5 tips for marketers starting out with GenAI

I’ve been with Jellyfish for over three years, contributing to our journey in embracing AI innovation. The rapid evolution of GenAI tools means our role - transforming these technologies into practical solutions for global brands - is as much about envisioning possibilities as it is about technical execution. With each new AI tool or workflow we integrate, we face two key challenges: first, understanding how it fits into our processes, and second, tailoring it to meet the unique needs of our clients.

In September, I was lucky enough to be invited to One Young World on behalf of The Brandtech Group. This convention promises to connect the world’s best young talent and industry over the most important issues this generation faces. This year, one of the topics was GenAI. This sparked significant existential debates about AI's role in society and its broader implications.

But talking to the leaders, philanthropists, founders, and technologists who visited the Brandtech booth, the conversation was less grandiose. These young leaders wanted to know: “How can I use it?” That question is much less complex. But seeing the outcomes: real, tangible benefits to the people doing good in this world, made me want to share some of those thoughts with everyone.

So, from the expertise of one of the world’s leading AI marketing groups (I have to say that, but it’s also true), here are some tips for getting started with AI.

1. Use the Tools Yourself

This is the first rule of GenAI for marketing: spend time with the tools before you make a plan. These tools are not like traditional tools. They don’t have complicated UI, and they don’t require much learning to get started. Plus, they’re everywhere now, and most are free of charge when starting. Even an hour messing around with ChatGPT or MidJourney will show you what they’re good at - and where they stumble. And no third-party description can beat the intuition you’ll gain by logging on and playing. You’ll quickly see that AI works well for certain tasks, like generating ideas or drafting content, but can be weak on the finer details, especially on things that require emotion or subtlety.

Getting hands-on with these tools makes you more realistic about what AI can and can’t do. You stop expecting it to write your next award-winning campaign (yet), and start using it to speed up the parts that don’t need your personal touch.

2. Let the Tools Teach You

It might be a bit unconventional, but isn’t the whole point of near-sentience to engage in a conversation? Challenge AI to teach you how to use it more effectively. You can ask ChatGPT to write better prompts, such as “How should I ask you to generate a social media post for X audience?” Let the AI break down its processes.

Ask it how it would approach a brief, or how it would rephrase something more effectively. Sure, AI isn’t a master of creativity, but it does know how it works. Leverage that. It can help you refine your prompts so you get better results faster.

3. Get Scientific

Beyond being such a rapid creative tool, the real calling card of AI is that the cost per output is low. That means you can generate a hundred variations of an ad copy, headline, or product description for the price of one. So you don’t need to hit the bullseye on your first shot.

Instead, adopt a scientific approach. Have clear variables you wish to test in your prompts. Tweak your inputs, see what works, then go back and refine. The beauty here is that AI doesn’t get tired. It’ll keep iterating as you keep asking. Test different tones, styles, and formats. Let data guide you to what works best until you develop a gut feeling about what works.

4. Plan for AI From the Start

AI is a new medium, not a direct replacement for any other, so it works best when you factor in its limitations from the very beginning. Don’t try to fix an AI-generated campaign at the end of the process. Start your project knowing where AI will struggle - like subtlety, creativity, or context -and build around that. Use AI to do the heavy lifting where it excels, and keep human creatives in the loop for quality control.

Creative solutions are needed for AI’s blind spots. If you plan for them upfront, you’ll save yourself the frustration of trying to force AI into doing something it’s not designed for. The more you respect its limitations, the better the outcome.

5. AI Shines in Long-Tail Marketing

AI’s low production costs mean you can keep campaigns going far longer than traditional methods allow. With AI, delivery doesn’t have to be the finish line. Once you launch a campaign, you can iterate, optimize, and extend your reach.

Need to tweak a product description based on new feedback? Easy. Want to generate variations of a blog post to hit different SEO keywords? Done. AI lets you keep refining and developing long after launch, which opens up tons of long-tail marketing opportunities.

And a gentle plug here - if you want the best platform for AI advertising, check out Pencil.

AI isn’t magic. It’s a tool. One that’s brilliant at volume, cost, iteration, and speed. But at times it’s our job as marketers to bring creativity and nuance. The trick is to know how to use it: get your hands dirty, ask AI to show you how it works, and plan for its flaws from the start. When you do that, you can tap into its power to scale your marketing and extend your campaigns long after delivery.