In the final episode of our “AI for Healthcare Marketers” series, we’re getting hands-on with AI platforms. While understanding how AI platforms function is important, it’s when you start using them in your workflows that the real value shines through. This interactive episode walks you through five practical exercises designed to make your daily tasks smoother and more efficient.
The exercises covered:
•Using ChatGPT for quick note transcriptions
•Brainstorming content ideas with Claude
•Conducting market research with Copilot
•Navigating privacy concerns with Perplexity
•How AI can improve communication
Connect with Jenny:
•Email: jenny@hedyandhopp.com
•LinkedIn: https://www.linkedin.com/in/jennybristow/
If you enjoyed this episode we’d love to hear your feedback! Please consider leaving us a review on your preferred listening platform and sharing it with others.
Jenny: [00:00:00] Hi friends. Welcome to today’s episode of We Are, Marketing Happy, a healthcare marketing podcast. My name is Jenny Bristow. I am your host and I’m the CEO and founder at Hedy & Hopp. Hedy & Hopp is a full-service, fully healthcare marketing agency, and we specialize in working with providers and payors across the U.S.
Today is episode three of our three-part series titled AI 101 for Healthcare Marketers. We designed this series to really allow you and then anyone on your team you forward the series to really understand how AI works, understand the different tools and models that are available to you, understand some core tenants of using AI within your marketing function.
And then really getting your hands dirty and doing some prompt engineering and getting some results. You got to move beyond theory in order to really understand something. So that’s today’s goal. If you are listening to me today, I strongly [00:01:00] recommend that you have me in one ear with an earbud and you have your laptop or computer in front of you, because I’m going to walk you through some prompts that are going to allow you to really get your hands dirty.
So this is really meant to be an episode that you listen to for a second. Pause, listen to again, pause. And it may take you an hour to get through it simply because you are being interactive and working through these different examples. And I think we have six different examples we’re going to be walking through.
But again, I strongly recommend do not skip this and please do not just listen to it and not do the prompts real time, because I know you say you’ll go back and do it later, but I assure you that you won’t because you are busy. An you have a million and one priorities so let’s please make this interactive. So we’re going to use a handful of different platforms.
The reason why we’re going to do that is because I want you to see what the output looks like. I want you to create a free account on a variety of platforms so you’ll go back to it and use it again. And I want you to see what they look like and the outputs that they [00:02:00] generate. So, the first one we’re going to do is Chat GPT, and this is actually something that you’ll need to use with your phone.
So take a moment, go and download ChatGPT, the app to your phone to your iPhone or Android, and it is as of the recording of this podcast so mid December 2024, it’s a little green icon that you’ll want to download. Once you download it, you’ll have to create a free account if you haven’t already. But this is one of my favorite uses, use cases to teach people that are kind of afraid of prompt engineering.
But it is a huge time saver. So I myself, I will show you for anybody watching the video, I am a handwritten note person. A study came out 15 years ago. I think that said, if you write during a meeting, it actually makes the information be cemented multiple parts of your brain. And, you know, I’ve never followed up on if that study is accurate or not, but I am [00:03:00] a believer.
So I’m constantly doing pages of handwritten notes. And then at the end of every day, I’m having to actually transcribe them and text them in in order to share them with my team. So this is a great hack. Using the ChatGPT app, open it up in the lower left-hand corner, hit plus button and go to camera.
So you can take a picture. Take a picture of your page of notes. You also can take a picture of a PDF. Let’s say a handout, like a marketing sales collateral piece. Take a picture. And then once you do that, hit submit, and then it will ask you, what do you want to do with this? And one of the options will be to transcribe.
So respond back. Yes, please transcribe. And then what it will do is actually then will transition that image, whether it’s handwritten notes, collateral, whatever, in to text. So it does such a good job reading handwriting. I have very bad handwriting and it does an excellent job reading my handwriting and trying to decipher the organization of content off my notes.
So if I had [00:04:00] something on the far right, it does a good job putting it where it needs to be in bulleted format. Once it does this, you can actually select the content. You can text it to yourself, jump into Slack, email it to yourself, whatever you need to do. But if you are a handwritten notes person like I am, this could be a great way to use AI technology to quickly eliminate that time consuming task of texting or typing in your notes.
Workshop number two, using AI for content brainstorming. I really like this from a content marketing perspective. So again, whenever we talked about the AI tenents, one of the core ones is not using output as a marketing collateral. Like we’re not going to take something that Claude writes and just use it as a blog post, right?
We’re not doing that for a variety of reasons, legal concerns, quality concerns, et cetera. But it’s great for brainstorming ideas. I really like it because it helps get out of my box in my own [00:05:00] head of the biases I have of what I think matters, what I think people are talking about. So we’re going to use Claude for this one.
So open your browser to Claude.ai, create a free account if you don’t have one yet. And then here is your prompt. So first we’re going to tell the platform we’re going to tell Claude their role, we’re going to give them a little bit of context and then we’re going to do the ask and we’re going to do a general ask and then we’re going to get more specific with the ask.
So here’s the prompt. And again, I’m going to read this. You can pause this podcast and then you can think about how you’d modify it for your own needs. You can type it in as is if you want, or you can modify it based off a project you want to do content brainstorming for your own organization and modify it.
So prompt, you are the chief producer for my podcast We Are, Marketing Happy. We create concise educational videos for healthcare marketers. Please create a visual mind map for all the video topics I [00:06:00] could write on about healthcare marketing for providers and payors. Please include the article title in a short summary for each topic.
Make the title and summary punchy and use lessons from David Ogilvie to craft them. So if you think about that prompt, we did a couple of things, right? Okay. Told Claude who they are, gave them context, and then said what we want the output to look like as far as the format and the copywriting approach.
This is really important because you could also have the copywriting approach be very professional and buttoned up or very casual using slang, modern slang language. There’s lots of ways that you can modify that to have it be appropriate for your brand. And for the output, I actually asked it to do a visual mind map.
Again, you don’t have to do that. It can, when it does the visual mind map, it also creates just like a piece of like written text to accompany it. That could be really helpful. Workshop number three, we’re going to use Copilot for this one. Again, just to get [00:07:00] examples using all the different platforms. So go ahead and open up Copilot, make an account.
If you don’t have one yet and here’s the prompt. This is around market research. I want to open an urgent care facility in St. Louis, Missouri, where my customers can get quick and easy access to care at a standard rate. Act as a market researcher and give me information on the demographics of people most likely to use urgent care in St. Louis, including age gaps or age groups, income, education, gender, and specific location. For queries like this, you’re likely to get back a lot of information that you already know. After all, you have a job as a marketer in healthcare. You’ve been doing this for a while. Some of you for a long time.
I’m 20 years into my career. I’ve been doing this a long time. Of course, I could do that on my own without using AI, but I like it for four key reasons. Number one, it usually adds perspective or other angles to things that I may have missed. Right? Assumptions that I may have [00:08:00] made. It uses language the community uses.
So specifically in that prompt, it ended up naming places like Central West End and other things that are just included in the day-to-day conversations that folks in St. Louis use. It challenges biases the team may have developed internally. So maybe with a specific service line, you think, oh, people come to us for pricing or people come to us because of this, and they always go to our competitor because of that.
Do they, or is that just what you’ve been telling yourselves for the last decade? So prompts like this are helpful for that reason. And then also when you tell it at the very end, if you add, please cite your sources and give citations it can find sources for you that are more likely relevant to your goal.
So there are general market research resources that you may use to pull data, but it also may share information for some lesser-known regional publications or additional contacts that maybe you wouldn’t have found otherwise. So it can be really helpful. The fourth [00:09:00] is around privacy. So for this one, we’re actually going to use Perplexity AI.
And actually, if you want to be if you want to be a little saucy, you can do it on Perplexity and you also can do it on Gemini and you can compare the results. But the prompt in this situation is around privacy. And once you opened Perplexity.ai, the prompt is going to be, please tell me the privacy and user concerns a healthcare business would have using Facebook pixel on their website, considering the most recent HIPAA guidance.
Now, this is interesting to look at the output for a few reasons. First of all, you can see if it’s biased or not, but also, so, again, I’m filming this mid-December. About three weeks ago, HIPAA, sorry, Meta dropped a bunch of additional guidelines around conversion tag usage for healthcare companies. Does Perplexity, Gemini, or any of the other tools share that with you?
How up-to-date and recent [00:10:00] is the information that they’re sharing? A lot of these models can access the Internet, but unless you tell it to, sometimes it just relies on the indexed data that has available. So, should you take any of these outputs as gospel? Absolutely not. But what it can do is actually summarize a lot of the issues around the topic in a tidy bundle.
Once you get that, then you can check into the sources that it cited, make sure that they’re accurate. You can learn some of the legal terms or issues you need to research. So, for example, maybe it will bring up some special considerations within your own state related to private state-level privacy laws, et cetera, so just flag some things you need to dig into and then also can give you a starting point to have a knowledgeable conversation with your other team members, like legal compliance analytics and development.
So whenever we are wanting to step outside of our own area of expertise there’s a whole nother culture and set of acronyms in all these other departments. So the more that you can [00:11:00] understand how to bridge that gap, the easier it will be and these tools can help you do that with queries like the privacy one we just shared.
And if we really want to dig into communication our final examples, we’re doing five, not six, I apologize, is around communication. So, for this one, you could really use any of the platforms that you like. I would probably prefer if I were doing this on my own, I would probably select Claude just because of the how human the responses are that Claude provides. But again, dealer’s choice.
You could even do it on multiple platforms to compare examples, but there are three separate prompts I’m going to give you. So you’d want to do each of them separately, but what we’re going to do is I’m going to tell you the prompt and then I’m going to tell you why you might want to do that prompt or other use cases within your organization or processes.
The first one is how do we speak about service line or disease, let’s say cardiology with the public? What is common nontechnical terminology that is [00:12:00] used? And what are the biggest concerns we should address? So, again, this challenges your organization’s messaging around service lines or diseases. It challenges biases or assumptions.
And it really can help you kind of get out of your own head whenever you’re thinking about how you’re talking about a service line. If you have service line campaigns launching in 2025, this is a great thing to run for all of them to make sure that your messaging is still relevant. Your landing page copy is still as comprehensive as possible.
And you’re being thorough for your patients. The second prompt. I think my business needs server-side Google Tag Manager. Can you explain that to me and help explain it to other people on my team? This is a great one if you’re trying to understand something that another team owns, and you need to be able to walk into a meeting and look like you kind of know what’s going on.
A great example of this is we were at, when our team was actually setting up server-side Google Tag Manager, we are setting it up on a Google Cloud platform. One of our clients. [00:13:00] Their internal team had only used traditional servers. They were asking us a lot of questions about using like traditional server language that just weren’t really relevant for Cloud servers.
And so our team was having some problem bridging the gap. There’d been like seven emails back and forth saying like, no, that’s not how these Cloud servers work. Instead it’s like this. And we just weren’t able to connect based off of that gap in vocabulary. Our team actually used a prompt that actually said, how can I explain this Cloud server set up specifically around X, Y, and Z to a person who’s more familiar with a traditional server structure?
And it gave all the answers, of course, our team read it, fact-checked the technical setup, et cetera, but once they did, they sent it over and the person on the other side of that, oh, awesome. Cool. Proceed. Right. So there was just a communication hump they couldn’t get over. And then the third one, this is great if you are walking into a meeting and you’re going to be expected to be knowledgeable about something [00:14:00] and you have no idea what’s going on, really great foundational knowledge.
So let’s say for example, here’s a prompt. We are interested in media mix modeling. What are some of the pros and cons of this that I should be discussing within my organization? So let’s say, for example, your manager put a meeting on your calendar for next week around a topic. And you’re like, Oh man, I’ve been so busy with these other topics I haven’t even really researched or learned about that yet.
You can type in a prompt like this to one of these platforms and say, give me the 101. What do I need to know to be able to speak on it at a foundational or intermediate level? And these platforms can give you the information and then you can even have a Q and A session to be able to make sure you’re rounding out your knowledge.
So a great way to use it. So why do you, why should we care about this? Right. When the first episode we talked about and we covered about how where we are with AI right now is sort of like where the internet was in the late nineties. People could still do business without it, but my gosh, once you started [00:15:00] doing it, it was easier.
And then once you got to, like, the mid 2000s, you really couldn’t do business anymore without understanding the Internet and being able to send an email. We’re kind of at that point with where, you know, you don’t have to learn it, but my gosh, and I don’t think it’s going to be 10 years from now. I’m thinking 5 years from now based on how quickly everything is growing and ramping up.
We need to learn and understand how to integrate AI into your workflows. And I strongly suggest you start now and going through this three-part series was already a great start, but what’s coming next. In 2025 is my guess we’re going to be seeing three key things.
We’re going to be seeing marketing campaign optimization. We’re going to be seeing a lot of ways that you can be able to optimize your campaign. Assets do a lot of additional testing within your campaigns to be able to optimize performance results. There’s going to be a big jump in the ability to leverage AI-based analytics analysis and reporting.
[00:16:00] So is an analyst job still going to exist? Of course it is. But they’re going to have all of these new AI tools that are going to take them from looking at data for seven hours to looking at it for 20 minutes and having a much better understanding about what is going on. And the third is creative asset development.
So right now, a lot of the creative tools are kind of crappy within AI, right? Like they’re nothing I would put out on behalf of our brand. Besides basically, like, the creation or extension of using a basic template, but in 2025, I really think that’s going to change. I think there’s going to be a lot of tools that are able to take a basic brand design and then framework and then really be able to roll it out to a lot of different assets.
So it’ll be interesting. And what you have to do now is get started now. Make sure you understand all of the guardrails. Make sure you’re really well versed on it. And then, of course, make sure that your organization is bought in and you have all of the different legal things in line. [00:17:00]
So thank you again for tuning in to our three-part series. I really hope this was helpful. We have found again the more you actually dig in and use these platforms, the less scary they are. Just remember to keep privacy top of mind, and I’d also recommend hopping on a couple of LinkedIn groups around using in your daily workflows.
There’s lots of great prompting examples that are shared. It’s just a way to be able, if you’re a LinkedIn user, like, I am, it’s a great way to really incorporate it into a daily bite size learning by using or joining a tool or sorry, a platform like that.
There’s also tons of newsletters you can sign up for, but I think at this point, most of them are created by so they’re not awesome, but it will give you information at least about the platform advancements if you’re interested in that. But thank you again for tuning in to our three-part series and for today’s episode of We Are, Marketing Happy.
My name is Jenny Bristow, and I so appreciate you tuning in, reach out for any questions. [00:18:00] jenny@hedyandhopp.com. Have a fabulous day. Take care.
