In case you missed it, AI was everywhere in 2023. It’s not going away anytime soon. There were more than 14,000 AI startups in the United States in March 2023. Hundreds of thousands of AI tools are available with almost any specialization.
AI can help with cybersecurity, scientific discoveries, communication, financial services, and healthcare. It’s also been widely adopted for marketing. Forbes states that 64% of businesses are adopting AI to improve production. One-fourth of companies are using AI to make up for labor shortages.
AI isn’t going away.
Marketers can either embrace AI or be out-maneuvered by those who utilize AI properly.
AI Marketing Statistics From HubSpot
We’re already purveyors of HubSpot, especially for our internal efforts and for creating inbound marketing campaigns for our clients.
In its 2024 State of Marketing Report, HubSpot, a world-class CRM platform, showed how AI has taken over marketing. In September 2023, HubSpot surveyed more than 1,400 global marketers from 14 countries across 23 industries.
Nearly two-thirds of marketers (64%) use AI and automation in their jobs, with 62% saying it’s essential for their jobs. However, 38% of those surveyed said they don’t use AI and will adopt it in 2024.
A large majority of marketers who used AI, 85%, said generative AI (i.e., content writing) improved their content. 84% said it improved efficiency when producing content. 77% said AI helped them craft more personalized content, one key to landing more conversions.
The point is this: AI is the new standard for marketers moving forward.
With so many affordable tools, the only obstacle to adopting AI is training your marketing team to use it.
What Is AI Good For?
AI is excellent at doing repetitive tasks over and over again. For example, it can quickly parse millions of data points and analyze them to help humans make decisions based on the data. AI can find disparate data points and make an analysis that would take humans or non-AI-powered tools much longer to determine.
Some AI tools can also create marketing campaigns, emails, blogs, and landing pages when prompted correctly. (More on this later.)
What Is AI Not Good For?
AI is unsuitable for decision-making, human expertise, determining if its sources are good or bad, and the accuracy of YMYL data (with a caveat).
For all of its time-saving tools, AI cannot be relied upon to replace human thinking in making decisions where human expertise is more valued. Nor should AI be counted on to be accurate for “your money, your life” information, such as how to write a will for your exact financial and family situation, what car you should buy, what stocks you should invest in, and the like.
These things require more nuances, such as a person’s background, that AI simply can’t know (we don’t live in the Matrix). However, AI can help you research YMYL topics, provided you see that the sources of information are trusted (maybe try not to find legal advice on Reddit).
The caveat we mentioned about YMYL data is that AI isn’t advanced enough yet. Accountancy Age predicts that 99% of accounting firms will adopt AI to help with audits by 2027. So, AI tools will continue to get better.
What Marketers Need to Know About AI Prompting
You’ve probably heard of ChatGPT from Open AI by now. This platform paved the way for AI adoption in 2023 and remains the gold standard. Chances are good that many marketing AI platforms integrate some form of ChatGPT into their platforms.
Google has Gemini, and it can integrate with enterprise-level apps, but it’s not (yet) nearly as integration-friendly as ChatGPT.
We’ll discuss AI prompting as if you’re using ChatGPT.
The key to AI prompting is to see every chat session as a blank slate. It already knows how to accomplish its given task based on how it was coded. It just needs to know what to look for.
That’s where prompting comes into play. Once the AI has its prompts in hand, it will save it for later use, so you don’t have to retrain it over and over again.
For example, we asked ChatGPT to create a pillar page outline on marketing. It gave us six chapters with several subtopics on marketing in general.
But then, in a separate chat, we asked ChatGPT to perform a string of requests that gave us better results. First, we asked the AI to define a pillar page. Second, we asked it to scan our website to determine our high-level services. Third, we asked it to create a buyer persona based on the target audience we would write for. Finally, we asked it to create a content outline for a pillar page for BigPxl’s target audience.
You can combine these four requests into one, but you might get more detailed information if you keep these requests separate and ask them in a sequence.
Rather than six chapters on generic marketing, ChatGPT gave us 10 chapters tailored to our specific services. We would then use this content outline to generate content for particular messages (landing page, ad groups, social media, emails, video).
The first chat, with a single prompt, took us 10 seconds, but the results could have been better. The second chat, with four prompts, took about a minute and gave us a much better response.
Think about how much time you would save when using AI tools for competitor research. You can ask ChatGPT to scan a competitor’s website, create a list of high-level keywords your competitor has, and then scan your website to get a list of keywords. Ask your AI tool to compare these lists of keywords to see where the gaps are, if any.
Here’s a simple way to think of “chats” versus “prompts” in ChatGPT. A chat is a book. A prompt is a chapter of that book. The prompts form a cohesive, larger story, while each chat adds to, changes, and supplements the main story. ChatGPT allows you to name (and save) each chat for future use if you need to tweak them later.
That’s the basics of AI prompting. We’ll dive into how AI prompting can boost your content creation.
How AI Prompting Can Boost Your Content Creation Process
Whether you’re writing blogs, social media, emails, or ads, AI prompting can reduce your time on content creation.
We’ll create a prompt for an email marketing campaign. First, ask ChatGPT in a single prompt what kind of campaign you want, the campaign goals, and what company you’re writing for.
We asked ChatGPT:
“Create a top-of-funnel nurture email series for BigPxl (link to the website: https://bigpxl.com/) to introduce BigPxl’s audience to its inbound marketing services. There will be ads that run to the inbound services page: https://bigpxl.com/services/inbound-marketing/. There is a lead magnet download entitled, 10 Ways Inbound Marketing Drives Fantastic Results, that BigPxl’s target audience will download on that landing page.”
You can download the PDF guide to ChatGPT to give it more information to work with.
Let’s say ChatGPT creates a six-part email series with different titles for each email. You can ask it in separate prompts (but in the same chat) to create verbiage for those emails. If the words it initially creates sound too technical, ask ChatGPT to rewrite the text using more conversational language.
How AI Prompting Improves SEO and Keyword Strategy
Just like you can create a content outline with ChatGPT, it can generate keyword ideas with AI prompting. You can ask the AI to scan a particular website to generate keywords or ask it to generate keywords about a high-level topic.
A prompt might be, “Create 30 keywords that will serve as content topics about coffee.” ChatGPT then creates an enumerated list you can copy and paste into a document to begin content ideation.
Then, you can drill down as deep as you want to on these 30 keywords to create pillar pages and cluster content that speaks to your target audience. You can take one of the topics about coffee, like ethically sourced coffee beans, and create a pillar page about that.
Then, each subtopic can be about the countries or groups with the most robust ethically sourced coffee programs to let your customers know what coffees to buy when they want to foster local growers in faraway places.
Pro tip: ChatGPT has limits on the amount of content it can create in a single prompt. We will discuss that later.
Enhancing Customer Engagement with AI-Powered Chatbots
Chatbots are one of the first AI tools marketers created. These simple tools follow prompts from the people typing in them to generate topics or questions to get to the heart of why someone is interacting with the chatbot in the first place.
It starts with the chatbot analyzing the person’s verbiage and then determining which of the pre-programmed responses it should choose to display. The beauty of an AI-based system is that it can remember similar language and responses from previous users to determine the best way forward. AI chatbots get better over time using machine learning from past interactions that were successes or failures.
Other Ways AI Prompting Works for Marketers
Content ideation and creation are two great ways to use AI for marketing. But the right tools can supercharge much more than content writing.
Streamline Marketing Workflows
The best marketing CRMs have automated workflows that can deliver personalized content to various audiences based on lead scoring, time-based automated tasks, or the actions an individual in your target audience takes. The goal of your workflows is to cover all stages of the marketing funnel to ensure no one gets left out.
Why use AI to streamline marketing workflows? Look at your current workflows. If you don’t have any workflows, read articles about how to create them. Workflows can be simple and enormously complex. Think of workflows as a tree. When done correctly, a single workflow can spawn several branches and grow into a gigantic tree of customer engagement.
You might see that workflow creation may not be as easy as you thought once you find the nuances of your customers. There can be several different steps and branches. For example, in the first email of our hypothetical nurture series on inbound marketing, there are three CTAs (download a guide, call, fill out a form) and four links to go to.
You should create seven workflows based on the actions people take on just that first email. These workflows will separate your target audience and score the leads based on what actions they take or how many actions they take with the first email.
So, seven workflows spawned from one email? Now, you can see the value of streamlining workflows using AI.
You can continue to refine your messaging based on the actions taken in those additional seven workflows. You see where this could go.
HubSpot has several ways to streamline marketing workflows using its tools.
- Improve lead scoring with intelligent classification tools
- Generate personalized content that resonates with individual customers
- Add relevant information to leads based on scanning their websites
You can prompt AI to generate a workflow that classifies a response, content based on a specific response, and what actions to take next. Based on HubSpot’s criteria, you can get as specific or general as you want.
Like any other workflow, you should test any new ones to ensure they function properly.
Analyze Marketing Data
Just like the thousands of AI tools, many data analysis tools out there can parse seemingly disparate data points to derive helpful insights for your marketing team.
You can prompt AI that’s integrated with your chosen data analysis tool to perform several tasks. Like with any prompt, the more information you feed it as it looks for data, the better it performs.
Increase quality data and reduce errors. Non-AI platforms can have errors and may create useless data. AI can find the correct data in the right places to better inform your marketing efforts.
Find information faster. Prompting AI to locate certain data points allows it to find the data faster. For example, you can ask an AI to discover all customers who bought $50 worth of sweaters in the past two weeks. Once you have this information, you can quickly use AI to generate an email series to nurture more purchases through a funnel you create for existing customers.
Making conclusions from seemingly disparate data. AI can find data points about your customers or potential customers that might not be found otherwise. For example, AI can determine consumer sentiment based on specific keywords found on social media feeds to determine how people feel about the brand. The AI can then make recommendations about which products or services might need improvement to increase positive sentiments.
Predictive analytics. The beauty of AI tools for analytics is that these tools can use historical data to make predictions. The more historical data you have on hand, the more accurate the conclusions are.
The AI platform you choose analyzes the data saved within the system. The more data it has access to, the more accurate its conclusions become.
If you don’t want to use an AI platform for data analysis, you can tap into ChatGPT to create spreadsheet formulas and spreadsheets themselves to organize the data in ways you need it. You could even ask ChatGPT to make conclusions based on the data you feed it from a spreadsheet. But ultimately, it’s up to humans to decide if the conclusion is valid.
Tools and Platforms for AI Prompting in Marketing
We’ve mentioned ChatGPT and HubSpot’s AI tools that work with its products. But there are plenty of others on the market, and each has a niche.
Gemini is Google’s free AI tool, and it works in similar ways to ChatGPT.
Some AI platforms for marketing are more robust than others. For content creation, many agencies go gaga over Jasper AI. It allows a lot of customizations, and you can create personas based on various customer types, which the system saves.
When you need to create content based on that persona, you can tell Jasper to create whatever content you need based on the information you input for each particular persona. But Jasper is pricey.
Other content-creation AI tools like Anyword use the same basic principle. Grammarly has an AI tool to create and edit snippets of content. The Hemingway App does the same thing as Grammarly.
Shutterstock and Adobe have AI tools for images. And there’s the DALL-E, the famous image-creation tool associated with Open AI’s ChatGPT.
AI tools are used to create ads for a myriad of platforms, from Google and YouTube to Facebook and LinkedIn. If a marketing channel exists, there is an AI tool to help you.
AI tools have various pricing models, depending on your needs. The lowest-level plan might work if you’re creating content five times a month. If you need 100 pieces of content a month, consider a higher pricing model for your chosen tools.
Overcoming Challenges When Implementing AI Prompting
Businesses face common challenges when it comes to implementing AI and prompting for AI.
The top one is that companies may misunderstand the limitations or use of AI. This is where choosing the right AI tool is vital. What data sources can the AI tool work with? Does the AI tool scan the Internet in real-time to find information? How accurate is the information? How quickly does the AI tool work? How do you know you’re prompting the AI tool the right way?
ChatGPT has limitations. For example, we prompted ChatGPT to write a 5,000-word pillar page. It stopped after 1,400 words. The workaround for this is to create sections of your pillar page at a time.
So, instead of asking it to write the entire pillar page on digital marketing, we asked it to create a 500-word explanation of digital marketing followed by 500 words about SEO. Even if it takes someone on your team 15 minutes to write 500 words versus 30 seconds in ChatGPT, you can see the time savings already.
However, there are no limits to the number of prompts you can create in a single chat or the number of chats if you have a pro plan with ChatGPT. If you have a chat that’s a mile long, simply use Ctrl-F (or Cmd-F) to find the right spot in a particular chat.
The trick here is to utilize the correct version of ChatGPT. While excellent, the free version has limits in speed and searching the web. The paid version is much better, so spend $20 a month to have your team use it.
Choosing the right tools. As we stated, there are many tools available with varying degrees of cost associated with them. Determine which AI tools meet your needs in terms of functionality and scalability while weighing costs and benefits.
Find ones that have free trials, encourage your team to use them, and get feedback from your staff. Schedule a demo to see your AI tools in action. Some AI tools have custom pricing for enterprise-level or unlimited use. Always remember to negotiate the prices to try to get the best deal.
Training your team to use AI prompting. Any AI tool is only as good as the information you feed it. The more details you give it, the better. Most AI tools save prompts for later use, so you don’t have to reinvent the wheel every time.
Once a prompt is saved, the AI tool doesn’t need to be retrained. For example, you create three separate prompts to create customer personas of Acme Brick (contractors, developers, retail stores).
Once those personas are created in an AI tool, your team can tweak them later based on the campaigns you want to make. An initial campaign to land new customers would look different than a re-engagement campaign to get repeat business. But the persona is the same. The messaging is just different.
You can add customer data to these personas to further refine your messaging based on how much money customers have spent. You could even determine how much of a discount repeat customers could get using AI. Just make sure the discount fits into your pricing model.
Using AI the right way. Artificial intelligence is not a panacea. Like a hammer, AI is a great tool to have. It still needs humans to vet the end product or make decisions. When using AI in your company, make it a collaborative effort.
Back to our Acme Brick scenario. If your marketing team determines that a 15% discount on bulk purchases over 10 tons for retailers who buy 50 tons in six months is the sweet spot for returning customers, make sure the marketing staff discusses these numbers with the sales and finance teams. Depending on the company’s cash flow requirements, a 10% discount might be too much or too little.
AI Insights with Human Thinking
BigPxl believes that AI is a great tool to use for your marketing efforts. But it’s just that—a tool. A hammer needs a human arm (and the expertise) to create a beautiful deck.
AI can accelerate your marketing efforts and make the laborious part of it easier. However, humans still need to review and approve the results.
You can ask AI to write a 2,000-word blog about the best way to make toast (Tom Wujec has some ideas). But a human still has to edit the copy, add images, and make it worthy to go onto a website.
Never, ever do what Sports Illustrated did.
Yes, we write articles based on keyword searches, competitor research, and what’s on other websites. And getting really good at making your content stand out with the GoogleBot Overlords is important.
But a robot has never purchased your product or service because robots don’t make money.
Humans are the ones you need to create content for.
The right tools, like AI, can help put the content together. The humans on our team are the ones who make sure the content resonates with target audiences, the marketing strategies are sound, and full-funnel campaigns bring more leads to our clients.
Accelerate Your Marketing with AI and BigPxl
BigPxl can help you build successful marketing campaigns from start to finish. Contact us or call (417) 799-2233 to start the conversation. We’re happy to help in any way we can!
