I didn’t start using ChatGPT for emails because I hate writing. I started using it because I was tired of spending 15 minutes trying to make a simple reply sound normal. A quick message to a coworker needs a different tone from a family group chat, and somewhere between “too blunt” and “weirdly overenthusiastic” is the version that actually sounds like me.

The results were mixed at first. Though technically fine, they often sounded just a little too much like a cold outreach from a marketing department, regardless of appropriateness. Every email had to start with “hope you are doing well” and often meandered into unnecessary small talk. Others sounded oddly enthusiastic for messages that were really just confirming coffee for Thursday afternoon.

Mapping tone

The answer was to draw a metaphorical map for ChatGPT, guiding it to sound more like me. Instead of “reply to this email casually/formally,” I found that adding a little more context dramatically reduced how much editing I needed.

I've perfected a simple template for most cases now. I ask ChatGPT to:

"Write a reply to this email" and paste the original message, then write:

One of the first times it really clicked for me was when I was planning a birthday dinner. The initial message included a long chain suggesting restaurants, dates, backup dates, and possible rain plans as though we were organizing an international summit. I needed to reply with a clear answer while also suggesting an earlier time because of early plans the next day.

Normally, I would spend several minutes trying not to sound either overly blunt or stiffly formal. I pasted the email into ChatGPT and said the goal was to confirm the plans, but suggest moving dinner earlier, and that the tone should be warm and relaxed.

The draft it produced hit those notes without any further editing. It got to the point quickly, acknowledged the original message naturally, and avoided AI's penchant for exaggerated friendliness.

Most importantly, it sounded like a real person trying to finalize dinner plans.

Speed emailing

The real benefit of the prompt structure is that it means I don't have to spend extra time going back and forth with ChatGPT to refine every message. ChatGPT is much better at understanding the social shape of the email once it knows the purpose behind the reply and the general feeling I want it to convey.

The “goal” section especially helps. Telling the AI what the response is actually trying to accomplish changes the results far more than simply asking for a polite email. A message declining an invitation needs a different rhythm than a message confirming plans. A quick update to another parent from school feels different from an email to somebody fixing your dishwasher.

The tone section shapes those answers to match your relationship with the recipient. “Casual but professional” works well for practical things. “Warm and brief” is useful for family messages. “Friendly but direct” has become surprisingly effective for service appointments and scheduling emails where you want clarity without sounding cold.

I still edit the replies before sending them. Usually, I change a sentence or trim a phrase so it sounds more natural to me. Even then, starting with a solid draft saves an enormous amount of time compared to opening a blank email and trying to figure out the exact wording from scratch.

ChatGPT hasn't eliminated my hand-written emails, but it does make the task less tedious. The AI speeds me through the repetitive parts of digital correspondence that tend to pile up during a busy week, but it doesn't mean I'm avoiding communicating with people. If anything, I can get in touch faster than ever.

Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews, and opinion in your feeds.

Eric Hal Schwartz is a freelance writer for TechRadar with more than 15 years of experience covering the intersection of the world and technology. For the last five years, he served as head writer for Voicebot.ai and was on the leading edge of reporting on generative AI and large language models. He's since become an expert on the products of generative AI models, such as OpenAI’s ChatGPT, Anthropic’s Claude, Google Gemini, and every other synthetic media tool. His experience runs the gamut of media, including print, digital, broadcast, and live events. Now, he's continuing to tell the stories people want and need to hear about the rapidly evolving AI space and its impact on their lives. Eric is based in New York City.