How to Write a Cover Letter with ChatGPT (That Doesn't Sound Like ChatGPT)

How to Write a Cover Letter with ChatGPT (That Doesn't Sound Like ChatGPT)

You've done it. You opened ChatGPT, pasted a job description, typed "write me a cover letter," and got back something that started with: *"I am excited to apply for the role of Marketing Manager at Acme Corp."* Technically fine. Completely forgettable. [Recruiters read that sentence dozens of times a day](https://www.eweek.com/news/5-best-chatgpt-cover-letter-prompts/). ChatGPT can write a good cover letter. The problem isn't the tool. The problem is that generic input produces generic output. This post walks through the exact process to fix that: what to prepare before you open the chat window, three prompts you can copy and adapt, and what the output still won't show you.

What You Need Before You Write the First Prompt

Most people skip this step. They open ChatGPT, type a rough description of the job, and hope for the best. That's why the output sounds like it was written for any job, not this one.

Before you type a single prompt, get three things in front of you:

  1. Your CV or the relevant experience section. Not a summary of it. The actual text, with job titles, dates, and specific achievements.
  2. The full job description. Copy everything: responsibilities, requirements, the "nice to haves." The more you give ChatGPT to work with, the better it can match your language to theirs.
  3. One specific detail about the company or role. Something from the job post, the company website, or a recent announcement. A product launch, a stated value, a line from the CEO's last interview. One real detail is enough.

According to The Muse, including quantifiable achievements like "increased sales by 15%" and specific motivations in your prompt directly improves how authentic and relevant the output feels. Vague input gets you vague output. Specific input gets you something worth sending.


The Prompts: Step by Step

Prompt 1: The Foundation

This prompt does the heavy lifting. Paste your CV excerpt and the job description together, then ask for a structured first draft.

I'm applying for [Job Title] at [Company Name]. Here's the relevant section of my CV:

[Paste CV excerpt]

Here's the full job description:

[Paste job description]

Write a cover letter in 3 paragraphs, first person, no filler phrases. 
Don't open with "I am excited to apply." Focus on matching my experience 
to the role's requirements. Keep it under 300 words.

The Muse recommends including the job title, company name, and key skills explicitly in your prompt. That specificity is what separates a tailored draft from a template.

Run this once. Don't edit yet. Read it for structure and see what it got right. The next prompt handles tone.


Prompt 2: Tone Refinement

The first draft is usually too formal. It tends to produce phrases like "I am passionate about contributing to your team" or "I would welcome the opportunity to discuss my qualifications further." These phrases don't say anything.

Use this prompt to clean it up:

Revise this cover letter. Make the tone conversational but professional. 
Cut any filler phrases. Replace vague claims with specific ones based on 
the CV I gave you. Don't start any sentence with "I am passionate about" 
or "I am excited to." Return the revised version only.

[Paste the draft from Prompt 1]

Here's what that refinement looks like in practice:

Before: "I am passionate about contributing to your team and believe my background aligns well with your needs."

After: "My three years running paid social campaigns for B2B SaaS companies maps directly to what you're looking for in this role."

The second version says something. The first one doesn't. Raw ChatGPT output often needs this pass to stop sounding like a template and start sounding like a person.


Prompt 3: Personalization

This is where the letter starts to feel like it was written for this job specifically. Take the company detail you gathered before you started and work it in.

Revise the second paragraph of this cover letter to include the following 
detail about the company: [your specific detail, e.g., "They recently 
launched a self-serve product tier aimed at SMBs" or "Their careers page 
emphasizes async-first culture"]. Work it in naturally. Don't force it. 
The goal is to show I've done my research without it reading like flattery.

[Paste the revised draft from Prompt 2]

Teal recommends copying company culture language directly from the company's website into your prompt so the reference feels grounded, not invented. A detail that's verifiable and relevant lands better than a generic line about "shared values."

After Prompt 3, read the full letter out loud. If a sentence sounds like something you'd never actually say, rewrite it yourself. ChatGPT gets you 80% of the way there. The last 20% is yours.


What ChatGPT Won't Tell You

You now have a well-structured, reasonably personalized cover letter. Here's what it doesn't tell you.

ChatGPT doesn't know how well your CV actually matches the job. It'll write the letter from whatever you gave it, but it can't tell you whether you're a 40% fit or an 85% fit. It can't flag that the job requires 5 years of enterprise sales experience and you have 2. It won't tell you which skills the hiring manager is likely to weight most heavily.

That gap matters. A well-written cover letter for a role you're significantly underqualified for still doesn't get a callback. Some recruiters and ATS systems also flag AI-generated content as low-effort, which is another reason the human edit in Prompt 3 isn't optional.

This isn't a criticism of ChatGPT. Any writing tool that doesn't have visibility into your full application context has the same blind spot. The letter is only one part of the equation.

If you want to know your fit before you spend time on the letter, GotHired generates a match score, a gap analysis, and a cover letter from the same job paste. It tells you what you're missing before you start writing.


Putting It Together

The process is three steps: gather your inputs (CV excerpt, job description, one company detail), run the three prompts in sequence, then do a final read-through and edit anything that doesn't sound like you.

Do this for every application. Reusing the same letter with small edits is slower than it sounds, and the output shows. A letter written from the actual job description takes 10 minutes with this workflow and reads like it.

The prompts above are a starting point. Adjust the tone instruction in Prompt 2 to match the company's register. Add a challenge-action-result structure in Prompt 1 if you have a strong story to tell. Teal's research confirms that storytelling frameworks like challenge-action-result, when fed into the prompt with real examples, produce noticeably stronger drafts.

Start with the inputs. The prompts do the rest.


If you want to know how well you match the job before writing the letter, GotHired generates a match score, a cover letter, and a gap analysis from the same job paste. Free to start at gothired.ai.

Try GotHired's cover letter generator.

Ready to apply smarter?

Get an AI match score, cover letter, and interview prep — free for your first 5 jobs.

Get Started Free

Posts Relacionados