Write a Data Job Description That Attracts the Right People
Turn a vague hiring need into a specific job description that filters in the candidates you want and out the ones you don't.
When to use this
When you're hiring for a data role and the generic "must know Python and SQL" JD will get you 300 applicants — and not the right ones.
The prompt
You are a hiring manager who knows the difference between a job description that lists tools and one that describes work.
Role:
- **Title and seniority**: [...]
- **The team this person joins**: [size, mission, current data maturity]
- **The 3 most important problems** they'll work on in the first 6 months: [be specific]
- **The 3 stakeholders** they'll work with most: [roles, what they need]
- **Tools we use** (just so candidates know): [list]
Write a job description with these sections:
1. **The team and the problem** — what the team is, why this role exists right now, what's broken or unbuilt without it. 1–2 paragraphs.
2. **What you'll do** — bullet list of 5–8 concrete things, framed as activities not skills. "Partner with product to design experiment readouts" beats "Strong analytical mindset".
3. **What you'll have in 12 months** — what success looks like a year in. Outcome-shaped.
4. **Who we're looking for** — 3–5 must-haves stated as past evidence ("You've shipped X" not "Familiarity with X").
5. **Bonus** — 2–4 nice-to-haves, honestly framed.
6. **Compensation and logistics** — range, location, remote policy. If you withhold range, write a one-line honest reason.
7. **How to apply** — what to send. Make this short.
Banned: "rockstar", "ninja", "10x", "passionate", "fast-paced". Anything from this list is replaced with what those words really mean here.
What you'll get back
A JD that describes real work and outcomes, with past-evidence must-haves, honest compensation transparency, and a clear application instruction.
How this is structured in English
Notice the English patterns this prompt uses — they're worth borrowing for your own requests.
- Activities, not skills Pair-contrast for job descriptions. Activities are observable; skills are claims. Activities filter better.