Why You Can't Hire a Data Person Until You Define What They'd Actually Do

The Job Posting That's Too Vague

A company decides to hire a data person. "We need someone to handle our data," they say. They post a job: "Data Analyst/Engineer. Handle data tasks. Work with tools. Do analysis." They interview 10 candidates. Nobody's right. After 3 months, they hire someone. After 6 months, that person doesn't work out. "Hiring data people is hard," they conclude. Actually, they just didn't define the role clearly.

The Role Confusion

"Data person" could mean: Data Analyst — Answers business questions. Pulls reports. Creates dashboards. Data Engineer — Builds data infrastructure. Maintains pipelines. Writes code. Data Scientist — Builds predictive models. Machine learning. Statistics. Data Architect — Designs systems. Decides how data flows. Data Administrator — Manages access. Maintains infrastructure. BI Developer — Creates dashboards. Works with BI tools. These are completely different roles. Different skills. Different salaries. If you post a job for "data person," you're inviting confusion.

Why You Can't Hire Until You Define

Vague posting attracts the wrong people. You can't evaluate candidates if you don't know what you need. The person you hire fails because they spend time on things that aren't important. Team dysfunction—you hire an analyst but needed an engineer.

How to Know What You Actually Need

What problems are you trying to solve? (Be specific.) Can anyone on your team solve this, or do you need to hire? What does success look like? (By month 6, you'll have automated our monthly reporting.) What skills do you actually need? (SQL? Python/R? BI tools? Business understanding? Communication?) What's the day-to-day work? (50% analyzing, 30% building infrastructure, 20% reporting? The balance tells you what type of person you need.)

The Role Definition Template

Create a document: Title — Specific role name, not "Data Person." Purpose — What is this person here to do? Key Responsibilities — Pull data (SQL), create reports (BI tool), analyze segments, present findings. Skills Required — SQL (required), BI tools (required), business acumen (required), Python/R (nice to have). Success Metrics — By month 1: can pull basic reports. By month 3: created 5 dashboards. By month 6: done analysis on 3 key business questions. Compensation — $X-Y range. This clarity attracts the right people and repels the wrong ones.

The Role Options (Most Companies Need One of These)

Option 1: Analytics-Heavy Data Analyst — 70% analyzing, 20% reports, 10% infrastructure. Good for understanding your business. Cost: $80-120k/year.

Option 2: BI-Heavy Data Analyst — 50% dashboards, 30% analysis, 20% BI/infrastructure. Good for getting insights visible. Cost: $80-120k/year.

Option 3: Engineering-Heavy Data Engineer — 60% infrastructure, 30% pipelines, 10% analysis. Good for scalable systems. Cost: $120-160k/year.

Option 4: Data Scientist — 60% models, 20% analysis, 20% infrastructure. Good for prediction and ML. Cost: $150-220k/year.

Most small companies need Option 1 or 2. They don't need a data scientist yet.

The Onboarding (Once You Hire)

Week 1: Business understanding. What do we do? Who are our customers? What metrics matter? Week 2: Data understanding. Where does the data live? How is it structured? Week 3: First project. Something small that produces value quickly. Week 4: Feedback and course correction.

The Downloadable Resource

We've created a Data Role Definition & Hiring Guide that includes: A role assessment questionnaire; role definition templates for 4 common data roles; job description templates; interview questions (tailored to each role); skill assessment rubrics; onboarding checklist; 90-day success metrics template.

Download it here: aiforbusiness.net/resources/data-role-definition-hiring

What's Next

The next article, "The Documentation Debt That Catches Up During Growth," covers what happens when you don't document systems and processes.