enough down the horizon to make sure you're working on what you're working what you need to be working on to be successful 24 36 months, 48 months from now and those are your strategic initiatives that you're building, they're they're a little tougher to forecast accurately and then you have some more of your stable uh you know, more mature businesses that may have less variability in them that that helped fund some of those investments in bets. I think in all cases on those, you want to measure quickly, you want to iterate fast and you either want to lean in where things are working or you want to fail fast for those things that you tried and you aren't as successful or didn't give you the results, you thought so, you know, I think we, we heard about being curious, you know, learn, you shouldn't be afraid to fail. I think that's important with respect to my focus on how I kind of weed out the noise, you know, I think it's important to identify for the CFO and for the business leaders to identify what are the key inputs, you know, so when things get a little hazy and during the pandemic, you know, it was very difficult to see very far ahead, even in today's world, we have a macro environment in the industry that uh you know, has a lot of tailwind to it um but also has very low inventory listings and um understanding the inputs and tracking those on a frequency that's daily or weekly uh and making sure you pick the right inputs so that if those inputs come to a result you you would hope for your outputs will be as expected, you know, that allows you to react a lot quicker than waiting for the monthly close review or waiting for the next forecast. So I'm a big input driver um and and these dashboards, you know, gets back to the use of data and you know what my team does, we have a list of data that we review regularly. We have control limits for most of that data where possible when we get outside the control limits. That would be where I interact with the business leaders to inspect and understand what's going on and what we could do about it quickly and for those controls that are are those inputs that are you know at trends are performing better than trends. I don't have to spend as much time, so that's kind of how I read out the noise. I try to identify the areas, I need to be spending time to help us adjust and fix and and single source of truth data that allows you to have great dashboards and to iterate quickly and escalate fast. Uh You know, I think it's very important in my role for helping the business when you're dynamic and you have so much, you know, interesting stuff going on like we did, that must be some dashboard. Um, I'd like to know we've got a couple other voices here. Um, do we have Tony Bowen from H and R block on the call? Um, thinking even though people got an extra month to file their taxes, I bet you are awash in data right now. So tell us a little bit about how you guys are managing. Yeah. And if you haven't filed yet, six days to go to get your block office, so hurry in now. I mean, I think everything everybody said today is exactly on point. You know, we're obviously a legacy brand and we have a lot of data that we've been building up for a number of years and you know, I was curious to see the answer to your, your survey question because I answered the legacy data architecture is the number one problem that we face. I'm jealous if Allen has the dashboard that he was talking about because for us it's definitely a struggle and you know, getting it to a data architecture that's streamlined and seamless that you can quickly then analyze is paramount for us. We've been going through a journey the last few years and moving all of our data to the cloud and that's been a big journey to one clean up our architecture and then to reduce our storage fees, make us more nimble and improve security. I mean there's just a lot of pieces tied to that. Once we're through that phase, then it's, how do you narrow down on the key KPI s to run the business? And that seems so simple. But it's so difficult because when you have so many data elements trying to narrow in on the few that really matter. Um, not only the ones that help you understand where you've been and the performance from, you know, last week or last month, but more ideally where you're going and for us because we have so many different data fields. Think of everything that's on the tax return and then all the business metrics around that, trying to figure out what are the ones that are truly indicative to understand where we're going and that we all agree are the key indicators while also still keeping some visibility to things that aren't the key indicators but would tell you there's a problem. Um, and you know, someone explained it to me once, a little bit like a dash dash in your car where you've got the MPH and the rpm and the, and the gas tank meter. But then there's a bunch of other things that are being monitored behind the scenes and when one of those are out of line and, you know, the check engine light pops on and that sounds so simple in a car and we all can relate to it, but trying to do that in business I think was what, what we're trying to do, but it's, you know, to be completely honest, it's much easier said than done, not just at Tony that and the driver needs to be looking a lot farther down the road than just in front of his hood and that's a strategic element that you have to have in order to get where you want to go uh safely. I'd like to try and leave in a couple more voices. Here is Davina from waste management on the call. I am. Hello lee Hi, how are you to tell us a little bit about how this all looks from your point of view? Well, it's interesting that Allen's most recent comments were about looking over a hood because that's what we do every day. And um talking about the magnitude of data and, and enterprise, I can tell you at waste management. It starts with the fact that we have on board computer units on each and every one of our vehicles read conversations with enterprises who are interested in using our data to learn more because as you can imagine whether it's a waste management or UPS or Fedex, we're driving down the streets of our communities each and every day and there's value to the information that can be gathered in that regard. What we really struggle with, I would say is with a decentralized organ organization and one that's hungry to use data in a purposeful way is a lot of people using that data in what I would just consider a rogue fashion. And so um you've got all of a sudden people taking data analytics classes and and trying to create their own data warehouses across the enterprise. And what we're trying to be is much more purposeful structured um in terms of feeding that data rather than using valuable enterprise resources in a less purposeful, less structured way than is good for the business. Um that requires a great deal of collaboration because while finance has data analysts within its organization so does our HR organization as does the digital organization and and I. T. Does not report to me as CFO though I am a tremendous partner to our chief digital Officer. So for us it's really about driving collaboration so that we're not walking on top of each other stepping on toes creating data in vacuums and instead being more purposeful and intentional with the data that we are accumulating and how that gets digested. That seems like a very sensible approach. And finally in the last minute I'd love to is present from analog devices on. Hi Yes I am high, we'd love to just quickly here from you in 60 seconds or less. What I think that that perhaps what is what hasn't come out of this conversation is the role of whether it be your FP and a function or some analytics team in modifying the the the scoreboards or the dashboard you're using to maintain that constructive tension that finance needs to drive when you pick on a couple of metrics and then you start flashing them then the organization begins to respond to drive to those metrics but there's always unintended consequences which starts showing up so then knowing how to once, once you've solved one set of problems figuring out. Okay now I need to pick a different metric because I want to highlight a different area or a different opportunity for improvement and that is the part that is hard for us to put into an ai mlps and that's where having business context and relying on analytics on an analytics team or in our case the FP and a team to be able to bring that judgment becomes becomes very important. Excellent. You did it. You managed to do it in 60 seconds. Amazing. Um thank you guys all so much. Alan Christina, thanks for joining us and everybody I'm dan um This is actually a perfect transition into our working group sessions in just a moment. We're going to send everyone into four different breakout rooms. Each of these rooms will be hosted by a fortune editor together with the CFO discussion leader and just a reminder that this part of the program will be off the record so feel free to just let it fly and as we shared with you before this gathering the prompt that we're all going to be talking about is the following. How can CFOs collaborate more effectively with I. T. And other teams within their companies to better navigate the explosion of data. So please get involved engaged and we'll see you all back here for our report backs at 4 15 Eastern.