What technical skills should a data analyst put on a 2026 resume?

Lead with advanced SQL (window functions, recursive CTEs, performance tuning) on a named warehouse (Snowflake, BigQuery, Databricks, Redshift), dbt for transformation, a BI tool (Looker / LookML, Tableau, Power BI, Hex, Mode, Sigma), Python or R for analysis (pandas, scikit-learn, statsmodels, tidyverse), and an experimentation or activation tool (Statsig / Eppo / GrowthBook / Hightouch / Census). For Senior+ roles, add semantic-layer authorship, data observability (Monte Carlo / Bigeye / Datafold) and AI tooling (Cursor / Hex Magic / text-to-SQL workflows). Recruiters use boolean search — list the named tool, not the category.

What soft skills should a data analyst include?

Stakeholder requirements gathering (the question framing skill), executive readout writing (1-page memos, 3-slide decks, async written communication), data SLA negotiation, hypothesis-vs-analysis distinction, and conflict resolution when stakeholders disagree with a read. At Senior+ add mentoring (juniors taken from first model to promotion), hiring loops, analyst capacity planning, and OKR ownership. Avoid soft-skill clichés ('detail-oriented', 'team player') — replace with evidence ('mentored 3 juniors to Analyst-II promotion in 9 months').

How do I show statistical literacy on a data analyst CV?

Quote the method by name (regression, A/B test, difference-in-differences, propensity-score matching, CausalImpact, survival analysis with lifelines, Prophet for time-series), name the library (statsmodels, scikit-learn, DoWhy, EconML, PyMC, lifelines), and always quote the sample size and confidence interval. 'Lifted activation 8pp at 95% CI, n=82,000 vs control' reads as Senior; 'lifted activation 8pp' alone reads as channel-specialist. For causal claims (especially observational ones), name the design (geo-holdout, switchback, instrumental variable, synthetic control).

How important are dashboard adoption metrics on an analyst resume?

Critical at Mid+. Replace 'built X dashboards' with weekly active dashboard users, hours saved per stakeholder, and time-to-decision. 'Built 12 dashboards' is weak; 'Built 11 dashboards consumed weekly by 134 GTM users replacing 22 hours/week of manual reporting (≈0.6 FTE)' is the screening pass. Adoption metrics signal the analyst owns the consumption side of the work — not just the build.

Should I list Excel on a data analyst resume in 2026?

Yes if it is in the job description (most FP&A and many ops / finance roles still require it), no if the role is product / growth / SaaS analytics where Excel is assumed. When you do list Excel, name the advanced features (Power Query, pivot tables, INDEX/MATCH, XLOOKUP, dynamic arrays, Excel-add-in for Snowflake / BigQuery) — not Excel as a noun. For modern SaaS analyst roles, leading with Excel reads as a junior signal; SQL + dbt + a BI tool should come first.

How do I show AI tooling on an analyst CV without sounding generic?

Name the workflow, the tool, the productivity gain, and the guard-rail. Weak: 'Familiar with ChatGPT.' Strong: 'Operate Cursor + Snowflake MCP for first-draft SQL; the dbt test suite (142 tests) plus a manual review-before-ship gate keeps the false-positive rate under 2%; saves ≈6 hours/week per analyst across the 5-person team.' AI fluency without guard-rails reads as junior; AI fluency with the validation layer + outcome metric is the Senior+ marker.

What is the difference between a data analyst and an analytics engineer skill list?

Data analyst skill list leads with SQL + a BI tool + Python; the bullets are about decisions shipped, dashboards adopted, experiments read out. Analytics engineer skill list leads with dbt (models, tests, exposures, contracts, semantic layer) + warehouse architecture + version control + CI/CD; the bullets are about model authorship, data SLA, downstream-consumption count, freshness, and trust. Senior analysts increasingly own analytics-engineering work too — calling out dbt mart authorship + semantic-layer ownership on an analyst CV is a Senior+ pull-up signal.

How many tools should I list on a data analyst CV?

Quality over quantity. List 8-12 tools you operate fluently and can defend in a live interview; group them by function (warehouse, transformation, BI, Python, experimentation, observability, AI). A list of 25 tools you touched once reads as padded; a focused 10 you actually use, with the modules and version specifics ('Looker LookML, Tableau Cloud, Snowflake Snowpark, dbt Cloud, Statsig with sequential testing'), reads as senior. Match the named tools to the job description before submitting.

Skills for Data Analyst Resume (2026) — SQL, dbt, BI, Python & Experimentation

The 2026 data analyst skills list has shifted significantly from 2018. The baseline is now SQL on a cloud warehouse (Snowflake, BigQuery, Databricks, Redshift) + dbt for transformation + a BI tool (Looker / LookML, Tableau, Power BI, Hex, Mode, Sigma) + Python or R for analysis + an experimentation or activation tool (Statsig, Eppo, GrowthBook, Hightouch, Census). Recruiters use boolean search — listing the named tool beats listing the category every time. This guide gives you the skill clusters that pass 2026 ATS and recruiter screens, with quantified bullet examples for product / marketing / finance / ops / healthcare analyst domains. Read the full role-by-role guide at /data-analyst-resume.

SQL & cloud warehouses

Advanced SQL (window functions, recursive CTEs, pivots, performance tuning)
Snowflake (Snowpark, Streams, Tasks, Dynamic Tables, clustering keys)
BigQuery (Standard SQL, ARRAY/STRUCT, scheduled queries, BI Engine)
Databricks (Delta Live Tables, Lakehouse, SQL Warehouse, Photon)
Redshift (RA3, Spectrum, materialised views)
Postgres / MySQL / SQL Server / Oracle (operational stores)
Trino / Presto / Athena (federated query)
Apache Iceberg / Delta / Hudi (open table formats)

Transformation, modelling & orchestration

dbt Core / dbt Cloud (models, tests, exposures, contracts)
dbt-utils, dbt-expectations, dbt Semantic Layer
Dataform (BigQuery-native)
Airflow / Dagster / Prefect (orchestration)
Fivetran / Stitch / Airbyte (managed ingestion)
Cube, Lightdash (semantic / metric layer)
Dimensional modelling (star / snowflake schemas, SCDs, fact / dimension)
Data contracts, exposures, model lineage, ownership matrix

BI tools, dashboards & notebooks

Looker (LookML modelling, Looker Studio, embedded analytics)
Tableau (Desktop, Server, Cloud, Tableau Prep, Pulse)
Power BI (DAX, Power Query, Fabric, composite models)
Hex (notebooks + apps + SQL + Python)
Mode (SQL + Python notebooks + reports)
Sigma (warehouse-native spreadsheet BI)
Metabase, Superset, Lightdash, Preset (open-source BI)
ThoughtSpot (search-driven BI)
JupyterLab, Deepnote, Databricks notebooks, Google Colab

Python, R & statistics

Python: pandas, NumPy, polars, scikit-learn, statsmodels, SciPy
Visualisation: matplotlib, seaborn, plotly, altair, Streamlit, Dash
Causal inference: DoWhy, EconML, CausalImpact, PyMC
Forecasting: Prophet, statsmodels SARIMA, fable, lifelines (survival)
ML basics: XGBoost, LightGBM, scikit-learn, MLflow
R: tidyverse, ggplot2, dplyr, broom, fable, lme4 (mixed-effects)
Statistical methods: hypothesis testing, regression (linear / logistic / GLM), A/B testing with confidence intervals, sequential / Bayesian testing, cohort and survival analysis, time-series, propensity-score matching, difference-in-differences

Experimentation, attribution & activation

Statsig, Eppo, GrowthBook (modern experimentation platforms)
Optimizely, VWO, AB Tasty (web experimentation)
Amplitude Experiment, Mixpanel Experiments, PostHog
Hightouch, Census (reverse-ETL / activation)
Segment, RudderStack, mParticle, Snowplow (CDP / event collection)
Iterable, Braze, Customer.io, Klaviyo (downstream lifecycle)
Triple Whale, Northbeam, Polar (DTC attribution)
MMM (media mix modelling), incrementality testing, geo holdouts

Governance, observability & AI tooling

Monte Carlo, Bigeye, Lightup, Soda (data observability)
Datafold (data diff, regression testing for dbt)
Atlan, Alation, Collibra, Select Star (data catalogue)
Immuta, Privacera (governance / RBAC)
Cursor + warehouse MCP (AI-assisted SQL drafting)
Hex Magic, Notion AI, Claude / GPT (notebook + narrative drafting)
Vanna, Defog, Julius (text-to-SQL)
dbt-test-backed validation gates for AI-generated code

Domain expertise & soft skills

Product analytics: funnel, cohort, retention, activation, pricing tests
Marketing analytics: attribution, MMM, channel mix, lifecycle revenue
Finance / FP&A: forecasting, variance attribution, unit economics
Operations & supply-chain: throughput, inventory turn, freight cost
Healthcare: HEDIS, claims, population-health, HIPAA-compliant warehouse
Fintech: transaction analytics, fraud signal modelling, SOX / SOC 2
Stakeholder requirements gathering, executive readout writing (1-page memo, 3-slide deck)
Mentoring junior analysts, hire / ramp loops, analyst capacity planning

Resume bullet examples

Built the LookML revenue mart and 11 executive dashboards (Snowflake + dbt + Looker) consumed weekly by 134 GTM users; replaced 22 hours/week of manual finance reporting (≈0.6 FTE) and surfaced a $1.4M unbilled-revenue gap patched in 30 days.
Designed and read out 28 Statsig experiments in 2025 across activation and pricing; the simplified-onboarding test lifted Day-7 activation from 31% to 39% (95% CI, n=82,000), the annual-default pricing test lifted ARR-per-signup 14% — together adding ≈$1.2M ARR run-rate.
Built a churn-driver logistic regression on 2.4M paid accounts (Snowflake + dbt + scikit-learn); identified the top 3 early-cancellation triggers and informed a feature-flag rollout that lifted month-2 retention from 71% to 78%, equivalent to ≈$3.1M ARR retained.
Authored 142 dbt tests + 18 Monte Carlo monitors across the marketing and revenue marts; detected a Salesforce sync outage 3 hours before quarter-end exec readout (preventing a $4.2M reported-revenue misstatement) and lifted weekly stakeholder-NPS on data trust from 6.4 to 8.7.
Built a 12-channel media-mix model in Python (statsmodels + scikit-learn) on 18 months of paid + organic data; reattributed $11.4M of annual spend, recommended a 28% reallocation; the resulting paid-CAC dropped from $4,180 to $2,940 over two quarters.
Owned the quarterly revenue + bookings forecast in a Snowflake + dbt + Excel-add-in stack across 6 product lines; lifted forecast-vs-actual accuracy from 81% to 94%; variance attribution informed the board's decision to pause hiring in 2 verticals, saving ≈$2.8M opex.
Migrated 9 legacy Tableau workbooks into a single LookML-modelled metric layer; reduced dashboard build time per ad-hoc request from 3.5 hours to 35 minutes; analyst sprint throughput rose from 9 to 17 requests/sprint at flat headcount.
Operate Cursor + Snowflake MCP for first-draft SQL on ad-hoc requests, validated against the dbt test suite (142 tests); saves ≈6 hours/week per analyst across the 5-person team without raising the false-positive rate above 2%.
Standardised the company-wide retention metric in dbt Semantic Layer; aligned 7 teams (Product, Marketing, Finance, CS, RevOps, BD, Board) on a single SSOT and eliminated 4 conflicting definitions across BI tools.
Mentored 3 junior analysts through their first dbt model + LookML view + read-out cycle; all 3 promoted to Data Analyst II within 9 months; team analyst-NPS on mentorship rose from 6.8 to 8.9 over 12 months.
Built a CausalImpact incrementality study on a $1.8M Q4 Meta-prospecting campaign (Bayesian counterfactual); identified that 64% of the reported lift was non-incremental, recommended a $1.1M reallocation into LinkedIn-ABM that returned 3.4× MQL volume.
Authored the company's first data SLA framework + RACI for the marketing + revenue marts; reduced incident MTTR from 18 hours to 4 hours and freed ≈8 analyst-hours/week previously spent on Slack-debugging upstream pipeline questions.

Get these skills onto your resume in the recruiter's language

Paste a job URL and your background into WadeCV. It maps your real experience against the posting, mirrors the exact skill keywords the ATS screens for, and writes quantified, recruiter-ready bullets — ATS-safe DOCX, free to try with 1 credit included.

Lead every analyst bullet with the question being answered and the decision shipped — not the report produced. The 2026 ATS and recruiter screen filters out 'analysed data', 'built dashboards' and 'wrote SQL' bullets before a human reads them. Use the question + method + decision + $ pattern: open with a strong verb (Built, Shipped, Designed, Owned, Authored, Lifted, Reduced), state the question or model, name the tool (Snowflake + dbt + Python, LookML mart, 28 Statsig experiments), state the outcome (lifted activation 8pp at 95% CI; reattributed $11.4M media spend; surfaced a $1.4M unbilled-revenue gap), and quote the dollar or percentage impact.

Domain framing matters. Product analyst CVs lead with experimentation and funnel analysis. Marketing analyst CVs lead with attribution, MMM and channel mix. Finance / FP&A CVs lead with forecast accuracy and variance attribution. Ops / supply-chain CVs lead with cost optimisation, throughput and inventory turn. Healthcare CVs lead with HEDIS, claims and HIPAA-compliant warehouse design. Fintech CVs lead with transaction analytics, fraud modelling and regulatory reporting. Mismatching the framing to the role is the single most common reason data analyst resumes get filtered into the wrong pipeline. WadeCV reads the job description and rewrites your bullets to match the domain.

Tailor your resume to highlight the right skills

Paste a job URL and WadeCV matches your skills to what the employer wants — then rewrites your bullets to prove each one.

1 free credit on signup
AI-powered fit analysis
ATS-optimised formatting

Common mistakes to avoid

Listing 'SQL' alone instead of 'SQL with window functions, recursive CTEs and performance tuning on Snowflake'
Leading with Excel on a modern SaaS analyst resume (acceptable for FP&A; reads as junior elsewhere)
Quoting 'increased conversion 23%' with no baseline, sample size, or confidence interval — reads as causally weak at Senior+
Listing 'familiar with' followed by a tool name instead of naming the workflow you operate with it
Omitting dbt + warehouse + semantic-layer literacy at Senior+ — these are now table-stakes
Padding the skill list with 25 tools touched once instead of 10 tools operated fluently
Soft-skill clichés ('detail-oriented', 'team player', 'data-driven') without evidence bullets
Domain mismatch — listing product-analytics skills for a finance / FP&A role or vice versa

Frequently asked questions

What technical skills should a data analyst put on a 2026 resume?
Lead with advanced SQL (window functions, recursive CTEs, performance tuning) on a named warehouse (Snowflake, BigQuery, Databricks, Redshift), dbt for transformation, a BI tool (Looker / LookML, Tableau, Power BI, Hex, Mode, Sigma), Python or R for analysis (pandas, scikit-learn, statsmodels, tidyverse), and an experimentation or activation tool (Statsig / Eppo / GrowthBook / Hightouch / Census). For Senior+ roles, add semantic-layer authorship, data observability (Monte Carlo / Bigeye / Datafold) and AI tooling (Cursor / Hex Magic / text-to-SQL workflows). Recruiters use boolean search — list the named tool, not the category.
What soft skills should a data analyst include?
Stakeholder requirements gathering (the question framing skill), executive readout writing (1-page memos, 3-slide decks, async written communication), data SLA negotiation, hypothesis-vs-analysis distinction, and conflict resolution when stakeholders disagree with a read. At Senior+ add mentoring (juniors taken from first model to promotion), hiring loops, analyst capacity planning, and OKR ownership. Avoid soft-skill clichés ('detail-oriented', 'team player') — replace with evidence ('mentored 3 juniors to Analyst-II promotion in 9 months').
How do I show statistical literacy on a data analyst CV?
Quote the method by name (regression, A/B test, difference-in-differences, propensity-score matching, CausalImpact, survival analysis with lifelines, Prophet for time-series), name the library (statsmodels, scikit-learn, DoWhy, EconML, PyMC, lifelines), and always quote the sample size and confidence interval. 'Lifted activation 8pp at 95% CI, n=82,000 vs control' reads as Senior; 'lifted activation 8pp' alone reads as channel-specialist. For causal claims (especially observational ones), name the design (geo-holdout, switchback, instrumental variable, synthetic control).
How important are dashboard adoption metrics on an analyst resume?
Critical at Mid+. Replace 'built X dashboards' with weekly active dashboard users, hours saved per stakeholder, and time-to-decision. 'Built 12 dashboards' is weak; 'Built 11 dashboards consumed weekly by 134 GTM users replacing 22 hours/week of manual reporting (≈0.6 FTE)' is the screening pass. Adoption metrics signal the analyst owns the consumption side of the work — not just the build.
Should I list Excel on a data analyst resume in 2026?
Yes if it is in the job description (most FP&A and many ops / finance roles still require it), no if the role is product / growth / SaaS analytics where Excel is assumed. When you do list Excel, name the advanced features (Power Query, pivot tables, INDEX/MATCH, XLOOKUP, dynamic arrays, Excel-add-in for Snowflake / BigQuery) — not Excel as a noun. For modern SaaS analyst roles, leading with Excel reads as a junior signal; SQL + dbt + a BI tool should come first.
How do I show AI tooling on an analyst CV without sounding generic?
Name the workflow, the tool, the productivity gain, and the guard-rail. Weak: 'Familiar with ChatGPT.' Strong: 'Operate Cursor + Snowflake MCP for first-draft SQL; the dbt test suite (142 tests) plus a manual review-before-ship gate keeps the false-positive rate under 2%; saves ≈6 hours/week per analyst across the 5-person team.' AI fluency without guard-rails reads as junior; AI fluency with the validation layer + outcome metric is the Senior+ marker.
What is the difference between a data analyst and an analytics engineer skill list?
Data analyst skill list leads with SQL + a BI tool + Python; the bullets are about decisions shipped, dashboards adopted, experiments read out. Analytics engineer skill list leads with dbt (models, tests, exposures, contracts, semantic layer) + warehouse architecture + version control + CI/CD; the bullets are about model authorship, data SLA, downstream-consumption count, freshness, and trust. Senior analysts increasingly own analytics-engineering work too — calling out dbt mart authorship + semantic-layer ownership on an analyst CV is a Senior+ pull-up signal.
How many tools should I list on a data analyst CV?
Quality over quantity. List 8-12 tools you operate fluently and can defend in a live interview; group them by function (warehouse, transformation, BI, Python, experimentation, observability, AI). A list of 25 tools you touched once reads as padded; a focused 10 you actually use, with the modules and version specifics ('Looker LookML, Tableau Cloud, Snowflake Snowpark, dbt Cloud, Statsig with sequential testing'), reads as senior. Match the named tools to the job description before submitting.

Related guides

Explore more guides

Compare AI resume tools

Build a resume that highlights these skills

WadeCV helps you tailor your CV to the role and surface the right skills and bullets for each application.