Job description template
A complete data analyst JD for Indian employers — with the analyst-versus-scientist scoping error that wrecks most of these hires, and the screening signals that separate a dashboard operator from an analyst.
More data analyst searches fail on scoping than on sourcing. The JD asks for SQL and dashboards in one paragraph and machine-learning model deployment in the next, sets a band somewhere between the two, and then the team spends eight weeks wondering why every shortlist is either overqualified and expensive or underqualified and cheap. An analyst turns questions into evidence a business owner can act on. That is a complete job. Write it as one, and the market responds.
Replace everything in [square brackets] before you post it.
Job title
Data Analyst
[Company name] operates [one sentence about the business]. This role sits in [team — e.g. Growth, Finance, Central Analytics] and supports [which business owners].
We are hiring a Data Analyst to turn business questions into answers our teams can act on. You will work with [function — e.g. sales, operations, product] to define the metrics that matter, build the queries and dashboards behind them, investigate why numbers move, and present findings to people who do not write SQL. This is an analysis role, not a modelling role: the output is a decision, not a model.
Experience: [x–y] years in an analytics or reporting role. Freshers with a genuine SQL portfolio may apply to [state whether you have a junior variant of this role].
Location and work model: [city, office locality] — [on-site / hybrid: n days a week / remote]. State whether the role must overlap with a specific business team's hours.
Reports to: [Analytics Lead / Head of Function]. Stakeholders: [name the two or three teams this analyst will actually serve].
Compensation: [band and fixed/variable structure, or a plain statement that it is benchmarked and discussed in the first call]. State the notice period you can accommodate.
To apply: send your resume and, if you have one, a dashboard or analysis you can talk through, to [email / apply link]. The process includes a short SQL exercise — [state whether it is live or take-home, and how long it takes].
The template is the easy half. These are the decisions that decide whether the posting works.
This is the adaptation that matters most, and it is the one most often skipped. A data analyst answers business questions with existing data. A data scientist builds models that make predictions in production. They share SQL and nothing else about the day job: different outputs, different stakeholders, different bands, different interview loops.
A JD that asks for dashboards, MIS and stakeholder management alongside model deployment and MLOps produces one of two failures. Either strong analysts self-select out because they read the modelling requirement as the real job, or you shortlist people who want to build models and hire them into a reporting role, where they leave inside a year. If you genuinely need both, hire two people or write the JD as an analyst role with a stated path into modelling — and mean it.
Analysts are hired for tools and retained or lost on stakeholders. An analyst supporting a single well-run product team does a different job from one supporting four functions with competing definitions of revenue. Say which it is. Candidates who have survived the second kind will recognise it and price themselves accordingly; candidates who have not will at least know what they are walking into.
Almost every serious analyst process includes a SQL test, and almost no JD mentions it. Announcing it does three useful things: it deters the fraction of applicants whose SQL exists only on the resume, it lets honest candidates prepare rather than fail on nerves, and it sets the expectation that this role is assessed on work rather than on interview presence. State the format and the time it takes.
Signals specific to data analyst applications — not generic screening advice.
Analysts leave — and decline — for reasons that are visible in the JD if you know what to look for.
SQL is on almost every analyst resume in the market, so it carries close to zero discriminating power in a keyword match — it will not separate your shortlist. The tokens that do discriminate are the specific ones, and they are the ones JDs most often leave out.
Other roles, hiring guides and recruiter tooling.
Stack depth over stack breadth. The JD that attracts engineers who ship and filters out keyword collectors.
Lifecycle ownership, not talent acquisition with a bigger title. Scope it by headcount supported.
Early attrition is a JD problem. Disclose field-vs-inside and lead source up front.
Six roles, each with its own screening guidance
The components a job offer letter usually carries, the difference between an offer letter and an appointment letter, and the omissions that cause offers to be declined or disputed later.
A working method for shortlisting at scale: build the scorecard first, run two passes, know which filters are real and which are proxies, and calibrate before you trust anyone's judgement
Score applications against your own job description
JD screening scores every application against your own job description and shows the requirement-level gaps behind the score.