Prompted: Is Claude or ChatGPT Better for Integrated Marketing Teams? (Part 2: PR)

For part two of our ChatGPT vs. Claude series — featuring a prompt-for-prompt showdown* on key marketing tasks — we’re pitting them head-to-head in two essential PR challenges. Which tool will reign supreme when it comes to interview prep and pitch research?

(Also check out part one of the matchup, where we put both tools through the copywriting and content creation wringer.)

Test 1: Prepping for a Media Interview

At Tier One, it’s a top priority to properly prep our clients before they meet with a reporter. With background on who they’ll be talking to, what the conversation will be about, and potential questions they might be asked, they’re ready to take on the interview with confidence. 

Recently, we explored how AI tools can help you master an upcoming meeting. But who should you turn to — ChatGPT or Claude? 

For this test, we input the example prompt below into both chatbots, formatting our question from the perspective of the interviewee to simplify our ask for the AI. To fill in the details, we used a hypothetical Tech Brew interview with a Salesforce SME on the topic of agentic AI.

Example Prompt: I have an upcoming meeting with [reporter] of [outlet], discussing [topic]. [My company] offers [solution]. Provide a background on the reporter and outlet, links to any recent relevant pieces they’ve published, and messaging to highlight in the interview.

The Results: The results were surprisingly different — Claude took the initiative to generate an entire meeting brief with background, suggested messaging, and potential questions. ChatGPT responded with a quicker, less-detailed overview.

But one thing stood out in ChatGPT’s favor: Our prompt specifically asked for article links, and Claude seemed to ignore that ask.

ChatGPT:

Claude:

The Winner: Despite the link mishap, we’re handing this one to Claude. The full brief included a lot of relevant and important details, and we’d feel more prepared going into our interview with Claude as our guide than ChatGPT. We thought its messaging was higher-level and more closely aligned with both our hypothetical company and reporter. 

One point that particularly stood out was its inclusion of the insight below, which added some extra depth to its analysis:

Test 2: Digesting the Headlines

Staying on top of what’s making headlines is PR 101. We listen every day for new storylines, relevant reporters, emerging outlets, and evolving trends that relate to our clients’ areas of expertise. A creative human eye can still catch connections our AI companions can’t, but collaborating with our tech tools can help speed up the brainstorming process. 

To test which chatbot is a better PR coworker, we asked both ChatGPT and Claude to come up with potential storylines or ways to connect an expert with the writers of a few articles we flagged. We prompted it to identify relevant details first to help guide the response.

Example Prompt: Identify the reporter, beat, and outlet for each of the stories linked below, then summarize its messaging or main takeaways. Suggest ways to connect [SME of company], which offers [solutions], to reporters who could benefit from their expertise. 

The Results: We tested this using HubSpot CEO Yamin Rangan as an example and included an article from The New York Times. Both bots’ answers were almost identical in how they were structured: starting with article summaries, then suggesting multiple connection points and throwing in a few additional tips at the end. But interestingly, only ChatGPT could access the actual articles. That made its information a lot more accurate.

ChatGPT:

Claude: 

The Winner: For accuracy alone, ChatGPT takes this one. There isn’t much use in connecting an expert with a reporter that doesn’t exist! But we actually thought Claude’s suggestions were a bit more creative and relevant to our specified SME, similar to its response to our media brief question. It could be worth going the extra mile to identify the topic and reporter yourself, and then asking for recommendations.

Who Wins the PR Showdown?

We don’t want to call Claude the outright winner on this one, since the bot continues to struggle with accuracy (something we also noticed in our content competition). But its insights did feel more useful to our PR pros, especially when compared to ChatGPT’s more simple and straightforward answers. In the end, it might just come down to how you use it. We tested using single prompts and answers, but AI shines in its ability to continue the conversation. For instance, you could ask Claude in a follow up for links, or request potential interview questions from ChatGPT.

Learn more about AI’s capabilities by reading part one of the series, which dives into how Claude and ChatGPT fare against our copywriters’ content creation tasks.

*A quick technical note for this series: Both chatbots offer multiple models, which have different strengths and weaknesses. Our tests were carried out on ChatGPT 4o and Claude 3.7 Sonnet, both using paid plans. 

Prompted, our monthly blog series, brings you relevant updates and insights on how to strategically add AI-driven solutions into your marketing toolkit. Subscribe to receive updates in your inbox.