Communicators are already prompt engineers — and we shouldn’t underestimate how naturally our skills translate.
Will Hodges is U.S. cross-commercial communications leader at PwC.
We’ve given an old skill a shiny new label. Prompt engineering may sound like a term born from Silicon Valley, but communicators have been doing it for decades. We just called it something else — briefing, messaging, media training. Same fundamentals, new name.
My role at PwC has given me a front-row seat to how artificial intelligence is changing the way communicators work. I’ve helped our communications and corporate affairs teams explore and adopt new tools, and I’ve seen firsthand how they fit into our daily craft. That experience led me to realize that prompt engineering isn’t a new technical discipline at all — it’s what we already do, and do well.
Communicators are professional translators. We give clarity, context and direction so others can act with confidence. That’s exactly what large language models like ChatGPT and Gemini need, and these are the same ingredients that make prompts so effective. In other words, communicators are already prompt engineers.
We’re overcomplicating the concept. It sounds like a brand-new skill, but in practice, it’s just good communication. Comms pros already do this when we prep spokespeople, build story arcs and brief executives — we shape direction and focus so the message lands where it needs to, the same way an effective prompt guides an AI tool to a strong result.
So what does that look like in practice? It’s less about mastering prompts and more about applying the same communication habits that have always served us well.
Think like a media trainer, not a coder
When you media-train an executive, you start with a crisp headline: “What’s the one thing we want the audience to remember?” Prompting AI works the same way — start with intent. Here’s what that could look like:
You are a corporate communications writer. Draft a 400-word press release announcing our new sustainability initiative for business media. Highlight our 30%carbon reduction goal and include a CEO quote on accountability. Tone: confident, plainspoken.
Why it works: The model now understands purpose, audience and tone — just like a spokesperson who’s been properly briefed.
Context is the secret ingredient
You wouldn’t drop a spokesperson into a story without background. The same goes for AI. When you give it context — what’s happening, what’s not relevant, what success looks like — you shape better, safer output. Here’s an example:
You’re preparing employee talking points for a town hall introducing our new project management platform. Context: Rollout begins next month; teams will receive phased training; the old tool stays available during the transition. Focus: Simplicity, support resources and why the change helps collaboration. Avoid: Vendor jargon or feature lists. Format: Five bullets, one sentence each, ending with a link to training.
That extra context prevents confusion and protects tone. It’s the communicator’s experience turned into structure.
Iteration is the new editing
In media training, you don’t stop after one run-through. You refine phrasing, tone and pacing. Working with AI should feel the same. The first draft isn’t the finish line. Don’t accept the first AI output — ask follow-ups and iterate just like you would with a teammate:
Round 1: Summarize this three-page strategy memo for employees in 200 words.
Round 2: Make it more conversational and emphasize what’s new.
Round 3: Add a clear opening that explains why it matters, and finish with how to give feedback.
Each iteration brings you closer to clarity and alignment. The communicator’s instinct to refine, not settle, is what makes AI collaboration work.
Lead with empathy — the human still matters
Sometimes the win isn’t what you say but how it feels. Picture crafting a leadership note about a major systems change: You would check that it sounds authentic and considerate, not robotic or overly polished.
Write a 250-word message from a division leader to employees about an upcoming change. Audience: team members curious about new reporting lines but committed to the mission. Tone: appreciative, transparent, forward-looking. Focus on what stays the same — mission, customers and core values — and where to go with questions.
You’re using empathy as a data point — something AI can’t easily apply without your guidance.
Different tools, same truths
None of this is about mastering code or becoming a technologist. It’s about recognizing that the fundamentals haven’t changed. The same habits that have always driven strong communication — clarity, context and connection — are the ones that make AI useful. The tools changed; the craft didn’t. Success leaves clues.
So the next time you open an AI tool and stare at that blinking cursor, don’t think like an engineer. Think like a communicator. You already know how to do this — you’ve just been calling it something else.
The post Prompt engineering is just good communication appeared first on PR Daily.
