Prompt Engineering for Communicators: You've Been Doing It All Along

Apparently, everyone needs to learn prompt engineering now. Tech blogs are full of articles about this supposedly critical new skill. LinkedIn is overflowing with people adding it to their resumes. There's probably a certification course somewhere charging $2,000 to teach you the ancient art of asking a computer nicely.

But here's what nobody seems to be saying: if you work in communications, you already know how to do this. You've been doing it for years. They just gave it a name that makes it sound like you need a degree from MIT.

OLD SKILLS, NEW PACKAGING

What I've learned is that prompt engineering sounds technical, but at its core, it's just communication. Good communication. The kind PR professionals have been doing since before anyone invented the term "thought leadership."

Think about what we do every day. We brief executives before interviews. We craft talking points that actually make sense. We take complex information and turn it into something a human being can understand and use. That's literally what a prompt does for an AI model. You're giving it direction, context, and clarity so it can produce something useful.

The only difference? Your audience used to be a journalist or a CEO. Now, sometimes it's ChatGPT or Claude

STOP OVERTHINKING THE TECH PART

When you prep a spokesperson for a media interview, you don't just throw them in front of a camera and hope for the best. You start with the goal. What's the one message that absolutely has to land? What does success look like?

Working with AI is the same concept. You tell it what you need, who the audience is, and what the output should accomplish. Here's how that might look:

Draft a press release about our company's new sustainability initiative. The audience is business reporters who care about measurable outcomes. Include our 30% carbon reduction target. Keep the tone straightforward and confident. Make it 400 words.

That's not coding. That's briefing. You're setting expectations the same way you would with any other project.

CONTEXT MAKES EVERYTHING BETTER

You wouldn't send a spokesperson into an interview without background information. You'd tell them what's relevant, what to avoid, and what the journalist probably already knows. AI needs the same treatment.

When I work with teams on AI projects, the ones who get frustrated are usually the ones who skip context. They type a vague request, get a vague answer, and then decide AI doesn't work. But the problem isn't the technology. The problem is that they forgot to communicate.

Give the model context. Tell it what's happening, what matters, and what you're trying to accomplish. For example:

Prepare talking points for an employee town hall about our new project management platform. The rollout starts next month. Teams will get phased training. The old system stays available during the transition. Focus on how this makes collaboration easier. Skip the technical specs. Format as five simple bullets with a training link at the end.

See what happened there? You eliminated confusion before it started. You gave boundaries. You were specific about format. This is exactly what communicators do when they're good at their jobs.

REFINING IS PART OF THE PROCESS

In media training, the first run-through is never perfect. You listen, you adjust, you try again. Maybe the message isn't clear enough. Maybe the tone is off. So you refine it until it works. According to IBM's prompt engineering guide, this iterative approach is fundamental to getting quality outputs from language models.

AI works the same way. Your first prompt probably won't nail it. That's fine. Ask follow-up questions. Adjust the parameters. Make it more conversational or more formal. Add an example. Shorten it. Whatever you'd do with a human collaborator.

First attempt: Summarize this strategy memo for employees in 200 words.

Second attempt: Make it more conversational and emphasize what's actually changing.

Third attempt: Add a clear intro explaining why this matters and end with how people can give feedback.

Each round gets you closer to what you actually need. The instinct to refine instead of settle is what separates decent communicators from great ones. It works with people, and it works with AI.

THE HUMAN ELEMENT STILL MATTERS

Sometimes what makes communication effective isn't the information itself but how it feels. When you're writing a message from leadership about organizational change, you're thinking about tone. Does it sound authentic? Does it acknowledge that change is hard? Does it feel like a real person wrote it, or does it sound like corporate jargon threw up on a page?

You can use that same instinct with AI. Tell it how the message should feel:

Write a 250-word message from a division leader about upcoming reporting changes. The audience is team members who are curious but also committed to the work. Sound appreciative and transparent. Focus on what stays the same: our mission, our customers, our values. Include where people can ask questions.

You're using empathy as a filter. AI can't do that on its own, but you can guide it there. That's the communications expertise at work.

DIFFERENT TOOLS, SAME FUNDAMENTALS

Look, I'm not here to tell you AI is some kind of miracle. It's a tool. A useful one, but still just a tool. What makes it work isn't the algorithm or the training data. What makes it work is how you communicate with it.

The same skills that make you good at briefing executives or training spokespeople are the skills that make you good at prompting AI. Clarity. Context. Specificity. The ability to iterate. Understanding your audience and your goal.

Effective prompting is about understanding how to interface and communicate with language models, not about learning to code.

Nothing about this requires a computer science background. It requires communication skills. Which, if you work in PR or corporate communications, you already have.

YOU ALREADY KNOW THIS

Next time you open an AI tool and stare at that empty text box wondering what to type, stop treating it like you're about to write code. Think about it the way you'd think about any other communication task. What's the goal? Who's the audience? What context matters? What does good look like?

You've been doing this your whole career. The tool changed. The principles didn't.

So maybe stop signing up for expensive courses on prompt engineering and start recognizing that you're already qualified. You've just been calling it something else this whole time. Brief the AI the way you'd brief anyone else, refine until it's right, and move on with your day.

Turns out, good communication is good communication. Even when the person on the other end isn't actually a person.

Ready to put these skills to work and see how naturally your communications expertise translates to AI? Contact Winsome Marketing to explore how we can help your team confidently adopt AI tools without losing the human touch that makes great communications work.