From time to time, we like to have a conversation with a subject matter expert about how Marketing is evolving and the effects of this evolution on practitioners. In this interview we talk with one of Type Marketingx’s most prominent digital marketing gurus, Mike Bradbury, about some of the ramifications AI will have on marketing.
Type Marketing: Mike, let’s just jump into what might be the biggest concern of many people working in marketing and that is will AI eliminate my job? To add some context, a study completed this past summer by Ravn Research found that 64 percent of marketers are fearful that AI might replace their jobs over the next five years. Do you think their fear is warranted?
Mike Bradbury: I do think that some marketing positions will go because of AI. We’re seeing it happening somewhat already. Take for example the person who is responsible for short content pieces like social posts and emails. Maybe they write them or maybe they work with a writer to devise messaging. There are instances where Chat GPT is now being used to create these emails and posts and even short ads. In many instances, AI can improve the quality of the content and the speed at which it is completed. Some lower-level marketing positions, for example, the job of setting up online ads, positions at that level will likely be made obsolete by AI. With that said, I don’t think that AI is going to reduce the aggregate number of marketing jobs.
Type Marketing: If AI is not going to reduce the number of marketing jobs, how will AI change them?
Mike Bradbury: I do think job descriptions will change as will the way some marketing tasks are performed. Marketers will have to become more versatile and open to ways of working they might not have anticipated. For example, think about the offshoring of marketing content creation. Some firms have been offshoring the graphic design of their marketing materials for years – but they wouldn’t dream of offshoring the writing of content because of the language barriers. With advances in AI such as with Chat GPT, offshoring the writing of some short-form content to lower-cost, non-English speakers (along with guidance on what is to be written) might become an option, especially for smaller firms looking to increase their content while tightly controlling costs.
Marketers are going to have to be open to new ways of working and managing activities and they are going to have to know how to harness the power of AI. At Type we’re hiring people who understand how to use AI to lower costs for our clientele. Here’s a real-world illustration of what I mean. Many marketing agencies when engaged to create video content for a client will use Adobe Premier Pro. Premier is a very good product but it can be time-consuming to learn and use. Someone must pay the cost of that time and it’s usually the client. I’m now using an AI-enhanced video production tool called Descript. Descript applies AI to streamline and simplify video production. We can now create many more videos faster and at a lower cost. I don’t doubt that Adobe will also offer an AI-enabled tool, but my point is that if you are a marketing professional who doesn’t want to be displaced by AI then you should embrace the tools that enable you to increase your productivity. Even if your employer is not now pushing you to do so your clients (both internal and external) soon will.
Type Marketing: Which specific skills will marketers need to acquire as AI becomes more powerful and widely embraced?
Although the term has all the appeal of a wet blanket, prompt engineering is one skill I’d recommend acquiring. For those unfamiliar with it, prompt engineering is creating the text (the prompt) that is used by the AI model to perform a task. Many users of an AI platform like Chat GPT simply ask a direct question but using prompts is a much more powerful method to get more value from AI. Think of a prompt as something like a reusable template that can save you time and, when used effectively, provides more accurate, tailored, and detailed results. I created a prompt in Chat GPT that edits a video in a style that I’ve earlier defined for the system. I don’t have to “re-teach” the AI every time I want a video to fit a certain format. Using prompts has probably cut the time I spend editing videos by two-thirds. Think about how much more exposure an organization acquires if it can now create three videos for its social posts in the time it used to take to create one.
Using AI to create and edit videos and other graphics is going to be a very useful and highly demanded skill. I’ve been using the Midjourney bot through Discord for image creation and although it’s probably the most popular AI imaging tool, it has its shortcomings. What Adobe is now doing with Photoshop is intriguing. One of the long-term digs with Photoshop has been that the learning curve is steep. Adobe has been making flattening the curve a priority. They have made it easier to do things like moving people around in the frame and changing the clothing of the people appearing in the photo. Adobe has the resources to become a heavyweight player in this discipline. Getty Images could also become a major player in this field as they have access to so many images and are partnering with Nvidia to move into the generative AI space. This ability to quickly and extensively create and edit graphics is going to allow marketers to do more in less time and reduce the cost of that component of the marketing budget.
Type Marketing: It sounds like you think that AI will be creating new marketing roles even as it makes some existing roles obsolete. What types of marketing roles do you think AI will require?
Mike Bradbury: Yes, we’re already seeing some of this happening. Marketers are going to have to know how to effectively apply AI depending on their specialty. You can take the primary roles and preface all of them with AI – AI designer, AI writer, AI social media, etc. I also think that marketing generalists will play a role vis-à-vis the organization’s data. Until recently, it would take a data analytics specialist to use SQL or BigQuery, etc. to organize and categorize data. Some of these tasks can now be handled by AI. Recently I input over a thousand entries that included job titles and job orders into Chat GPT and had the platform categorize them. Had I wanted to, I could have had the system analyze this data by job title. Smaller organizations that don’t have a data analyst to manipulate or query data will be able to turn to AI for help. Additionally, Chat GPT can help analysts write SQL queries thereby improving their productivity.
Type Marketing: How should marketers get started with AI?
Mike Bradbury: Start using Chat GPT. I recommend paying the $20 a month for GPT Plus. You get a lot more value and capabilities for the paid version including priority access to upgrades, new features, and plugins; faster response times; and greater access to the system – which is important during peak hours. You can also upload pictures and graphics which means that you could take a diagram, and upload it, and the platform can then describe the important points contained in the diagram. If I wanted I could have it create a social post from the diagram and the description it generates. Start by having it draft emails for you or use it to conduct research. I do suggest when using it for research that you have it provide citations for the information it finds. I also suggest that you keep a repository of prompts and chats. Having a repository will eliminate your having to waste time every time you use the system. Don’t confine your use of AI to just Chat GPT – there are many other platforms like Google’s Bard or Microsoft’s Bing – and some are probably better at some tasks than others. It’s a quickly evolving field so it’s important to stay current about developments with AI platforms. Subscribe to a newsletter or two, attend a session focused on AI advancements – the more you know the more valuable you become.
If you track the progression of digital marketing moving from SEO and PPC to UX and UI – the people who developed skills in these fields early found that they could choose the trajectory of their careers. I think we’re at the same kind of inflection point now with AI. Those people who are most adept at understanding what the AI platforms can do for them and how to use them to generate value for their organization will have more control over their careers.
Type Marketing: Some recent surveys and studies indicate that many firms aren’t getting the value they anticipated from their data collection and analysis investments. Do you think AI can change that?
Mike Bradbury: From a digital marketing perspective, I think AI can help marketers more quickly get insights into why consumers behave in a certain way – or why they suddenly change their behaviors. When organizations develop algorithms to leverage AI for consumer insights using “bad” or the wrong data, the output won’t be very useful. To your point, a survey by the Mobile Marketing Association found that almost 4 out of 5 marketers had challenges with data validation and quality. Instead of relying too much on data that might be faulty or untimely, perhaps a focus on KPIs would be more fruitful.
If you are now measuring KPIs you can see when an indicator changes but understanding the cause of that change can be difficult. Using AI to dig into KPIs over time – in conjunction with whatever marketing programs were in place over the same period, can provide insights that can help guide my strategic thinking. What I believe is vital to understanding what is going on with my business is to speak to my customers and prospects – and then apply AI. For example, one technique that offers value is to take the transcript from a customer conversation and use an AI platform to analyze and summarize the important points. Once you’ve had enough of these conversations you begin to see trends and better understand the “why” behind the changes in your KPIs. These insights are more valuable and more useful than those I’m going to attain from ordered data that might tell me what happened but not why.
Type Marketing: How do you see organizations now most effectively using AI in their digital marketing initiatives?
Mike Bradbury: I see it used mostly for writing and video creation. For example, using Chat GPT to draft an email allows someone who might not be a strong writer to create and design a well-conceived email as long as the prompts used to create the email are appropriate. Using AI for writing pieces like emails also compresses the time required and eliminates some of the time spent in meetings for planning, review, and revisions. The output is also easy to edit. If the email comes out longer than desired, just have Chat GPT shorten it. Using an AI platform to help in routine writing assignments offers efficiency compared to how such work is often performed.
With this said, a word of caution about using any AI platform for writing and/or designing content, conducting research, or whatever is in order. These platforms are still works in progress and users must be aware of their limitations. They have been known to create inaccurate or off-the-wall sentences. Plagiarism is also a concern so users should check to make sure that their end product isn’t copied, and that attribution is given (there are several online plagiarism checkers). Some of what the AI platform might provide could be outdated or even inaccurate – make sure that the sources are credible. Keep in mind that the US Copyright Office has ruled that the output from AI systems like Chat GPT cannot be copyrighted. If the AI platform uses copyrighted material without permission and you use that material, you could find yourself in hot water for copyright infringement. Also, research or other material that resides behind a paywall or is “academic” research will likely be inaccessible to the AI platform.
AI platforms can be extremely useful writing assistants – but they have yet to be able to consistently replicate human creativity, empathy, humor, critical thinking, and the nuance that human writers provide when creating content.
The bottom line for those concerned about the effects of AI on their jobs is to learn as much as possible about AI in Marketing so that you are able to leverage these platforms to improve your productivity and increase your value.
