Role Definition
| Field | Value |
|---|---|
| Job Title | Data Analyst |
| Seniority Level | Mid-Level |
| Primary Function | Queries databases (SQL), builds dashboards and reports (Tableau, Power BI, Looker), cleans and transforms data, performs basic statistical analysis, and communicates findings to stakeholders. Sits between business stakeholders who need answers and the data infrastructure that holds them. |
| What This Role Is NOT | Not a data scientist (doesn't build ML models or run experiments). Not a data engineer (doesn't build pipelines or data infrastructure). Not a business analyst (data-focused, not process-focused). Not a BI developer (more analysis, less tool development). |
| Typical Experience | 3-5 years. Bachelor's in business analytics, statistics, or related field. Common tools: SQL, Excel, Tableau/Power BI, basic Python/R. Optional certifications: Google Data Analytics, Tableau Desktop Specialist. |
Seniority note: Junior/entry-level data analysts doing basic reporting would score deeper Red. Senior/lead analysts who own analytics strategy, define KPIs, and drive business decisions would score Yellow (Urgent) — the stakeholder influence provides moderate protection.
Protective Principles + AI Growth Correlation
| Principle | Score (0-3) | Rationale |
|---|---|---|
| Embodied Physicality | 0 | Fully digital, desk-based. All work happens in SQL editors, BI tools, and spreadsheets. |
| Deep Interpersonal Connection | 1 | Some stakeholder communication — presenting findings, understanding business questions. But the core value is the analytical output, not the relationship itself. |
| Goal-Setting & Moral Judgment | 1 | Some interpretation of data and recommendations. But works within business questions defined by others rather than setting strategic direction. Follows analytical frameworks, doesn't create them. |
| Protective Total | 2/9 | |
| AI Growth Correlation | -2 | Strong Negative. Self-service BI tools (Power BI Copilot, Tableau AI, Gemini in Sheets) directly enable managers to query data themselves. Every AI adoption dollar makes it easier for non-analysts to self-serve — the exact function this role exists to provide. |
Quick screen result: Protective 2 + Correlation -2 — Almost certainly Red Zone.
Task Decomposition (Agentic AI Scoring)
| Task | Time % | Score (1-5) | Weighted | Aug/Disp | Rationale |
|---|---|---|---|---|---|
| SQL querying & data extraction | 25% | 5 | 1.25 | DISPLACEMENT | Natural language-to-SQL is production-ready across multiple platforms. Power BI Copilot, ChatGPT Advanced Data Analysis, and Gemini generate SQL from plain English questions. Managers ask "show me Q4 revenue by region" and get results without a data analyst. |
| Dashboard & report creation | 20% | 5 | 1.00 | DISPLACEMENT | Tableau AI auto-generates dashboards from data. Power BI Copilot creates visualisations from prompts. Looker AI suggests optimal chart types. ThoughtSpot provides search-driven analytics. The entire dashboard creation workflow is agent-executable. |
| Data cleaning & transformation | 15% | 4 | 0.60 | DISPLACEMENT | AI tools handle deduplication, format standardisation, missing value imputation, and outlier detection. ChatGPT Advanced Data Analysis executes end-to-end cleaning workflows. Domain-specific edge cases retain some human judgment, keeping at 4 not 5. |
| Ad-hoc analysis & investigation | 15% | 3 | 0.45 | AUGMENTATION | AI assists with pattern detection, anomaly flagging, and hypothesis generation. The human leads the investigation — asking follow-up questions, understanding business context, deciding what matters. Requires domain knowledge AI lacks. |
| Statistical analysis | 10% | 4 | 0.40 | DISPLACEMENT | Descriptive statistics, trend analysis, correlation analysis, basic regression — fully automatable. AI tools produce statistical summaries end-to-end. Mid-level analysts mostly do descriptive work; inferential statistics is rarer at this level. |
| Stakeholder communication & presentations | 10% | 2 | 0.20 | AUGMENTATION | Presenting findings, reading the room, knowing which insights resonate with which audience, navigating organisational politics. AI drafts slides and summaries — the human interprets, persuades, and builds trust. |
| Documentation & process improvement | 5% | 4 | 0.20 | DISPLACEMENT | AI generates data dictionaries, process documentation, and methodology notes. Human review needed but minimal editing required. |
| Total | 100% | 4.10 |
Task Resistance Score: 6.00 - 4.10 = 1.90/5.0
Displacement/Augmentation split: 75% displacement, 25% augmentation, 0% not involved.
Reinstatement check (Acemoglu): Limited. AI creates some new tasks — validating AI-generated queries, auditing dashboard accuracy, training business users on self-service tools. But these are lower-volume and lower-skill than the tasks being displaced. The "data analyst as AI output validator" is a real but narrow reinstatement path. Net headcount effect is strongly negative.
Evidence Score
| Dimension | Score (-2 to 2) | Evidence |
|---|---|---|
| Job Posting Trends | -1 | Generic "data analyst" postings contracting as self-service BI reduces demand for dedicated analysts. LinkedIn analysis: "Data analyst is not an entry-level job anymore." BLS projects 23% growth for operations research analysts (SOC 15-2031) but this is a different, more technical role. The specific "data analyst" title market is shrinking while adjacent titles absorb the work. |
| Company Actions | -1 | Companies restructuring analytics teams — smaller teams doing more with self-service BI tools. Not mass layoffs specifically citing AI, but headcount compression as managers self-serve through Copilot and Gemini. Data teams being reorganised toward fewer analysts with more strategic focus. |
| Wage Trends | 0 | Mid-level salaries $90,000-$110,000, stable. Tech sector midpoint $117,250 (Robert Half). 16.6% premium for analytics/BI certifications. Not declining in real terms but not growing above inflation. Premium shifting toward AI tool proficiency. |
| AI Tool Maturity | -2 | Production tools performing 80%+ of core tasks autonomously: Tableau AI (auto-dashboards, natural language queries), Power BI Copilot (prompt-to-report), ChatGPT Advanced Data Analysis (end-to-end analysis from upload), Google Gemini in Sheets (formula generation, data analysis), Looker AI, ThoughtSpot (search-driven analytics). These are mass-market tools — every knowledge worker has access. |
| Expert Consensus | -1 | Broad agreement that the "report builder" data analyst is being displaced. McKinsey: AI automates 60-70% of data collection and processing activities. YouTube data professionals: "bad data analysts are being replaced." LinkedIn experts: role shifting from tool-level skills to domain context and decision-making. Consensus: transformation with significant headcount compression. |
| Total | -5 |
Barrier Assessment
Reframed question: What prevents AI execution even when programmatically possible?
| Barrier | Score (0-2) | Rationale |
|---|---|---|
| Regulatory/Licensing | 0 | No licensing required. Optional certifications (Google, Tableau) are voluntary, not mandated. No regulatory barrier to AI performing data analysis. |
| Physical Presence | 0 | Fully remote/digital. An AI agent can execute every data analysis workflow from a cloud environment. |
| Union/Collective Bargaining | 0 | Tech/analytics sector, at-will employment. No union protection. |
| Liability/Accountability | 0 | Low stakes if analysis is wrong. No personal liability for incorrect dashboards or reports. Organisational consequences are mild — a bad chart doesn't trigger lawsuits or criminal liability. |
| Cultural/Ethical | 0 | Zero cultural resistance. Companies actively want AI to do this work. Self-service BI is a selling point, not a concern. Managers prefer instant AI answers over waiting for an analyst's queue. |
| Total | 0/10 |
AI Growth Correlation Check
Confirmed at -2 (Strong Negative). The data analyst role has a uniquely direct negative correlation with AI adoption. Self-service BI tools are not a side effect of AI advancement — they are the primary product. Power BI Copilot, Tableau AI, Gemini in Sheets, and ChatGPT Advanced Data Analysis exist specifically to let non-analysts do what data analysts do. Every enterprise AI deployment that includes Copilot licensing reduces the queue of requests going to the analytics team. This is not "AI might eventually affect this role" — it is "AI was specifically built to replace this function."
JobZone Composite Score (AIJRI)
| Input | Value |
|---|---|
| Task Resistance Score | 1.90/5.0 |
| Evidence Modifier | 1.0 + (-5 × 0.04) = 0.80 |
| Barrier Modifier | 1.0 + (0 × 0.02) = 1.00 |
| Growth Modifier | 1.0 + (-2 × 0.05) = 0.90 |
Raw: 1.90 × 0.80 × 1.00 × 0.90 = 1.3680
JobZone Score: (1.3680 - 0.54) / 7.93 × 100 = 10.4/100
Zone: RED (Green ≥48, Yellow 25-47, Red <25)
Sub-Label Determination
| Metric | Value |
|---|---|
| % of task time scoring 3+ | 90% |
| AI Growth Correlation | -2 |
| Sub-label | Red — Task Resistance 1.90 ≥ 1.8, does not meet all three Imminent conditions |
Assessor override: None — formula score accepted.
Assessor Commentary
Score vs Reality Check
The 10.4 places this deeper Red than Data Scientist (19.0) and well below Data Engineer (27.8). This is honest. The data analyst's core work — SQL queries, dashboard creation, basic statistical analysis — is more directly targeted by production AI tools than either adjacent role. Data scientists at least have experimental design and model building complexity; data engineers have architecture decisions. The data analyst's value proposition — "I translate data into reports for you" — is exactly what self-service BI was designed to eliminate. Zero barriers confirms there is nothing structural preventing displacement once technical capability arrives — and it has arrived.
What the Numbers Don't Capture
- The squeeze from both directions. From below: self-service BI tools let managers do their own reporting. From above: data scientists and analytics engineers absorb the complex analytical work. The mid-level data analyst occupies the exact space being compressed — too simple for a specialist, too complex for a tool two years ago, not complex enough for a tool today.
- Function-spending vs people-spending. Enterprise BI spending grows 15-20% annually — on Tableau licences, Power BI subscriptions, Copilot seats. The market for data insights grows; the human share of that market collapses. More dashboards than ever, fewer humans building them.
- Title rotation masking decline. "Data analyst" postings decline while "analytics engineer," "BI analyst," and "AI data analyst" grow — sometimes for overlapping work. Some of the posting decline is relabelling, not pure elimination. But the relabelling itself signals what the market values: the old title carries less weight.
- The BLS aggregate data problem. BLS projects 23% growth for "Operations Research Analysts" (SOC 15-2031), but this is a more technical role than the generic "data analyst." No BLS-specific SOC code exists for data analysts, so aggregate projections systematically overstate demand for this specific title.
Who Should Worry (and Who Shouldn't)
If your daily work is writing SQL queries, building dashboards, and generating weekly reports — you are in the direct path of self-service BI. Power BI Copilot, Tableau AI, and ChatGPT Advanced Data Analysis do exactly this, available to anyone with a licence. The analyst who is valued for "pulling data" or "building the Monday report" is competing against tools purpose-built to eliminate that queue. 1-3 year window.
If you understand the business deeply, shape the questions worth asking, and translate complex findings into decisions — you're safer than the Red label suggests. Domain expertise, business judgment, and stakeholder influence resist automation because they require context AI lacks.
The single biggest separator: whether stakeholders need you to get data, or need you to interpret what data means for the business. The "get me data" function is being automated. The "tell me what this means" function persists — but it's a smaller, more senior role.
What This Means
The role in 2028: The surviving data analyst is unrecognisable from the 2020 version. Less time writing SQL and building dashboards — those are self-served. More time defining metrics, designing analytical frameworks, validating AI-generated insights, and translating findings into business strategy. The job title may persist, but the headcount drops 40-60% as self-service BI matures. The analysts who remain are de facto analytics consultants, not report builders.
Survival strategy:
- Move from data retrieval to data interpretation. Stop being the person who pulls data and become the person who explains what data means for the business. Domain expertise and business judgment are the 25% that resists automation — invest there.
- Master the AI tools, don't compete with them. Become the person who configures self-service BI for the organisation, trains business users, validates AI outputs, and handles the complex analyses that Copilot cannot. Use AI to 10x your output, not as something to outrun.
- Specialise in a regulated domain or complex analytics. Healthcare analytics (HIPAA-governed), financial analytics (SOX compliance), or advanced statistical work creates specialisation moats. The generic "data analyst who can use Tableau" is commoditised; the healthcare data analyst who understands clinical workflows is not.
Where to look next. If you're considering a career shift, these Green Zone roles share transferable skills with data analysis:
- AI Auditor (AIJRI 64.5) — Quantitative analysis, data quality expertise, and model evaluation skills transfer directly to auditing AI systems for bias, accuracy, and compliance
- Data Protection Officer (AIJRI 50.7) — Data governance knowledge, analytical skills, and understanding of how organisations use data map to privacy and regulatory oversight
- AI Governance Lead (AIJRI 72.3) — Stakeholder communication, understanding of data systems, and analytical rigour provide a foundation for governing AI deployments
Browse all scored roles at jobzonerisk.com to find the right fit for your skills and interests.
Timeline: 2-4 years for significant headcount compression. Self-service BI tools are already in production at enterprise scale — the gap between "technically possible" and "organisationally adopted" is closing as Microsoft pushes Copilot across every Office 365 subscription.
