UC5: Job Description Generation & Optimisation

The Problem

Hiring managers write job descriptions with inconsistent language, vague requirements, and sometimes unintentionally biased terminology (e.g., "ninja coder," "aggressive go-getter" may appeal disproportionately to certain demographics). Legal and HR teams spend 3 to 5 hours reviewing JDs to catch compliance issues (missing ADA accommodation notices, wage transparency gaps, discriminatory language). Poorly written JDs attract unqualified candidates and repel qualified diverse candidates. Many organisations lack standardised JD templates and compensation benchmarking, leading to internal pay equity issues.

What the Agent Does

Receives job requisition details from hiring manager (role title, department, reporting line, key responsibilities, required skills, experience level, location, compensation range)
Generates initial JD draft using curated template library (by role family,engineer, finance, sales, etc.) with standard sections (job summary, key responsibilities, required qualifications, nice-to-have skills, benefits overview, EEO statement, ADA accommodation notice)
Compares proposed compensation range against market benchmarking data (Radford, Mercer, PayScale, Glassdoor) and flags if range is significantly outside market norms (e.g., 30% below market), with recommendations for adjustment
Scans JD text for biased language (e.g., gendered pronouns, age-coded terms like "digital native," personality traits that may disadvantage neurodiverse candidates) and suggests inclusive alternatives
Checks for legal compliance: confirms ADA accommodation notice is present, verifies wage transparency compliance (Colorado, California, New York require wage ranges in some contexts), ensures EEO/OFCCP statements are included
Routes final JD for approval (hiring manager review → HR legal check → recruiter sign-off) with edit tracking and comment resolution

Data Requirements

Data Sources:

Role requisition form (internal form or ATS) , role title, department, responsibilities, required skills, compensation budget
Compensation survey data (Radford, Mercer, PayScale, Glassdoor, internal salary bands) , market rates by role, location, experience level
JD template library (internal Confluence/Notion or ATS JD bank) , approved JD templates by role family
Bias language dictionary (curated list of age-coded, gendered, neurodiversity-unfriendly terms) , mapping to inclusive alternatives
Legal compliance checklist (EEO, ADA, wage transparency requirements by state/country) , jurisdiction-specific requirements

Data Classification:

Public: Approved JD templates, role titles, general competency frameworks
Internal Structured: Compensation survey data (market rates), salary band ranges, JD approval workflow
Internal Unstructured: Role requisition details (hiring manager narrative), previous similar JDs
External Third-Party: Market compensation survey data (Radford, Mercer, PayScale) if purchased
Sensitive/Regulated: Approved compensation ranges (internal equity data), legal compliance policies

Data Quality Requirements:

Compensation survey data must be current (updated annually or quarterly for volatile markets)
JD template library must be reviewed and approved by HR/Legal annually
Bias language dictionary must be regularly updated (informed by EEOC guidance, academic research on bias)
Role requisition form must capture essential details (department, location, responsibilities, budget)
Compliance checklist must be jurisdiction-specific (US federal, state-by-state, EU GDPR, etc.)

Integration Complexity: Medium , Requires access to role requisition form (ATS or internal form), compensation survey system (likely via API or exported data), and JD template library (Confluence, Notion, or ATS storage). Most of this data is static or slow-moving. Integration is 2 to 3 weeks.

Score Breakdown

Criterion	Weight	Score (1-5)	Weighted
Time Recaptured	15%	3	0.45
Error Reduction	10%	4	0.40
Cost Avoidance	10%	3	0.30
Strategic Leverage	5%	4	0.20
Data Availability	15%	4	0.60
Process Clarity	15%	4	0.60
Ease of Implementation	10%	4	0.40
Fallback Available	10%	5	0.50
Audience (Int/Ext)	10%	5	0.50
Composite	100%		3.85

Why It Scores Well

JD generation is on-demand knowledge work: hiring managers request new JDs frequently (every hire), and the process is repeatable but currently manual. The agent improves compliance (legal review built-in), reduces bias (automated language scan), and ensures market-competitive compensation (benchmarking integrated). The fallback is simple (HR can always edit JD after generation). Data availability is good (requisition details, templates, compensation data are accessible). Score reflects medium impact (JD writing is not a massive time drain, but consistency and compliance gains are valuable).

Regulatory Alignment

EEOC Equal Employment Law , JD language must not discriminate on protected characteristics; agent must flag potentially age-coded, gendered, or otherwise biased language
ADA (Americans with Disabilities Act) , JD must include notice that accommodations are available; agent must verify this statement is present
State Pay Transparency Laws , Colorado Equal Pay for Equal Work Act, California SB 1041, New York Fair Pay Act all require wage ranges in job postings; agent must confirm ranges are included if jurisdiction requires
FCRA/Background Check Compliance , If JD mentions background checks, must clearly disclose; agent should flag if JD requirements (e.g., "must have clean driving record") exceed what is actually needed for the role

Sprint Factory Fit

Sprint 0 (2 weeks) + 2 build sprints (4 weeks)

Scores 3.85. This is a solid use case but ranks below higher-impact recruiting automations because: (1) time savings are moderate (3 to 5 hours per JD = 15 to 30 hours per quarter for typical organisations), (2) most hiring managers don't write that many new JDs (many reuse old JDs), and (3) the fallback is always a human editor. However, the compliance angle (bias detection, wage transparency, legal review) is valuable and reduces HR legal risk.

Comparable Implementations

Findem: JD optimisation , market benchmarking and bias detection in job descriptions
Radford/Mercer: Compensation benchmarking integration , market data for JD salary ranges
LinkedIn: JD best practices , AI-assisted JD writing (limited benchmarking in-platform)

Deploy This Use Case with the Sprint Factory

From zero to a governed, production agent in 6 weeks.

Sprint Factory Schedule a Briefing

Related Use Cases

Workflow Talent Acquisition

Interview Scheduling & Coordination

Score: 4.3/5.0

Batch Talent Acquisition

Resume Screening & Candidate Shortlisting

Score: 4.25/5.0

Workflow Talent Acquisition

Offer Letter Generation & Approval

Score: 4.1/5.0