Episode Highlights:

Ibrahim Albaba: “BI, in many ways, is the glue that holds together a lot of different departments. Sometimes we exist in silos, and the data brings us together towards a main vision. You want to connect the marketing people to the finance people, the marketing people to the revenue cycle, the IT operations people, the actual care operations, and the physicians. The way you’re gonna do that is through business intelligence because marketing isn’t just a marketing problem… There are all these factors and facets—you wanna get everybody on the same page behind one thing. BI is the department that’s gonna bring that all together.”
Episode Overview
In this special episode of Ignite, we’re bringing you an exclusive session from Scaling Up: The Healthcare Performance Marketing Summit 2024. Cardinal’s CGO, Lauren Leone, and CSO, Rich Briddock, sit down with four healthcare marketing leaders to discuss how data-driven decision-making is revolutionizing the future of healthcare marketing and how leveraging better business intelligence (BI) can significantly enhance marketing strategies and outcomes.
A standout takeaway is the value of integrating data from diverse sources, such as call centers and scheduling systems, to boost conversion rates and patient engagement. For instance, introducing a scheduling calendar early in the patient interaction process can increase completion rates by an impressive 50%. This underscores the need for marketers to evaluate every touchpoint in the patient journey and its influence on key metrics like ad spend and customer acquisition costs.
The conversation also highlights the importance of defining and tracking meaningful KPIs tailored to healthcare organizations, such as return on marketing investment (ROMI), return on ad spend (ROAS), cost per acquisition, and brand health. Focusing on these metrics allows marketers to better allocate resources, optimize media spending, and run more impactful campaigns.
Segmentation also takes center stage, with the panel discussing the use of tools like Experian and Mosaic modeling to identify and target specific patient groups. This refined approach enhances campaign relevance, increasing returns while controlling marketing costs by ensuring the right message reaches the right audience.
This episode will equip you with a deeper understanding of how to leverage data for smarter decision-making, improve patient acquisition strategies, and effectively communicate marketing value to organizational stakeholders.
Featuring:
Andrew Chang, Chief Marketing Officer of UChicago Medicine
Tom Stanton, Director, Marketing Analytics at Northwell Health
Clay Nickens, Vice President, Strategy & Growth at Action Behavior Centers
Ibrahim Albaba, Director of Growth & Strategy at CLS Health
Announcer: Welcome to the Ignite Podcast, the only healthcare marketing podcast that digs into the digital strategies and tactics that help you accelerate growth. Each week, Cardinal’s experts explore innovative ways to build your digital presence and attract more patients. Buckle up for another episode of Ignite.
Alex Membrillo: I know you’ve been waiting around for six days hoping another episode of Ignite would drop in. Here it is my friends. This is going to be a phenomenal 45 minutes. I hope you’re going to enjoy it. We are only rehashing two episodes from scaling up, and this is one of them. This one’s all-around BI, Business Intelligence, driving insights from data. We felt this was so important because 2025, the macroeconomic outlook is not super rosy like it has been. Driving more efficiency and profit out of every dollar spent on advertising is critical. Over the next 45 minutes, you’re going to hear from Andrew Chang, the head of marketing, the CMO of UChicago, Clay Nickens, the head of growth and strategy at Action Behavior, the largest ABA group in the country, Ibrahim Albaba, the head of marketing at CLS Health and a guy that Ashley has a big crush on. Tom Stanton, Northwell Health’s head of analytics. They’re going to be diving into everything from creating the right dashboard to the right integration, understanding the business metrics that matter after talking with COOs and CEOs and CMOs, aligning on all that for a shared understanding and shared goal set, and talking about how you do big integrations with technology and teams over hundreds and hundreds of employees.
Finally, we also do cut into what are the metrics that matter for driving growth. At the end of the day, what do these four smart gentlemen wake up and look at when they’re eating their Wheaties? We are going to hear about all of that over the next 45 minutes. Let me stop talking so you can start learning. Have a blessed day.
Lauren Leone: Hey, everybody. Welcome to our second panel discussion on day one of scaling up. I have the pleasure of having Rich here with me. You guys probably all know Rich from our podcast, some of our events. Rich is our chief strategy officer. Rich and I have been doing this together for 10 years. It’s great to have him as a co-host. Last year I hosted all my panels solo and this year I get to have a friend.
Rich Briddock: Thanks for inviting me. I really appreciate being here.
Lauren: Who better to bring from the Cardinal team maybe other than Alex Kemp, to our data to decisions, transforming healthcare marketing with better BI panel than Rich to be our co-host? I’m super excited to introduce you all to our guests today. I’m going to call them out one by one. We’re going to let them tell you a little bit about themselves and their organization and then we’re going to get into our panel discussions. First up on our list of panels for today, we have Andrew Chang of UChicago Medicine. Andrew, hey there. Can you hear us?
Andrew Chang: Hey, thanks for having me.
Lauren: Awesome. Thanks for joining us. Andrew, tell everybody listening a little bit about yourself and just about your organization, what you guys have going on, your scale and scope.
Andrew Chang: Happy to be here again. Thank you for including me and inviting me here. I’m in charge of marketing and PR and external communications at the University of Chicago Medicine. I’ve been here for just over one year now. We’re right now in the build phase of our new marketing operating system where we’re building up a new tech stack analytics and attribution modeling, what have you. It’s been a lot of fun, a pretty big challenge, but having a great time.
Lauren: Awesome. We’re going to spend some more time talking about your build process here in just a minute. Next up, let me introduce you all to Tom Stanton of Northwell Health. Let’s see if we got Tom on here. Hey, Tom, can you tell everybody listening a little bit about yourself and Northwell?
Tom Stanton: Sure, Lauren. Thanks. Hey, everybody. I’m Tom. I’m the director of marketing analytics at Northwell Health. We’re a very large health system based out of New York. We’re 20 hospitals, over 80,000 employees, pretty big footprint. Where we are is within what we call an internal agency. We have a centralized marketing department. We work with people inside the department and across all the different areas of New York City and Long Island primarily.
As far as what we have going on right now, we’ve got a few data migrations going on. We’re moving our EHR over to Epic. We’re ramping up with Salesforce as a CRM and moving a lot of our cloud databases as well. Certainly, a lot going on. One other interesting thing just about our health system is, a big part of our marketing that’s been growing out is within the entertainment space and doing content. Partnering with streaming services to create original documentary content. A lot of cool–
Lauren: The content on Netflix, 10 out of 10 recommend. We can give those some shout-outs. We spoke this morning about how marketing wears many hats and noticing that every single one of you all have slightly different titles. Next up, we’ve got Clay Nickens, vice president of strategy and growth at Action Behavior Centers. Hey, Clay.
Clay Nickens: Hey, everybody. Really excited to be here today. Action Behavior is the nation’s largest provider of applied behavior analysis for children who are on the autism spectrum. My responsibilities include, as Lauren was mentioning, growth and strategy. Business development, new service line development, strategic partners, and marketing all within that umbrella. Within marketing, our organization serves not just potential new patients and families but also supports our position liaison and recruiting teams.
In terms of where we are as a marketing organization, we’ve been super focused on bottom-of-funnel people who are really ready to start services. Our journey over the next year or two is to have a more fulsome relationship with potential patients and using data to do that is part of our day-to-day. Again, excited to be here.
Lauren: Speaking to Rich and I’s hearts here on the V.I. from Clay. Last up, we’ll bring Ibrahim Alibaba into our panel of CLS Health, and then we’ll kick off the discussion. Ibrahim, can you tell everyone a little bit about yourself and CLS?
Ibrahim Alibaba: Hey guys, my name’s Ibrahim. I’m the director of growth and strategy at CLS Health. CLS Health is Houston’s largest physician-owned, physician-led healthcare organization. That’s central to our belief and our core value as an organization. My responsibilities like Clay’s very similar, but I like to say that my job is to grow CLS both in revenue and value, and anything that touches that, I touch it. That includes business intelligence, V.I., business development, mergers and acquisitions, and then finally the marketing team. V.I. is really central to everything that we do as a company, and that’s where I actually started and then eventually grew into a larger role. Excited to be on this panel [crosstalk]
Lauren: Something really cool about all of you, you all have, whether it’s passion or by trade, a background in V.I. and analytics. Then you’ve found the best way to leverage that in marketing growth and strategy at your organizations. Rich, I’m going to let you kick off with our first question, which we want to put out to each of you just to understand, and you all touched on it a little bit already, your current state of your business intelligence function, but want to dive a little bit deeper into what you guys are working towards. Rich, I’ll let you put out the question to our first panelist and then we’ll go around the circle.
Rich: I feel like this is the trick for every analytics professional and data professional. Which is what is the promised land and how do you get there and how do you prioritize that workload? [unintelligible 00:07:05] deploying media mix modeling, multi-touch attribution, there’s so much that you could do and to leverage data to make your organization more effective. We’d love to know from you guys what your focus is and what your priorities are for 2024 and 2025 to make your organizations more effective in terms of the way that they use data and what the biggest challenges are that you face to achieving those things. In no particular order, I guess maybe I’ll just pick on someone, but Tom, why don’t you kick us off if you may?
Tom Stanton: All right, sure. Thanks, Rich. You mentioned tools and techniques. Where our focus is going to be is a little bit more on process and relationships within our department. I think what we want to do is do a better job of speaking the same language as our business partners and making sure that we are very aligned on our thinking so that it’s not request and response. It’s more of a collaborative strategic conversation where we make it more apparent what is possible and they make it more apparent what the problem is that they want to solve.
Getting BI more directly tied to decision-making is the main thing and how we measure the value and the actions that we’re taking more directly and thinking about future decisions, not past decisions and what happened. That’s the main direction that we want to go. That’s going to take form and obviously reaching out and setting up some more opportunities to collaborate, but that’s going to be getting really in the weeds of what people’s processes are and how they make these decisions now.
Rich: Yes. Makes total sense. What about Clay? How are you thinking about [crosstalk]
Clay Nickens: I’m just going to go and start with the jet stream here that you’ve established. The phrase that we use is democratization of data, but it’s very similar in terms of being– Like you were saying, you have to get the details right, whether that’s multi-touch attribution or just cleanliness of the day of a well-governed set of data sources. Then putting that in a position where users in the organization, decision-makers in the organization can access that and are really ready to serve in fast basis is our focus for the remainder of 2024 and probably early 2025.
Lauren: Clay, who are the stakeholders in that for you at Action?
Clay Nickens: We have a pretty tight-knit team here in Austin. We also have folks all over multi-site locations. In terms of making that democratized, ensuring that our C-suite is fully read in on what’s happening, even at a very granular level, allowing them to do that dig and dig and dig and deep dive, that’ll probably resonate with a lot of people who are listening in on this as executives always want to go deeper, but then being able to serve up the right level of information for somebody who’s in an individual center and trying to make a decision about how to work through the pipeline of prospective families.
Lauren: That’s really everybody. You’ve got this BI is for everybody, not just the person with the strategic analytics title at the organization.
Clay Nickens: Absolutely.
Lauren: Awesome. [crosstalk]
Clay Nickens: 2024
Lauren: Yes, this year. We’ll get into next year in a minute. Ibrahim, what about you? What are some of the challenges you guys are facing? What’s the end state look like for you?
Ibrahim Alibaba: To start with the second part of the question, the end state. I’ll talk about this from a data science perspective of me wearing the data science hat, but I think the genesis of any data and business intelligence team is this. Is first you want to know the past. How good are you at analyzing the past? Then after that, you really want to get to what’s happening day to day. Let’s get monitoring tools out there.
Then once you get really good at analyzing and the monitoring, the next step is the future and it’s predictive analysis. I think that’s generally the genesis. I think what happens sometimes is people try to jump ahead. What we want to do is really nail down those previous two things. I think prediction comes later. Now to segue to Clay’s point, what’s the end state? What’s the best ideal situation? From a business intelligence perspective, I think BI in many ways is this glue just like he said, but it’s glue that holds together a lot of different departments.
Sometimes we exist in silos and the data brings us together towards this main vision. I guess what I mean by that is if you really think about it, you want to connect the marketing people to the finance people and the marketing people to the revenue cycle, to the IT operations people, to the actual care operations, the physicians. The way you’re going to do that is through business intelligence because marketing is not just a marketing problem. Marketing is a finance problem and RCM is a marketing problem.
There are all of these factors and facets. You really want to just get everybody on the same page behind one thing, I think BI. Business Intelligence is going to be that department that’s going to be able to bring that all together. I think that’s the end state, bringing people together, analyzing, monitoring, and then maybe one day when you get really good at it, let’s start predicting stuff.
Lauren: Yes, absolutely. We’re going to talk a little bit about predictive competency modeling here in a minute. Andy, I’m curious and I’ll ask this back to the rest of the group too, you’ve got to get those first two steps really good and clean and solid. You’re going to need the right technology and the right team on whether they’re BI professionals by trade, external partners, you’ve got marketing individuals that you’re training up. What does an ideal infrastructure look like for you to be able to accomplish that? Do you have any tools or technologies that you’re leveraging and what might a good team look like? The right roles and responsibilities?
Andrew Chang: First of all, I’m into what you’re writing. We’re just saying very passionate speech. I’m very inspired. Thank you for that. Second is in terms of the roles and responsibilities and how we build the team, there’s a few things that I require every single person that has worked for me to go through, which is a course that I had to go through when I was at UPS and then eventually teach called Insights to Action. It’s basically how to interpret data and then turn that into insights, which is not just a regurgitation of the data. Then lastly is how do you turn that into actionable next steps and making sure that you’re not just telling a story with no end?
It doesn’t matter what department or what team you’re on. You may be a graphic designer, you still have to understand how data works. That’s number one. Number two is I’ve worked in places where BI has been centralized under finance or under IT, or it’s been a very federated model with– Everyone has their own analysts and data scientists. At the end of the day, the way that the model should work is what Clay was saying earlier is the data easily accessible from a democratization standpoint.
Second, is that data trustworthy? If I took results from one dashboard that’s built and showed it to the service line, the service line should say, “Yes, I saw that too.” Not, “Where did you get this from?” Same thing for finance. Making sure that data is trustworthy, it doesn’t mean it has to be 100% accurate. As long as everyone knows what the assumptions are and what could be a little off, that’s fine. That’s what should be happening, but there’s got to go be a trust level and layer on the data itself.
Then lastly is we try to make everyone on the team a business owner. The question that I’ll ask is if this was your own business, if you owned this business as Ashley’s knee replacements are us or whatever it is, what would you be looking for? Not just from a social media campaign standpoint, the whole experience. What are you looking for to make sure that your business is running in the best way possible? It makes you think beyond campaigns.
It really forces you to think from the customer standpoint as well as financials, everything else. Try to just have an ownership mindset, learning how to tell stories with data and making sure that everyone’s on the same level and the same page in terms of the data trust aspect as well.
Lauren: Clay, you talked about democratizing data. Any tips on how to best do that? Have you solved that from a platform perspective? Any advice on what organizations might want to put in place basic data warehouse? How can people access the data? How can you make it available for the democracy?
Clay Nickens: I don’t think we’ve solved it to start with. The second thing is that our organization, I have the advantage of a single service line today. A relatively standardized model. If you compare our organization to Andy [unintelligible 00:15:25] and Tom’s, but that permits us to do a more purist approach to this, which is single data warehouse, serve that up in a controlled format depending on the user’s sophistication. Our finance team is going to have better access, more access to the raw underlying data without assumptions or cleanup necessarily than say a user who’s in one of our centers who needs a more distilled and curated experience. No, I don’t think we’ve [unintelligible 00:15:52]
Speaker 1: It’s an evolution. I don’t know that either. No one’s ever going to probably just point to a time and say it was solved. It’s going to evolve. There’s going to be new data points and KPIs that you hadn’t thought of before that you want to build into it. You get great questions from the field that help you realize I need to make that data available too. I appreciate that mindset.
Clay Nickens: Maybe one thing to add here is this is enabled by getting the boring, small details right. What are your standardized center codes? What are the keys that you need to basically connect your various data sources and being realistic about what’s possible to interconnect and not? If you don’t have a clean setup of those keys and the crosswalk associated, which is unique immutable IDs is probably the key phrase most often, then you won’t be able to do any of the other cool stuff that you want to. It’s not fun, but you got to go through every single one of those and make sure they’re right in all of your systems. Then you have the feedback process to catch when one system may fall out of compliance with your standardization scheme.
Rich: Do you guys think that is the biggest hurdle in terms of building a great BI function inside of an organization is data cleanliness, data accessibility, or is it resourcing? Is it time? Is it ever changing goals and objectives of an organization, the different needs of stakeholders? What are the biggest challenges in terms of getting to that promised land of that perfect BI function?
Lauren: Tom, I’ll ask you first.
Tom Stanton: I’ll go for it because I can build on what Clay was saying. I think that they’re all inspirations, but for sure, a lot of the inefficiency that would come out of everybody’s BI function is going to be those things that Clay called out. It’s the lack of stable foundation to build on top of, it’s data quality, it’s defining things once. There’s a term in data engineering, it’s called dry, D-R-Y, Don’t Repeat Yourself. We want to define these things once in our data warehouse and never have to go back to it and rewrite the same SQL queries or pull the data again.
That’s absolutely where we are right now is knee deep in data quality stuff and modeling these different pieces of the business. One of the tools that we’re hoping to bring on soon is a semantic layer, which sits between the data warehouse, even the clean data sets, and the actual reporting tools. The idea with that is that you define metrics and what’s called entities, whatever these business objects are one time, and then it’s something that you can not have to go back and waste your analyst time of just data wrangling and cleaning it up later on. The idea is that it’s all ready to go and it’s just poke around until you are looking in the right place and then you can really accelerate what you can do from there. It is the boring overhead that has to go first.
Lauren: Ibrahim, I know you mentioned you’re in phases one and two as well. Does that resonate with you? You’re at that point. That’s the hurdle at the moment.
Ibrahim Alibaba: I guess we have a different challenge because like I said, we have a centralized VI team. I always think about what’s the centralized VI team’s challenges. Data accessibility in general is always going to be that one. I think what we’ve done to solve that like what Andrew said, is make them business owners. I think there’s an [unintelligible 00:19:13] I don’t know how to say this, but most of the time you’re going to maybe be wrong. There’s probably things you’re wrong about. The data might be wrong. You have to be humble enough to understand that and you have to be curious enough to find it.
I think by creating business owners, you empower them to challenge your thing. I’ll just give an example. I know I may not be answering the exact question, but it’s something to think about. A lot of people’s calculations are completely based on their own assumptions. The biggest drivers to improving our calculations to any sort of metric, believe it or not, has been physicians. That’s because we sit with physicians one-on-one and we explain to them exactly how we did it and then they ask us questions that we’ve never thought of.
They say, “Did you consider this and that or that and this.” The big one is how do you calculate customer acquisition costs? People ask, “Did you exclude X segment or Y segment?” I’m like, “Wow, that’s a really good point.” Instead of in those QBRs or whatever you’re doing, you’re just glossing over, “Hey, [unintelligible 00:20:09] is this, LTV is that, using that jargon. Really breaking it down and sitting with those really end stakeholders and being humble enough to accept the fact that you might be wrong. You’re [unintelligible 00:20:19] The classic thing in data science or engineering is junk in, junk out. The thing that’s coming in, is it junk, or am I somewhere in the middle? Am I missing some calculation steps? I think that’s my two cents on this subject.
Lauren: Ibrahim, I think that’s a perfect segue because I want to bring this discussion down into the marketing layer. A lot of our listeners today are marketing leaders at their organizations and they’re wondering, “I’ve got an individual perhaps at my organization, like Clay, like Tom, like Ibrahim, like Andrew, who’s owning this step of it. I’m the marketer sitting here trying to figure out how am I going to use this data? How am I going to ask the right questions? Improve data literacy at my organization.”
The key being to glean the insights to inform the next decision on where to put the next dollar, the next audience to find the next channel next decision. I would love to hear from each of you what the holy grail of KPIs for marketing effectiveness look like at your organization right now. This is probably going to look different for everyone. It might be just based on where you are in your evolution, where your organization cares about. Andrew, I’d love to hear from you. What are the marketing KPIs that you’re most interested in looking at?
Andrew Chang: There’s four that I’m trying to hold us accountable for. Number one is Return on Marketing Investment, ROMI. Number two is Return on ad investment, ROAS. Number three is cost per acquisition. Number four is brand health. You try to measure the whole funnel the whole time. ROMI is important because that is ultimately what we would be reporting back to our stakeholders. ROAS is important because normally the largest part of anyone’s budget is the media spent. We have to always make sure that we’re using it efficiently and effectively.
Third is cost per acquisition because I think of healthcare, especially with the primary care, urgent care side as the milk that grocery stores will sell. They put it in the back of the aisle. [crosstalk] Primary care is that way. Then it’s usually a break-even proposition. Managing costs is pretty important. Lastly, is, of course, you can’t forget about brand health. That includes awareness, perception, preference, et cetera. Those are the main four key metrics.
By definition, they’re incremental and revenue-based. You have to have a system that’s able to measure incrementality and also revenue, not just clicks or form completions, lead submits, et cetera. You got to get down to the revenue and how deep and how connected you can make that down to the patient level by campaign or by whatever tactic you have is all part of that equation.
Lauren: Andrew, do you have mechanisms to think about those holistically? Are you looking at a last-click attribution, multi-touch attribution? I think that’s a big challenge for marketers nowadays is to measure something like ROMI or ROAS without necessarily having each channel’s direct contribution to be a straight line.
Andrew Chang: Everyone starts out with last touch and then move on. Then typically that moved to MTA multi-touch and then marketing mixed modeling. We’re building out a CDP right now. We’re about to launch it in the next two weeks with a Salesforce data cloud. All of our campaigns are going to start from there and end there so that we have the data flow all connected, which is also connected to our EHR so that we can track it down to the patient level.
All of our agencies are connected to it and patient data is connected to it, CRM is connected to it, et cetera. That’s really where we use that as the main source for measuring the incrementality, et cetera. Plus other AB tests that we’ll do along the way for different types of campaigns. I think I answered your question. Actually [crosstalk]
Speaker 2: You answered it and then some. That was incredible. Thank you. Super helpful to understand how an organization like yours with so many different business units and stakeholders is thinking about data in a single fashion. You’ve got all your agencies plugged in, your business owners and I like that you mentioned the tech that you’re using too. Thanks for sharing that. I think a lot of organizations are wondering where do I start? What are the best of the best using? Is it worth investing in something like a Salesforce or do I try to patch together my own systems to get myself off the ground?
It’s nice to see what a large organization at the state might look like. Ibrahim, I’d love to ask you the same question, marketing KPIs, what you guys are measuring, and challenges in that [unintelligible 00:24:36]
Ibrahim Alibaba: We may start saying the same answers, but it’s I think very similar to what just Andrew said. I will tag on lifetime value. Although lifetime value is built into the previous functions. However, the way that we as an organization operates is on a physician account level, and so what we’re trying to do is get every physician to a profitability, not the systemic probability.
In many ways, Andrew’s example of the milk in the back, it won’t work for us because we need the milk to be in the back and the milk to make money because we’re working on an account basis. We really have to understand lifetime value. Then a little bit below that is what we’re trying to now do is move towards can we understand LTV, obviously the channel by especially whatever, but then below that is LTV by maybe a condition or LTV by a service. I guess that’s the one add-on I’ll add to Andrew.
What we’re trying to do is once you understand that LTV, and then you understand your customer acquisition costs, you can work with your finance team to figure out what’s it going to take to get us to profitability? What’s the actual spend is going to make it safe to bring us to what a private care provider [unintelligible 00:25:41]? How many patients can you get? Or whatever specialty we may be talking about. I guess that’s my only add-on to Andrew’s. Obviously, there’s some other ones I’d look at that you’re on a paid media site. What’s your conversion rate down every single step of the funnel?
I think maybe there’s a question coming on later about that specific topic so I won’t get into it. Then that’s on the paid site. Then there’s some other aspects that people might want to look at. Sometimes it’s not touched upon. I think this is fascinating because it’s not really marketing necessarily, but marketing touches it, which is the functions that are touching your marketing dollars. That might be a referral team.
You may be putting dollars spent to get a referral into the door and then your referral team is not a convert them. You might start building out certain KPIs and I’ll just start listing some of the things that we look at when we’re talking about referrals, because for us, the CLS and multi-specialty clinic, 50% of our volume comes from referrals. If we’re not focusing on that 50%, then you’re really lost as a marketing team that we have to really hone in on. We’re working with that BI team to come up with those key performance indicators.
KPIs, I’ll just list them. We could say percentage of patients that are outreach within 24 hours, percentage of patients that are booked, percentage of patients that are booked within seven days. Once you can get those metrics, guess how easy it is to go back to those referring systems or doctors and go back to the village medicals and show the real data, say, “Look, you know that every single time a patient comes through the door we reach out to them 24 hours. You’re seeing it seven days, blah, blah, blah.”
It’s powerful stuff, but it also aligns your operational teams on the things that might matter to marketing as well. They do impact marketing and the same thing can be said about call center. The same thing could be said about maybe if your IT manages your scheduling form if you’re using Hilo or Epic, or whatever the heck you’re using. That scheduling flow, what’s the conversion on every single step of the way?
I’ll say one last tidbit. We found that when you show patients the scheduling calendar at the beginning rather than the end, you increase your completion rate by 50%. That’s just fascinating. You got to look at all these things because, at the end of the day, they affect all the things that Andrew said downstream, which is turn on that and spend and plus my acquisition costs and whatnot.
Lauren: I’m going to talk about propensity modeling here in a minute and how some of those things impact it. Clay, just really quickly before we move into that, your business is very different. There isn’t milk in the back and these other things. Like you mentioned early on, we have a single set of services that we offer to a very similar set of individuals. What are your metrics that you guys are looking at and each of your locations being a separate business in it and part of a market and part of a system?
Clay Nickens: We hold our marketing team to three metrics. The first is access, which could be measured by qualified lead generation. Expanding access, making sure that families are hearing about us and considering us in a fulsome way if they’re ready for services. The second is efficiency. How much are we spending for each of those leads cost per lead? The third is what you’re referencing on us being multi-location with several hundred locations that each need families who are considering services to be considering them.
It’s not just the lead generation, but it’s the accuracy versus the availability of each of those centers. Some centers have been there longer. They’re more mature. They only have a handful of openings. There are others that just opened this week and could accommodate a couple of dozen families over the course of the next few months.
Lauren: Absolutely. It’s interesting. There’s a lot of overlap and you guys each are in different phases of determining what’s most important to the business. I’m going to stop collecting this baseline information like I always just do all the talking and let Rich ask a little bit. I know we want to get into forecasting propensity modeling, and some of the really cool stuff that’s to come maybe once you hit phases one and two like Ibrahim mentioned early on.
Rich: Lauren makes my job really easy. She saves my voice. Obviously, we’ve talked about a lot of the foundational layer and a lot of the groundwork that has to be done to get to a point where we have the data that we can then enable some of the cool, more fun stuff to do. Rather than forcing the conversation down a specific path, I’d love to know from you guys, what is the most high-impact thing that you do with your data as an organization once you have it readily available to you?
We talk about ROMI, we talk about ROAS. What’s really helped you with data enablement to move the needle the most as an organization by leveraging that data? Again, just to not break with the pattern, Tom, I’ll pick on you first.
Tom Stanton: All right. Sure. I think, for us, since I didn’t get in on that last question, also the brand health is a big one that we take a look at. That is a lagging indicator that’s very meaningful to us that have used a lot. Especially I mentioned that we are spreading out in the entertainment marketing space, which is a new frontier completely for this industry to look back and see the positive effects on those [unintelligible 00:30:34]. Then we did also some modeling of what that impact looked like too based on what we’ve got.
It was a forecasting-based measurement of what was the traffic that was coming into our website, projected out what a typical curve looks like at that time of year, and then looked at what actually happened and it’s just a flatline versus a mountain. It was an impactful visual [unintelligible 00:31:01] that we were able to distill down into– It’s an estimate of the upper mentality of just the exposure of that show on not only those brand health metrics but also people actually coming to check us out and making their way into appointment flows too.
Those are the [unintelligible 00:31:19] stories that we have. It’s mostly slow wins here and there, but those are some of the interesting ones that we want to keep building on and use that for smaller-scale campaigns. Just one of the areas recently my team was working on doing, we were at the close of budget season here for 2025 and using actual historical return on channel investment to try and get to the answer of what’s the right spend level for paid search, for example. We go and take a look back at over the course of time, we’ve invested at these different levels and we build out a diminishing return curve, which is nice because the theory matched what the data was actually telling us.
We were able to reach that point of what’s the break-even because we’ve got an acquisition cost that we can put on top of it. We could say what we used was a marginal acquisition cost. Every additional X number of dollars that you spend, how efficient is that? You keep going up and up and then we identify where that sweet spot is. That was a pretty cool and very recent one.
Andrew Chang: Tom, I’m curious, and Rich cut me off of if this is too far, but how noisy was that as you’re building towards it because there’s so many things that can happen that influence the outcome of your measurement. How do you balance signal noise and getting an answer that’s acceptably accurate?
Ibrahim Alibaba: This one, we aggregate it at a monthly level. The more you zoom out, the more you can smooth out that curve for sure. We were good because we had a lot of variation in one area of the business that we were focusing on where it proved out to follow the pattern we would expect. In another area, yes, we couldn’t go further because there wasn’t enough variation in how the business spent before.
We can say this section of Long Island, we know what the [unintelligible 00:33:14] but in Manhattan where we are pretty stable with what we’re spending, we said, “We don’t really know what the incrementality looks like. We can’t make a recommendation there.” This we also like because it leans more into the conversation of, “Hey, you need to get more experimental. Let’s work together. Just test and learn.” We’ll try things, we’ll collect signal and then we’ll know more so that we can inform the future
Rich: On that and picking up on something that you said earlier Andrew about the deployment that you were going through currently with Salesforce and tying everything together and then providing access to your agencies, how do you empower your agencies to help with some of this data actioning and some of the better decision-making based on the data that you’re making available to them? What has that successful partnership with agencies look like between your BI team and their teams to make the most of the data that you’re making available to them?
Andrew Chang: The way that we make it actionable is in a few different ways. Number one is that from an existing patient standpoint, which is the majority of most healthcare’s patient base, it is really trying to understand at a very granular level because of today’s day and age of how many different segments we can create. I believe we should be creating a lot of segments down to potentially ICD-10 or CPT codes to define who we’re really going after. That’s part of the data that’s being fed into our CDP today.
Based on that, we’ll make specific segments like one campaign could have 10 different segments. Yes, it does increase your cost, but also guess what? Because it’s relevant it also increases your return based on the very defined segments that we create. Then we also use third-party tools with Experian and Mosaic modeling to tell the agency these are the types of psychographic or mosaic segments that you should be going after, or these are the lookalikes that you should be creating to go after for new patient acquisition, or here’s the IDs of the actual people that we really need to click to confirm that appointment or book that appointment.
They started down the path and they didn’t finish. Now we’re going after them, trying to get them to book that appointment. Obviously, data connection to the agency, whether they use Trade Desk or whatever platform they’re using, is absolutely essential because again, it feeds back into our CDP and then we’re able to tell the agency or the agency can see for themselves which segments are performing for which creative and how we should be tweaking our different strategies for whatever platform we’re on and whatever the goals are.
Rich: Got it. I know we’ve only got three minutes left. I was going to say each of you tell us all what your biggest focus is with data in 2025, but it has to be within 30 seconds. I’ll do the countdown. No, I’m just kidding.
Lauren: Awesome. We’ve got Andy off mute. We’ll start with you.
Andrew Chang: Mine is pretty simple. It’s successful execution of our new marketing OS, which involves what I just talked about, starting with Salesforce, with definition of segments, and then being very strict and disciplined on sticking to that process for creation, strategy, measurement, everything.
Lauren: Awesome. Tom?
Tom Stanton: It’s finishing out a lot of the data quality work for sure. What we want to start building on top of that is where we model our business, where we plot out, to Abraham’s point, every point along the conversion funnel, we identify what are the value drivers, what are the KPIs, if anyone’s interested to dig in more, it’s the concept of a metric tree where you actually get into what every single piece is and how that ties into your business model and customer journey.
Lauren: Ibrahim, I know we had a similar line of questioning when we were discussing with you the metrics along the patient journey, but I don’t want to put words in your mouth. Your biggest focus for 2025.
Ibrahim Alibaba: I think two focuses. Attribution modeling, improving that across the board. Then lastly is exactly what Tom just said and what I’m really focused on right now is we’re building a new website and making sure we understand that site and how users interact with it at a very granular level so we can track all those things that we just talked about.
Lauren: Always a huge project. Clay, I’ll have you bring us home. Biggest focus for next year, any tools or tech that you love. What are you focused on?
Clay Nickens: Is it bad if I say all of the above? The core mission that we’re on from a marketing perspective, what I mentioned earlier is being able to reach earlier into the funnel, build those relationships, and have data-oriented with it. We have a great entrepreneurial culture. People are excited to go out and test and try things. Having the clean data, the multi-touch, the ability to do uplift analysis on mid or early-funnel tactics that enables our team to have a very high cycle rate on those tests.
Lauren: Awesome. You guys heard it here first. If you are looking for your takeaways from today’s segment, do not feel the pressure to leap all the way to I need to be building a propensity model. Let’s start by getting the right data, the right clean data. Let’s do it once as Tom would say, and then grow into those things. Don’t try to leapfrog steps one and two. It can be overwhelming. You end up feeling like you’re a hamster on a wheel and never really reaching a desired [unintelligible 00:38:27]. We had a million more questions for this panel. I could talk to you guys all day long. My favorite subject of the two days, but we will continue those conversations offline. Thank you, guys, so much for joining us and thank you everybody for listening. I think we’ve got a little lunch break now in the panel sessions and we’ll resume back in about an hour with our afternoon sessions. Thanks guys for joining us.
Andrew Chang: Yes. Thanks for having us.
Announcer: Thanks for listening to this episode of Ignite. Interested in keeping up with the latest trends in healthcare marketing? Subscribe to our podcast and leave a rating and review. For more healthcare marketing tips, visit our blog at CardinalDigitalMarketing.com.