X
CLOSE
Subscribe for Free
Join over 5,000 doctors, providers, and staff by subscribing to im+, the largest digital educational resource for medical.
Thank you! Your submission has been received!
Oops! Something went wrong while submitting the form.

How Digital Marketing Agencies Should Be Using AI (AI Part 5)

Published on
June 20, 2024
|
Last Updated
|
00
minute read
Christian Shepherd
Christian Shepherd
Content Strategist
/ Founder

This article is one of a five-part series exploring the fundamentals and practical application of artificial intelligence for professionals in the medical aesthetics industry. We are not responsible for any hostile AI takeovers, Armageddon, human extinctions, or cheesy dystopian plots. We will, however, take full credit for any incredible improvements to your marketing program. That we are used to.

Learning what artificial intelligence is capable of and then learning how to implement it into a workflow is not an easy task. It takes hours, days, weeks, even months of dedicated time to become proficient in these tools. So, naturally, a lot of organizations are getting it wrong.

When you get into the weeds of AI-assisted content, there are very clear limitations, capabilities, and best practices that should guide anyone exploring the systems. 

So, whether you are a medical aesthetic professional looking to see if your agency is using the tech in the most efficient way they can or a marketer looking to optimize your workflow, Part 5 of our series is a straight-to-the-point guide on what agencies should and should not be doing with artificial intelligence. 

What Agencies Should Be Doing With AI

We’ve said it a few times throughout this series, but AI is not here to replace creative professionals. Maybe one day we might be in a world where artificial creativity trumps the human variety, but we aren’t there yet. So, what can AI be used for in a practical setting?

1. Ideating More Efficiently

Ideation, or the process behind finding out what a project will look and feel like, is often a difficult and time consuming task. You need an art and style direction on big marketing projects, and unfortunately, art and design departments at agencies are one of the most bogged down.

So, one of the ways you can help the process is to have an art model generate some general ideas and direction for them to work with so they can skip the “hey do you like this approach” phase and move straight into asset creation.

Here is an example of something I just threw at Incredible’s Chief Creative Officer, Zack. I was working on some imagery for an email marketing article that played on the theme of “the lost art of email marketing.” 

I thought it would be cool to have some old-school Indiana Jones-style art to go along with the article for social media and other promotion. So, I had Midjourney generate about 50 images of old school movie posters. These were the three I landed on:

indiana jones generated art work
indiana jones generated art work

As you can see, the art is great, but the words are literal nonsense. So, I threw it to Zack, our Chief Creative Officer, who told me it would be no problem to get the words adjusted as long as I didn’t hit him with some ridiculous title like this:

indiana jones generated art work example

Rather than asking the design team to create from scratch, I was able to produce a solid foundation to work from to help speed along the ideation process. This is hugely impactful, especially since these images can easily be used as the marketing theme for this entire campaign.

This is just one example — you can also just produce a library of visual content to scour through until you find a general style and direction that suits your needs. It’s a blitzkrieg kind of approach, but sometimes starting with that can save you trial-and-error time in the long run. 

2. Brainstorming More Productively

Brainstorming content ideas is perhaps the biggest area of advantage for AI-assisted workflows. Models like GPT-4 can generate hundreds of topic angles and ideas in a matter of seconds.

This isn’t anywhere near hyperbole, either. Here was a GPT-4 prompt: please generate 100 blog topics on dermal fillers.

And in about two minutes, here is my list:

  1. The Basics of Dermal Fillers: What You Need to Know
  2. The Different Types of Dermal Fillers and Their Uses
  3. How to Choose the Right Dermal Filler for You
  4. The Science Behind Dermal Fillers: How They Work
  5. Understanding the Risks and Benefits of Dermal Fillers
  6. A Comprehensive Guide to Hyaluronic Acid Fillers
  7. The Role of Dermal Fillers in Anti-Aging Treatments
  8. Common Myths about Dermal Fillers Debunked
  9. Preparing for Your First Dermal Filler Appointment
  10. Post-Treatment Care: Maximizing the Effects of Dermal Fillers
  11. Dermal Fillers vs. Botox: Which One is Right for You?
  12. Exploring the Cost of Dermal Fillers: Is it Worth the Investment?
  13. Dermal Fillers: Temporary Solution or Long-Term Investment?
  14. The Art and Science of Lip Augmentation with Dermal Fillers
  15. Non-Surgical Nose Jobs: The Power of Dermal Fillers
  16. How Dermal Fillers Can Enhance Your Cheekbones
  17. The Importance of Choosing a Qualified Provider for Dermal Fillers
  18. Natural-Looking Results: The Goal of Modern Dermal Fillers
  19. Dermal Fillers: The Secret to a Youthful, Radiant Glow
  20. Avoiding the Overdone Look: Tips for Subtle Use of Dermal Fillers
  21. Addressing Common Concerns About Dermal Fillers
  22. What to Expect During a Dermal Filler Procedure
  23. How Long Do Dermal Fillers Last? An In-Depth Look
  24. Using Dermal Fillers to Treat Under Eye Hollows
  25. The Role of Dermal Fillers in a Comprehensive Skin Care Regimen
  26. Can Men Benefit from Dermal Fillers? Exploring the Options
  27. Innovative Uses of Dermal Fillers in Cosmetic Medicine
  28. The Impact of Lifestyle on the Efficacy of Dermal Fillers
  29. Aging Gracefully: The Supportive Role of Dermal Fillers
  30. How to Talk to Your Doctor About Dermal Fillers
  31. Dermal Fillers: Transforming Smiles Without Surgery
  32. Side Effects of Dermal Fillers: What You Should Know
  33. Dermal Fillers: Combating Signs of Aging on the Hands and Neck
  34. The Safety Profile of Dermal Fillers: An Overview
  35. Understanding the Differences Between Filler Brands
  36. Navigating the World of Dermal Fillers: A Glossary of Key Terms
  37. Customizing Your Cosmetic Treatment: The Versatility of Dermal Fillers
  38. Case Study: The Impact of Dermal Fillers on Self-Esteem
  39. The Environmental Impact of Dermal Fillers: A Consideration
  40. How Dermal Fillers Have Changed the Face of Cosmetic Treatments
  41. The Relationship Between Dermal Fillers and Skin Elasticity
  42. The Psychological Benefits of Dermal Fillers
  43. The Role of Dermal Fillers in Facial Symmetry
  44. The Future of Dermal Fillers: What to Expect in the Next Decade
  45. Correcting Acne Scars with Dermal Fillers: A New Approach
  46. Achieving Natural Facial Contours with Dermal Fillers
  47. Reducing the Appearance of Marionette Lines with Dermal Fillers
  48. Dermal Fillers: A Non-Surgical Approach to Facial Rejuvenation
  49. Breaking Down the Procedure: A Step-by-Step Guide to Dermal Fillers
  50. Before and After: Real Stories of Dermal Filler Transformations
  51. The Psychology Behind the Growing Demand for Dermal Fillers
  52. Dermal Fillers: A Practical Approach to Aging Skin
  53. The Role of Dermal Fillers in Correcting Facial Asymmetry
  54. Combining Dermal Fillers with Other Treatments for Optimum Results
  55. How to Handle Potential Complications from Dermal Fillers
  56. Exploring the Latest Trends in Dermal Fillers
  57. Dermal Fillers: A Key Part of Holistic Facial Rejuvenation
  58. Celebrities and Dermal Fillers: The Influence on Society
  59. The Pros and Cons of Permanent vs. Temporary Dermal Fillers
  60. When Are Dermal Fillers Not Advised: A Comprehensive Guide
  61. How Technology is Changing the Application of Dermal Fillers
  62. Can Dermal Fillers Help with Migraines? An Investigation
  63. The Use of Dermal Fillers in Medical Conditions: Beyond Aesthetics
  64. Dermal Fillers in Your 20s, 30s, 40s and Beyond: A Timeline
  65. Can Dermal Fillers Be Reversed? An In-Depth Look
  66. The Journey to Recovery: What to Expect After Dermal Filler Treatment
  67. How to Make Your Dermal Fillers Last Longer
  68. The Power of Dermal Fillers for Non-Invasive Jawline Contouring
  69. The Growing Popularity of Dermal Fillers Among Young Adults
  70. Are Dermal Fillers Painful? Dispelling Fears and Misconceptions
  71. Biodegradable vs. Non-Biodegradable Dermal Fillers: The Pros and Cons
  72. How to Spot Quality Dermal Fillers: An Essential Guide
  73. The Role of Dermal Fillers in Treating Facial Scarring
  74. Dermal Fillers: Frequently Asked Questions Answered
  75. Hyaluronic Acid: The Magic Ingredient in Dermal Fillers
  76. Exploring the Psychological Impact of Dermal Fillers
  77. Minimizing Discomfort During a Dermal Filler Procedure: Tips and Tricks
  78. The Role of Dermal Fillers in 'Liquid Facelifts'
  79. A Closer Look at Poly-L-Lactic Acid Fillers
  80. Are Dermal Fillers Suitable for Everyone? Understanding the Limitations
  81. How to Avoid 'Filler Fatigue': A Comprehensive Guide
  82. Case Study: Correcting Lip Asymmetry with Dermal Fillers
  83. The Impact of Dermal Fillers on Skin Texture
  84. Combination Treatments: PRP Therapy and Dermal Fillers
  85. Understanding the Legal Aspects of Dermal Filler Treatments
  86. The Versatility of Dermal Fillers: Beyond Wrinkle Treatment
  87. Personal Experiences: A Dermal Filler Journey Diary
  88. Dermal Fillers for Chin Augmentation: A Less Invasive Approach
  89. The Pioneers of Dermal Fillers: A Look into the Past
  90. Handling Dermal Filler Missteps: What to Do If You're Unhappy with Your Results
  91. The Use of Dermal Fillers in Earlobe Rejuvenation
  92. Dispelling Fears: Addressing Needle Phobia in Dermal Filler Treatments
  93. Can Dermal Fillers Improve Skin Hydration? An In-Depth Analysis
  94. Nonsurgical Rhinoplasty with Dermal Fillers: Procedure and Results
  95. Dermal Fillers: What Makes a Good Candidate?
  96. Understanding the Role of Dermal Fillers in Tear Trough Treatment
  97. Addressing Nasolabial Folds with Dermal Fillers
  98. Is There a 'Best' Time for Dermal Filler Treatment?
  99. Celebrity Dermal Filler Fails: What We Can Learn
  100. A Look into the Future: The Next Generation of Dermal Fillers

But this is definitely not a very manageable list to work with. So, let’s ask GPT-4 to identify some potential themes: can you put the themes in a list

Here is its work:

  1. Understanding Dermal Fillers
  2. Procedure and Post-Treatment Care
  3. Dermal Fillers and Aesthetics
  4. Comparisons and Choices
  5. Risks, Safety, and Side Effects
  6. Impact and Benefits of Dermal Fillers
  7. Special Applications of Dermal Fillers
  8. Miscellaneous Topics
  9. Personal Experiences and Case Studies

Much more manageable. Now, it’s just a matter of choosing whatever theme I want our marketing campaign to cover. For now, let’s go with “Special Applications of Dermal Fillers.”

GPT-4 prompt: can you give me all of the topics in the Special Applications of Dermal Fillers theme?

  1. Using Dermal Fillers to Treat Under Eye Hollows
  2. Dermal Fillers: Combating Signs of Aging on the Hands and Neck
  3. Correcting Acne Scars with Dermal Fillers: A New Approach
  4. Reducing the Appearance of Marionette Lines with Dermal Fillers

Now we have an easy, concise list to work with in only a few minutes. You can also ask for topics in this theme if you want to dive deeper. Also, if you had a theme or topic in mind already, it would be even easier to prompt the language model for topics about those parameters specifically. 

Keep in mind as well that GPT-4 and other models do a great job at finding similarities and relationships between two different topics. So if you wanted to write a blog about how the development of an aesthetic product mimics that of the lightbulb, you could begin your research and outline process by asking them to find those relationships specifically.

3. Outlining More Robustly

If you ask any professional writer or content creator, outlining is one of the most important parts of the process. It gives you a focused line to work within and ensures that you are hitting all of your intended points without straying too far from the topic at hand. 

For medical aesthetic professionals specifically, I see all too often intense medical digressions that get into the specifics of science that very few lay people could understand. Outlining helps avoid that and hundreds of other pitfalls. 

But it is time consuming, especially when you weave in the research aspect. Luckily, language models are great at helping to expedite this process. Let’s take a prompt from our list of 100: A Closer Look at Poly-L-Lactic Acid Fillers. 

It’s a general topic, so we will have to do some work to fill out what this blog will actually cover. 

Right off the jump, here are four sections I would add:

  1. What Is Poly-L-Lactic Acid?
  2. Why Is Poly-L-Lactic Acid Used?
  3. Which Products Have Poly-L-Lactic Acid?
  4. What Are the Downsides to Poly-L-Lactic Acid?

This is with no research or help from AI. Let’s plug in the following prompt to ChatGPT and see what it comes up with:

Not only do we have a fully developed outline to perfect, if you are using the web plugin, we also have a list of sources to cross reference for accuracy — which you should absolutely be doing when getting AI assistance.

Ideally, this is an even better starting point. You will still need to go back and add to it, but as you develop more and more, you can continue to lean on AI for additional ideas and concepts. 

4. Supplementing Content With More Content

For a lot of agencies, this is going to be a sensitive topic for several reasons. To start, there may be some shady players in the arena who have completely outsourced their content teams in favor of relying solely on AI-generated content. Bad move #1.

Alternatively, there may be other agencies who are providing the same content scope in shorter amounts of time — this means less focus on persuasive writing, voice, and all of the other little intricacies that take writing from passable to incredible. Bad move #2. 

With a thorough understanding of AI potential and content production for the medical aesthetic industry, I can confidently say the only way to effectively, fairly, and ethically use AI right now is to supplement your content with… more content. 

Whether that means higher word counts of meaningful information, bolstered design, additional graphics and infographic assets, or more in-depth research and thought leadership, you should be seeing more value in your deliverables if AI is part of the production process. 

Quote from Elyse, Incredible Director of Content:

"Since implementing AI into Incredible's content production process, we have seen an increase of 20% more content on average for a medical aesthetic page, plus a surge in visual assets like infographics. With this approach, our content becomes more immersive, interactive and engaging, and we are able to craft a more distinctive and memorable brand identity for our clients.”

5. Automating Time-Consuming, Non-Creative Tasks

Regardless of what your job is, there are always those busy work tasks that need to be done in order for the overall process to be completed properly. Parsing through long blocks of information for highlights, summarizing, consolidating, listing, creating bullet points, generating themes and motifs — all of these tasks take a ton of time from a marketing or medical professional, and they are really just the groundwork for the more creative final asset.

But platforms like ChatGPT can expedite a lot of this for you. Summarization, specifically, is something that ChatGPT excels at. Here is the model at work summarizing a page of a medical paper on gynecomastia:

For this example, let’s focus on the “Evaluation” section. Here is the summary:

Now, this is by no means meant to replace the need to read the article or section, but if after reading it you need to pull out the highlights, ChatGPT can take care of that for you in seconds.

Or, maybe you are in a situation where you need to develop a list of potential treatment very quickly, so you turn to ChatGPT to generate this list for you in less than 10 seconds:

You really can think of language models as an assistant to help you sort through those time consuming tasks that are both necessary and tedious. You’ll still need to keep a close eye on the final product you put out, but these types of automations are what make this new tech so damn powerful. 

6. Repurposing Content for Different Platforms

Let’s get this out in the open from the start: repurposing does not mean plagiarizing or self-plagiarizing. Everything you put out should be unique and tailored for the platform it is living on. Now that we have that out of the way, let’s take a look at some examples of how AI can assist you in getting the most out of your content.

The workflow I have found is the most efficient for me is to work from the largest part of the project downward. Let’s say we are planning a marketing campaign around preventative Botox usage. That project will include quite a few components that need to be produced, but the main three look generally something like this: a series of articles, social media posts for four major platforms, and a two-email campaign.

In this case, I’d work on the articles first, using AI where applicable to help create informative and relevant content. This gives me the biggest pool of content to then run through AI.

Once those were finished, I would immediately turn to ChatGPT to get the written content for the social media posts. But I wouldn’t go from scratch — I would feed the entire article into ChatGPT (or at least the section I want to highlight on social media) and ask it to create some social media caption options. 

So, let’s give it a shot and see what happens:

And here is where you begin to see the strength and weakness of AI: it gives you three distinct options to choose from, but all of them are generally pretty subpar. You’ll need to spice them up a lot to bring them up to the right standard.

Regardless though, your jumping off point wasn’t a bank page, and now you can focus on collaborating with the AI generation to get a truly great caption for social. Rinse and repeat this process for all of your platforms and suddenly you’ve managed to greatly improve the mileage of your content efforts. 

7. Preparing for Search Engine Changes

One of the biggest priorities of any digital marketing agency right now should be preparing your site for the upcoming changes that come from SGE. (You’ll know all about this if you read Part 4 of this series.) But how exactly should that be happening? Here are a few things that you should be seeing from your agency.

Upscaled Keyword Research. Google’s AI is becoming remarkably good at figuring out the intent behind the search. That is, what information is the user actually looking for and what content will they spend the most time on? It simply isn’t enough moving forward to find queries and answer them directly. Your content will need to read between the lines and answer the questions your user isn’t asking. 

For example, “breast augmentation cost miami” is fairly straightforward. An SEO-optimized answer for this query in the past might have been something like, “A breast augmentation in Miami can cost anywhere between $3,000 and $8,000.” A simple answer to a simple query. Or maybe even something that could land you on one of these snippets:

But that user likely has other questions. What constitutes the $5,000 difference? What breast augmentation options might I need, and where do those fall on the price scale? How extensive is my condition considered? Is the entire payment due at once? Is financing available? If I want to go with fat transfer, is that more or less expensive than an implant?

These are some of the “behind the scenes” questions that Google is associating with users looking up cost. If you answer all of these questions, it is more likely that you will be picked up by an AI-generated answer and get linked at the top of the page.

In this example, you can see that Google is pulling a ton of information and summarizing it from the web:

Updated Titles and Meta Descriptions. This one is true whether you are optimizing for SGE or not, but having fresh and strategic titles and meta descriptions go a long way in enticing Google to prefer your content. If it starts off with a confusing title or if you can’t grab attention with meta descriptions (the little blurb of content that shows up next to a link on Google) you are putting yourself at a significant disadvantage.

Reworked “Snippet” Strategies. If the current structure holds through to final launch, then SGE blocks are going to be more valuable than Featured Snippets. This means it is very likely that the energy your agency was investing into winning snippets should be redirected to SGE. This is mostly going to land on your agency’s content team since they are the ones responsible for compiling the information in an SEO friendly way, but every level of your team should be aware of the direction. 

What Agencies Should Not Be Doing With AI

Unfortunately, because of the excitement that models like ChatGPT and Midjourney have generated, there are a lot of people out there who grossly overestimated the usability of modern AI tools. Content and social media departments have been reduced in size, likely to only find later down the road that the quality of the work being produced is no longer acceptable.

Couple that with the good, old fashioned bad actors, and it is possible that you will at some point be exposed to some unethical AI usage if you are surrounding yourself with the topic. So, to protect yourself from agencies who aren’t using this tech responsibly, let’s go over some hard “do not do” rules for agencies.

1. “Outsourcing” Workload 

Your agency should never, under any circumstance, downsize its content production staff in favor of AI-generated content. Without human oversight and expertise, you are missing out on professionals who not only know how to communicate persuasively and personally with your audience, but also know the medical industry well enough to find factual errors. 

You’ll find that a lot of agencies who are willing to outsource content production to AI were likely those agencies that were more likely to outsource to content mills, where the focus of the production is on word count and speed, not quality or persuasiveness. 

And this type of content is, straight up, a waste of time. You need to understand the nuance of medical aesthetic content creation and persuasive writing, as well as how AI can operate within those requirements. You cannot just have AI shotgun all of your content — this is basically a surefire way to tank your SEO ratings, traffic, and conversions. 

2. Expediting the Content Creation Process

Even if your agency held onto its content or social teams, it is possible they are still using AI in the “copy and paste” sense to get your projects completed quickly. Your agency should still be creating unique and hand-crafted content, and 100% of the content that is produced and used by AI needs to be manually reviewed for accuracy, tone, persuasiveness and overall strategy.

If it is being used for topic creation or some other kind of assistance, these places should always be starting points to improve on — never the final direction. Why is this so important?

It’s simple when you understand these two rules:

  • Rule 1: Artificial intelligence is not powerful enough yet to understand the intent and needs of your audience.
  • Rule 2: Your content should always be smart enough to address the intent and needs of your audience.

If an agency is able to recreate their product through AI, then the content you have been receiving is not good enough to compete at the top levels of search engines since, as you’ve heard a million times, Google is getting increasingly better at understanding the needs of its users and their search queries.

If you can’t get into the mind of your audience in your content, you don’t really have a fighting chance at SEO success.

3. Replacing Traditional Research 

AI algorithms are data-driven and can't yet mimic the intuition or creative problem-solving capabilities of a human researcher. Humans are capable of making sense of context, understanding societal implications, and applying ethical considerations that AI models currently can't fully grasp. Things that are hugely important when it comes to medical aesthetic content.

Plus, AI research is only as good as the data it's fed. Poor-quality data can lead to inaccurate or biased outcomes, and no model is currently strong enough yet to understand how to differentiate between good and bad data.

Ultimately, you are responsible for the content that is on your site, and you will be the one who has to clarify the misinformation that a patient reads on your site (and deal with the subsequent damage to your reputation). 

4. “Repurposing” Other People’s Content

It’s unfortunate that we even have to bring this up, but bad actors do exist, and with the current tools available, copying someone else’s work into an AI language model and asking for a rewrite takes seconds.

But there are several problems with this. Firstly, it's unethical. Secondly, the content rewrite will almost never be as good as the original, which had the benefit of human oversight and intuition. So it is unlikely that it will ever outperform the original. 

Thirdly, borrowing structures or approaches is nothing new in the content world — but the point is that something is always changed, usually added, in order to provide even more value to the person reading. Something that has just been repurposed without strategy and intention is lacking any meaningful usefulness to the audience.

Lastly, every time you put out a piece of content, it should be viewed as an opportunity to improve your brand standing. If you are consistently only putting out rehashes of previously created content, why would anyone follow or look to you over another medical professional who is answering way more cutting-edge topics?

Remember AI Is a Tool, Not an Employee

The theme of this final part of our AI series is that AI should not be a replacement for any part of your agency’s process. It should be an open discussion about how it is and is not used. This is especially true after you read this entire series and now have some insight into how you could be implementing AI in your own marketing program. 

Just remember: AI is a tool to use, not an employee you can count on churning out reliable work with no breaks, no pay, no benefits, and no PTO.

If you jumped into this part of the series without reading the other four, I highly recommend going back and taking a look at the other parts if AI is something you are interested in. There is a ton of useful content to get you up to speed on the conversation.

TL;DR Marketing agencies should leverage AI as a tool to enhance their content creation processes, not replace human creativity. AI can assist in generating ideation, brainstorming content ideas, outlining, and automating time-consuming tasks. It's important to use AI to supplement content with value-added elements and optimize for search engine changes. However, agencies should avoid outsourcing their workload solely to AI, expediting content creation without human oversight, replacing traditional research, repurposing others' content, and expecting AI to fully replace human expertise. AI should be seen as a collaborative aid that requires careful strategy and ethical use.