Patient Acquisition Needs Programmatic Infrastructure

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.

Lauren Leone: Hey, everybody. Welcome back to Ignite, Healthcare Marketing Podcast. Today, I have Lisa Fisher, who is our director of programmatic and social media. Lisa, an interesting topic today. When I read the prompt for this episode, “thinking about programmatic as infrastructure,” I asked myself what exactly does that mean, and who better to ask before we dive into the episode. What does the phrase “programmatic as infrastructure” mean?

Lisa Fisher: That’s a really great title for the episode that we’re doing today because there’s a shift in consideration of programmatic from just blasting banner ads and having ads that are hitting people maybe occasionally with a low frequency, and shifting that into more of a sustained infrastructure that’s always on and that’s responsive to the needs of each individual marketer in real time. Rather than just having a display banner that pops up on a website and then has no context and it’s not following up with that audience or cultivating that audience, it’s really building and establishing cookie pools of who your most relevant consumer or patient is going to be, and then having a responsive system in place programmatically to be able to handle any market changes as they come up.

Rather than having just one-off siloed campaigns for programmatic where you may be putting 5K here, 5K there, and it’s not really having a consistent thread of performance that the algorithm is able to optimize off of, it’s having something that is more sustained and long-lasting and able to respond in real time to the needs of your business, and also continually driving users closer to the conversion event over time through a narrative and a storytelling format that really makes the brand more personal to them and connects that storytelling a little bit more.

Lauren Leone: Yes. What are some of the pitfalls? We see this used as these really short-term pushes. Examples of that, because I think people listening might be able to say, “Oh, yes, that’s me. That’s how I’m treating it right now.”

Lisa Fisher: Most of the DSPs have some forecast tool where you can go in and play around with the audiences that you want to reach and the GOs that you want to reach and figure out what that audience size looks like, and look at the scale and look at the reach. Those tools are often overlooked because they spit back numbers at you that may be higher than what your budget is. You just say, “Oh, I don’t have that budget. I’m not going to go in that direction.” Ultimately, you can use that directionally to show you, based on the amount of this audience size, really, does it make sense for me to put these dollars into programmatic if I’m not putting enough in for a timeframe that’s long enough where it’s really going to have an impact?”

Those forecast tools, I would say, should definitely be used in every case to look at your reach and to look at your feasibility in a market. If the reach and planning tool is not showing you that it’s going to hit less than 10% of your total audience, it’s really just putting money into something that’s not going to do much. You want to make sure that the amount of impressions that you’re able to receive from that investment is decently sized within the overall total reach of what your audience is.

Maybe you’re trying to hit 50% of audiences within geography and then having a frequency and place that is not overwhelming them with messaging, but that’s also keeping it top of mind. Really using these reaching and planner tools to calculate the amount of investment that’s needed because otherwise you’re going to get a few hundred impressions here and there. They’re not ultimately going to drive anything in the algorithm that is a sustained growth for your business. It’s not going to convert anywhere at the bottom of the funnel. It’s just going to go off into the ether.

We really want to make sure that programmatic is as smart as the information you’re feeding it. If you’re feeding it consistent data at a higher volume, it’s going to get smarter, and it’s going to do more for your business. It is an act of trust in the beginning. If you’re not prepared to invest that much at the beginning, I would say if you only have $5,000 here or there, to look first at what you’re currently doing on search, social, and other channels and see if you can invest or increase anything there first rather than branching off into programmatic with a really small budget, which is what a lot of people try to do.

I would say, wait for programmatic until you’re really fully prepared to invest in that channel to the right degree, where you’re going to see measurable results that are beneficial for your business. Otherwise, it’s not beneficial. You’d be better served putting that back into search, maybe doing demand gen on search or other channels that are already operating for you.

Lauren Leone: Lisa, you mentioned $5,000. Obviously, it’s all relative. If you’re operating in a small DMA, that could very well be enough to sustain a three-month run. Oftentimes, at least our clients are thinking bigger, but the budgets are such a small percentage of what they’re doing.

Lisa Fisher: Yes. My $5,000 reference is more for a national campaign. I think that’s why the reach and planner tools are so important, because sometimes the audience is very niche, sometimes your geography is small, and that will serve the purpose of what you need it to do. It’s all completely relative. I would say when you have a larger, more national campaign, definitely you need to make sure the investment is there and the timeline is there for that true infrastructure to be in place. For some smaller and more agile markets, it can definitely make sense to test into it. Make sure that you’re working with the DSP, working with your partners in the platform to run the numbers first and understand who you’re actually going to be able to reach with those dollars.

Lauren Leone: Now, I want to go back. You mentioned conversion. It won’t convert. It won’t do anything for the business. It’s really hard to measure when programmatic is doing something for the business. Can you help define that a little bit better, and what you think it should look like? What does good look like?

Lisa Fisher: Yes. This has been a constant thread of programmatic since I’ve been working in it since 2009 is how do you give credit where credit is due to programmatic, because a lot of the activity is coming through as a view-through and last-touch type of engagement rather than last-click engagement. A lot of models use last-click for defining your conversion event. In the case of programmatic, you actually don’t see a lot of last-click activity coming through. That’s a tough pill to swallow for more traditional marketers who are used to search marketing. That’s last-click.

Programmatic often happens earlier and throughout the journey rather than coming at the very end, which a direct search would do. Oftentimes, search is getting the credit for that last click. For programmatic, what we can do is there are incrementality measurements that a lot of the DSPs offer that will show you of the amount of people who ended up searching and converting. It was X% higher incrementality-wise because of the amount of media that we were running in the upper funnel.

Lauren Leone: You’re withholding certain portions of the pop.

Lisa Fisher: Exactly.

Lauren Leone: That’s when we feel like this is what happened, but there’s a real measured holdout group.

Lisa Fisher: Right. They do like a ghost bidding methodology to say, “Of the people who were not exposed to our ad, this is what the conversion would look like had they not had any exposure from programmatic.” That’s an actual portion of the population that they’re testing, versus, “Here are the users that we know have been exposed to a CTV ad.” Then what their conversion behavior looks like. It is significantly higher, the amount of incremental lift and relative lift that’s coming from those channels.

We can also look at the actual impact that upper and mid funnel and lower funnel on programmatic is having on paid search. We actually did this recently for one of our larger clients, where we increased investment year over year. They had a website traffic problem. They wanted more website traffic, so we increased–

Lauren Leone: Organic and paid.

Lisa Fisher: Yes, organic and paid. It wasn’t just coming– Actually, paid was doing pretty well. It was organic that was suffering. They weren’t getting organic volume to the site. We started running more awareness and consideration messaging. This was programmatic and paid social, but both of those channels combined doing more video storytelling, CTV. We ran this for about six months. After that point, we measured all of the results. These are actual results from this case study of what we ran. New users to the site organically increased 3X. Then 22% of those new users came back again for another visit.

We also saw the average time spent on site from organic visitors increased 119% year over year. That’s a really interesting metric just to show that people who had a CTV exposure had more awareness of the brand. When they got to the page organically, they thought about it. They came back later. They were spending more time on the site because they already had an idea of what they were interested in. Then also very cool because it integrates with paid search, we had a 44% increase in brand search impressions.

Lauren Leone: We love those.

Lisa Fisher: Yes. Overall interest in the brand increased to the point that we were able to spend more in brand search. Then brand search is very bottom funnel, ready to engage, ready to convert. We were able to see higher conversion rates on search because of the upper funnel we were running in programmatic.

Lauren Leone: It all creates this environment that works well together because you talk about metrics, more organic traffic, more engaged organic traffic, more time on site, more pages visited. Those metrics are the metrics that inform the organic algorithm to show the page more. You create this snowball effect that compounds when you layer these things in. It’s not, “Did this channel convert? Did it not?”

Lisa Fisher: It’s not in a silo at all. It’s not, “Was there a click and then a conversion event from programmatic?” It’s impacting your organic traffic. It’s impacting your engagement. It’s increasing your brand awareness, your overall trust with your community as well. It’s doing storytelling. Then there’s also retargeting that you can do from programmatic into paid social. Not all DSPs can do this, but there are a bunch that can, where you can actually ingest your audience of people who’ve been exposed to CTV and programmatic into Meta, and then continue the story from there. You can really do things with-

Lauren Leone: Keep engaging.

Lisa Fisher: -with them across platforms. Exactly. Keep engaging with them across platforms, sequential messaging, storytelling, building your narrative.

Lauren Leone: Lisa, you’re saying a few words like retargeting and this full funnel measurement that might seem a little scary to people who are from a HIPAA compliance standpoint. In order to measure incrementality, you would have to have a pixel on site. You would have to be measuring how is that possible? What kind of DSPs are we working with that allow this to happen? Because not all of them will enable this.

Lisa Fisher: The first thing is whenever you have a pixel on a site, it can never be used for targeting. It can only be used for reporting. We’re not ever using pixels to retarget directly from a site visitation. That’s an instant no-go. The pixel is only used for measuring and then allowing the algorithm itself to see what your audience looks like without any PII attached to it for modeling purposes.

Also, we work with DSPs that have a BAA, which is a business authorization that allows us to have some more of that tracking in place that is completely safe and has privatization that we’re not sharing it. It’s basically the company that we’re working with, the DSP, accepting full responsibilities and saying that they will not use it in any way, shape, or form. It’s completely brand safe to have a BAA.

Lauren Leone: It informs this brand’s campaign, but it’s not part of informing their broader algorithm, which is really important.

Lisa Fisher: Exactly. It’s not to be used in any other case. There’s a lot of legalese behind it saying that the data is completely private, not used in any type of marketing or being taken off for other cookie pulls, anything like that. It’s just strictly used for that campaign. It’s not used for retargeting again, even when we run with BAAs. We’re not actually doing site retargeting. We’re only using that data to inform the algorithm who is actually converting so that it knows if its models are correct or not. The retargeting that I’m mentioning is people viewing a video on CTV and then being served another video. It’s never a proprietary private interaction that they’re making on the web. We don’t retarget off of that type of tracking.

Lauren Leone: Yes. I think short and simple here is check what your DSP, if you’re using one, is capturing and how they’re using the data. There are plenty of DSPs out there that serve the healthcare industry specifically that we’ve vetted. We have a range of them because everyone’s risk spectrum is different, too. There’s different degrees of what DSPs we would use based on the organization and their compliance team.

You mentioned the six-month test that you ran. I’m sure you get results like that, and you want to keep it going. You think about this traditional buy, and you place this buy, and it runs for six months, and it gets stale, and it doesn’t adapt. What if something comes up in the business, Lisa? What if our service line priorities change? Can it be adaptable and fluid? Do you have to set it up for six months and let it do its thing? Talk a little bit about the real ideal in between.

Lisa Fisher: I think that works. I think at the minimum, you would want to run it for four to six weeks. I think the tipping point of when you start to have statistical findings that are indicative of a deeper sense of performance and not based on the whims of the market. I would say four to six weeks minimum to start. Then what you’re really looking for is performance will start to stabilize and become more in line with your benchmarks. You’ll start to see the measurement become more refined over time. The performance will improve, but it typically takes two to three weeks for performance to start to settle down and get smarter.

Lauren Leone: Really, what I’m getting at is just because you have a buy that has a long duration doesn’t mean within that you’re not testing creatives, testing audiences. It’s a fluid and adaptable buy.

Lisa Fisher: Absolutely. I would say when you’re running a programmatic campaign, you want to be testing something. My rule of thumb is every month or two. I wouldn’t test too many things because then you don’t know which variable is impacting your success. Testing a PMP deal or testing a different type of audience group, but always keeping that business-as-usual campaign running at the same time with the majority of investment. I usually only put 10% to 15% into the testing opportunities, but it’s going to ultimately unlock capabilities that you didn’t even know existed.

We’re testing a new bid optimizer that uses AI to determine when and where to bid. We’re still using audiences that we’ve powered and determined, but it’s more the automation of the bidding system. We tested that with 10% of budget, ended up rolling it out to everything that we’re running because the performance was so strong. You need to make sure that you’re testing, but that you’re allowing enough time for the results to truly become fully baked. Then from there, you can make the determination of what you’re going to continue with. That’s why a four to six-week period at the minimum is what’s recommended because it takes time for your audience to start to engage with your ads and to be able to really reach the right audience.

Lauren Leone: Yes. Again, it’s all relative. We can give a guideline of four to six weeks, but let’s say you did your audience planner and your budget came in at the low end of your recommendation of, “Hey, maybe 20% of the potential reach.” You’re going to run a much longer campaign to get meaningful statistical significance than someone who maybe has the budget to capture 50% or reach 50% of the target audience within a planned buy. All the numbers work together. These are guidelines.

Lisa Fisher: That is a really good call-out, though, that the less you spend, the longer it’s going to take for you to get results. If you’re able to spend more upfront on testing, you will have a much better starting point later. Also, totally understand that’s not the case sometimes. Most times that’s not the case. You have to start small. That’s fine. Just know that you’re not going to be able to test that many variables early on. You really want to pick one. Let’s say you want to drive phone calls to a call center. Pick that one goal and go with that for weeks, if not months, before you’re moving on to a second goal, expanding your markets, expanding your audiences.

The system is as smart as we allow it to be. When we’re giving it that dedicated time for learning and intelligently organizing our thoughts and telling it how we want it to run, it’s going to do the best work for us. That’s how it’s going to be long-term, sustainable platform.

Lauren Leone: Yes. When you talk about that sustainable, really thinking about this, I get the four to six-week testing, the six-month test, but once you’ve done those, you’re generally adopting this as an evergreen portion of the strategy. What metrics do you want to see growing, shrinking? What to you tells you that this is continuing to be an effective part of our strategy?

Lisa Fisher: Obviously, we would look at conversion rates, first and foremost, and see if our conversion rates are increasing, not just on programmatic, but also look at the other channels. Programmatic has a huge impact on organic search and organic website visitation. Is overall conversion from an organic search to a conversion improving? Are people searching more for certain terms? Looking at overall conversion rates across all channels.

Also, looking at the pool of users that are qualified. Those retargeting pools that we discussed, which are not based on pixels, but are based on engagement with the video or engagement within the programmatic ecosystem. It’s not tracking any of their on-website behavior, but more their in the DSP, what they’re ingesting. Able to look at the overall cookie pools and see if those are growing and see if we’re getting more qualified users. That will help the conversion rates increase on all of our other channels.

Looking at overall lead quality, too. A lot of our clients have certain insurance that’s required in order to become a lead. Making sure that the leads coming in are quality leads, that they’re actually qualified. In order for the algorithm to be working properly, it should only be getting the most qualified audiences that have the right insurance, that have all of the right parameters that fit our needs. You should ultimately start to see your qualified lead rate increase as well. Then, if you invest a little bit more and you’re still seeing the conversion rates hold and you’re still seeing the CPAs hold, you know that you’ve got a scalable system that is optimizing correctly and can scale.

Lauren Leone: If the business need is there and there’s desire to grow or to improve efficiencies, you can start to potentially increase the investment in that channel as part of the overall mix. Absolutely. Once you’ve created this stability, it’s a great place to think about, where does my marginally inefficient dollar from search or somewhere else go when I have this pressure of, “Hey, I need more patients in this market this month?” Adding that dollar to search is not going to drive anything else. You have to have other places to go with it.

These are the kind of systems and infrastructure we’re talking about building beyond bottom funnel that help you have a program that can really grow and scale for years to come.

Lisa Fisher: I have one thing to add. If you have a demand problem and you’re maxed out on what you can do on search, you should look at programmatic to activate from a demand gen perspective. You’re going to be getting more qualified audiences because we can use third-party targeting, which you can’t use in Google as well. Google is only based on segments that Google has because it’s a walled garden, but within programmatic, we can use third-party audience segments that are HIPAA compliant to target people based on specific needs, and we can also do things like disease awareness, which is really important for a brand to be an authority in their space and to establish trust with users, so that brand awareness component, that demand generation is really ideal for programmatic to have on in the background if you’re maxed out on the search side.

Lauren Leone: Yes, absolutely. This episode, really talking about the infrastructure of programmatic, but third-party audiences and what we can do to build and continue to grow those audiences, a whole other topic that I know we’ve covered and will continue to cover because it’s so important. I think, Lisa, you had a really great lead-in to what will be an episode in the weeks to come, as well, around the go-to-market strategy itself. This is about the buying itself, but this has to be paired with creative go-to-market strategy, and that is, you can stand up this infrastructure Lisa’s talking about all day long. If you don’t understand your user and the journey that they’re on, this will fall flat.

Unfortunately, what you will end up doing is ruling out a channel, potentially writing it off for the short or long term, that really should be considered. That’s the hard part. There’s so much that goes into making this successful. The barrier to entry does feel a little bit higher than some of the other channels that are a little bit more straightforward. I’ll ask you, Lisa, to round out the episode. First step for a marketer who’s like, “You just described me perfectly. I’m stuck. I keep running flights. I keep running silos. I only use programmatic to support de novos.” Really, here and now, problem statement is what’s one thing they can do today to inch their way towards thinking about programmatic as part of their infrastructure long term?

Lisa Fisher: I would say, depending on who the audience is for this client, I would explore third-party audience targeting to first narrow in on the people that you want to be reaching. Then I would say, look to your creative that you have that’s been working on other channels. If you have any video creative, I think video is a really great place for programmatic to start because video takes a little bit longer to work. You need to see it a few times, and it’s telling a story, ultimately, that’s not going to just change minds overnight. You need a little bit of a ramp time there.

I would say, look at the audiences that you want to run, figure out how much money you’re working with, and then from there, define the audience size so that those dollars are actually going to have an impact. I would recommend activating on a video-first channel to start with programmatic, something like online video or CTV. Then from there, you can build pools of people who are engaging with and viewing that video and target them with more display, static imagery, native imagery. Within programmatic, still, that is a little bit more stronger call-to-action, more educational, but you really have to start from that awareness perspective, starting with video, and then ultimately leading those users on a journey to learn more, be educated, and then ultimately click through to the website.

Lauren Leone: Absolutely. It was great to have you, Lisa. I feel like we could talk about this topic. There’s so much to unpack, and we will do more of it. For those of you listening, challenge yourself, your internal team, your agency to consider whether this is the right time to think about how to build it in as part of your infrastructure. Big advocates for programmatic over here, and hope to see more of it in healthcare in the year to come. Thanks, Lisa.

Lisa Fisher: Thank you.

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.