Maximise your marketing budget with AI: How artificial intelligence can help you work smarter, not harder

A picture of a robot drawing a robot, drawn by a robot

Are you feeling overwhelmed with a never-ending to-do list and limited resources? You’re not alone! Many marketing teams are struggling with the same challenges. That’s why I’ve been exploring the potential of Artificial Intelligence (AI) to help us make the most of our time and get more done. How can AI assist your team in tackling your workload? Let’s find out together!

The challenge

Let’s take a look at a simplistic example. Your marketing team has been asked to deliver three concepts to illustrate the high levels of sugars that are found in breakfast cereals to be concept tested by a focus group.

It is roughly five days of work; Reading the insight, brainstorming ideas and the designers working up some concepts.

Can AI help support the initial process?

We’ll start by asking ChatGPT to generate some slogans. Within ten seconds it returned the following ideas:

1. “Sugar overload! Don’t pour on the pounds with your cereal bowl”
2. “Cereal killer: all that sugar sneaking in at breakfast time”
3. “Say goodbye to a sweet start: check those labels before pouring”
4. “More sugar than a donut? Think twice before digging into that cereal”
5. “Start your day off right, not hyper and wired from all that cereal sugar”
6. “A spoonful of sugar might make the medicine go down, but a bowl of sugary cereal can send your health spinning”
7. “Cereal: not as innocent as it seems with all that added sugar”

We’ll look at some ideas to help visualise number four first.

Cencept 1 : More sugar than a donut? Think twice before digging into that cereal

We’ll ask Midjourney to generate some concepts based on the following prompt:

/imagine small boy eating donut, sitting at the breakfast table,cartoon,simple

Within 60 seconds it generated the following:

Not a bad start, and while I’d asked for something simple, these are maybe a touch too simple, so we’ll tweak the prompt slightly:

/imagine small boy eating donut, sitting at the breakfast table, gravity falls

60 seconds later:

They work, but we’ll ask ChatGPT to generate a few more variations of the slogan before moving onto the next concept.

Here is what we got back:

1. “Is your cereal sweeter than a donut? Time to reconsider your breakfast choice”
2. “Don’t be fooled by the healthy-sounding name, that cereal could have more sugar than a donut”
3. “Sneaky sugar alert: is your cereal packing more of a sweet punch than a donut?”
4. “Donut or cereal: which has more sugar? You might be surprised”
5. “Cereal or donut: which is the healthier choice? Don’t let added sugar trick you”
6. “Think twice before diving into that bowl of sugary cereal, it could have more sugar than a donut”
7. “More sugar than a donut? It might be time to switch to a healthier cereal option”

Concept 2 : Say goodbye to a sweet start: check those labels before pouring

The first thing that popped into my head was a bowl of sugar, so we’ll ask Midjourney to generate some concepts:

/imagine small boy eating large sugar cubes from cereal bowl, breakfast, comic style

60 seconds later:

The first image is closer to what I was thinking but I’ll ask for a few variations of it to tighten it up, 60 seconds later:

Again, we’ll ask ChatGPT to generate a few more slogans before looking at a third concept.

1. “Before you pour, check the label to avoid a sugar-filled start to your day”
2. “Sugar lurking in your cereal? Check the label before pouring”
3. “Start your day off right, not with a bowl of sugary cereal. Check those labels before pouring”
4. “Don’t let hidden sugar ruin your morning. Check the label before pouring your cereal”
5. “A sweet start to the day can quickly turn sour if you don’t check those cereal labels before pouring”
6. “Be a label detective: check for hidden sugar before pouring your cereal”
7. “Start your day on the right foot, not with a sugar rush from your cereal. Check those labels before pouring”

Concept 3: Cereal: not as innocent as it seems with all that added sugar

The first thing that popped into my head was a menacing-looking box of cereal, so we’ll see what Midjourney generates for us:

/imagine an evil cereal box, comic style

Not quite what I was thinking so we’ll tweak the prompt:

/imagine a cereal box with a evil sugar cube on it, comic style

Again back to ChatGPT to generate a few more slogans:

1. “Don’t be fooled by the innocent-looking cereal box. Check for hidden sugar before you pour”
2. “Cereal: not as pure as it seems with all that added sweetness”
3. “Beware of the sneaky sugar in your cereal”
4. “Cereal: it might look harmless, but all that added sugar could be harmful to your health”
5. “Sugar hiding in your cereal? It’s not as innocent as it seems”
6. “Cereal: a not-so-sweet surprise when you see how much sugar is added”
7. “Innocent-looking cereal or sugar trap? Check the label before pouring”

Conclusions

Have you ever been in a creative rut, trying to come up with catchy slogans and visuals for your marketing campaigns? Fear not! With just 30 minutes of brainstorming, you can generate a ton of ideas to get the ball rolling. While AI may be able to assist with some tasks, it can’t fully replace the human touch when it comes to creating unique content and forming personal connections. Trust in your own creativity and let AI be your helpful sidekick, rather than a replacement.

About Midjourney

Midjourney is an independent research lab dedicated to advancing the creative capacities of people by examining new thought mediums.

Midjourney’s AI bot creates images based on textual descriptions similar to Open AI’s DALL-E.

Midjourney’s team is led by David Holz (Founder Leap Motion, Researcher at NASA, Max Planck) and 11 full-time staff and with advisors from Apple, Github, etc.,

You can use Midjourney’s bot from the Discord server (https://discord.gg/midjourney) as well as on any other Discord server where it has been set up.

About ChatGPT

ChatGPT is a large language model chatbot developed by OpenAI based on GPT-3.5. It has a remarkable ability to interact in conversational dialogue form and provide responses that can appear surprisingly human.

Large language models perform the task of predicting the next word in a series of words.

Reinforcement Learning with Human Feedback (RLHF) is an additional layer of training that uses human feedback to help ChatGPT learn the ability to follow directions and generate responses that are satisfactory to humans.