Five polished, well-documented projects can strengthen an entry-level data analyst application—but no project can guarantee a job. The best portfolio demonstrates a complete workflow: defining a business question, cleaning imperfect data, analyzing it with relevant tools, validating the results, and explaining what a stakeholder should do next.
This set covers customer analytics, profitability, job-market data, workforce reporting, and product analytics. Together, the projects can demonstrate SQL, Excel, Python, data visualization, business judgment, and responsible communication. Choose the tools that appear in the roles you want; you do not need to force Python, Tableau, and Power BI into every project.
As an Amazon Associate I earn from qualifying purchases.
What makes a data analyst project job-worthy?
A portfolio project is more convincing when it resembles work an analyst might actually do. It should answer a specific question, use data that requires some judgment, show the path from raw data to conclusion, and make its assumptions easy to inspect.
Current portfolio guidance generally favors three to five complete projects over a large collection of unfinished tutorials. The goal is not to prove that you can create attractive charts. It is to show that you can define metrics, find data-quality problems, produce reproducible analysis, and communicate a recommendation. See Dataquest’s project guidance for additional portfolio criteria.
Quick comparison
| Project | Primary skills | Main output | What it demonstrates |
|---|---|---|---|
| Customer churn and retention | SQL, Python, segmentation, BI | Retention dashboard | Customer and KPI analysis |
| E-commerce profitability | Excel, SQL, BI | Executive dashboard | Commercial reasoning |
| Job-market monitoring | Python, text processing, SQL, BI | Skills dashboard | Data collection and sampling judgment |
| Employee exit analysis | Cleaning, joins, privacy, visualization | HR insights report | Data quality and responsible analysis |
| Product funnel and cohorts | SQL, event data, cohorts | Product dashboard | Conversion and retention metrics |
1. Customer churn and retention analysis
Business question
Which customer segments are most likely to churn, why are they leaving, and which retention actions should the company prioritize?
This is a strong general-purpose project because it combines customer behavior, segmentation, KPI design, SQL, cleaning, and business recommendations. It is more valuable than simply reporting one overall churn percentage.
Data to look for
- Customer and account identifiers
- Signup and cancellation dates
- Plan type and monthly revenue
- Usage or engagement measures
- Support contacts
- Geography, acquisition channel, or customer segment
Telecom, subscription-software, streaming, membership, and loyalty datasets can work. Public sources include Kaggle and Data.gov. Check the dataset’s license and document its coverage, time period, and limitations.
Analysis and deliverables
- Overall churn rate and monthly churn trend
- Churn by plan, tenure, geography, and acquisition channel
- Revenue at risk, not just the number of customers lost
- Cohort retention by signup month or quarter
- Relationships between usage, support contacts, and churn
- High-value segments with both elevated churn and meaningful revenue
Use SQL for joins, aggregations, cohort calculations, and reusable views. Use Python and pandas for cleaning and exploration, then Tableau or Power BI for the final dashboard. An optional Excel summary can make the result easier for a nontechnical stakeholder to review.
Publish the SQL scripts, a data dictionary, a cleaning notebook, dashboard screenshots or a live dashboard, a one-page executive summary, and three prioritized recommendations. Include what additional data would be needed before launching a retention campaign.
Interview angle
Be ready to explain how you moved from “customers on plan X churn more” to a cautious recommendation such as testing an onboarding intervention for new customers on that plan. A correlation is not proof that the plan caused the churn.
2. E-commerce sales and profitability dashboard
Business question
Which products, regions, and customer segments drive revenue while weakening profitability?
Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minuteWindows Errors? Fix Them Before They Spread
Repair common Windows errors and clear accumulated junk for a smoother, more stable PC - no reinstall needed.Free scan · no reinstallMany beginner dashboards stop at sales. Adding profit, margin, discounts, returns, shipping costs, or product costs makes the project more commercially realistic.
Data to look for
- Orders and line items
- Products, categories, and costs
- Customers and regions
- Dates, discounts, returns, and shipping information
Before calculating anything, determine whether profit is supplied directly or must be estimated. If you use assumptions, label them prominently. Never describe revenue as profit.
Analysis and dashboard
- Revenue, orders, average order value, and profit
- Gross margin by category
- The relationship between discounting and profit
- Top and bottom products
- Regional performance
- New versus returning customers
- Return-rate impact and monthly or quarterly trends
- High-revenue products with poor margins
Build KPI cards, date and region filters, a category-profitability view, a revenue-versus-margin scatter plot, product or customer drill-down, and a clearly labeled recommendation panel. Use Excel to demonstrate pivot tables and formulas, SQL for the core analysis, and Tableau or Power BI for interactive reporting.
Common errors
- Double-counting orders after joining order-level and line-item-level tables
- Failing to define whether discount is a percentage or currency amount
- Hiding negative-margin products with filters
- Ignoring duplicate orders and outliers
- Presenting a dashboard without a business action
Interview angle
Explain how a merchandising manager could use your findings to review discount policies, product assortment, or regional promotions.
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
3. Job-market skills monitoring dashboard
Business question
Which skills, tools, industries, and locations appear most often in data analyst job postings?
This project connects directly to your own job search while demonstrating collection, text processing, SQL or Python, visualization, and communication. A similar job-trends concept was highlighted by KDnuggets.
Collect the data responsibly
Use a legally reusable public dataset, a downloadable dataset, an API whose terms permit your intended use, or a clearly labeled manually collected sample. Do not tell recruiters that your results describe the entire job market. Check terms of service, robots policies, authentication requirements, and rate limits before collecting listings from any website.
Analysis and documentation
- Skill frequency and common skill combinations, such as SQL plus Power BI
- Skills by job title, industry, location, and remote status
- Salary ranges only when the source defines them consistently
- Posting dates and the exact collection window
- Duplicate and reposted listings
- Differences in employer vocabulary
Create a taxonomy that maps variations such as PowerBI, Power BI, and Microsoft Power BI. Distinguish required from preferred skills, exact tools from related terms, posting counts from unique employers, and one geography from the market as a whole.
The Tool Desk
Outbyte Driver Updater FREEScan for outdated or missing drivers - takes under a minuteDriver Scan →Outbyte PC Repair FREEClear out junk files and repair common Windows errorsFree Scan →Publish your collection method, taxonomy, duplicate-removal logic, dashboard, and a “how to read this” section. Present it as a dated snapshot. Skill counts do not prove importance: a tool may be optional, legacy, or merely part of a standard job-description template.
Interview angle
Discuss sampling bias and data quality—not only which technologies ranked first. This shows the judgment employers need when working with imperfect observational data.
4. Employee exit-survey and workforce analysis
Business question
What patterns appear among employee departures, and which factors should an HR team investigate further?
Rank #3
- Includes 24 permanently bound, top-loading sleeves that display up to 48 letter-size pages.
- Designed for standard 8.5" × 11" documents: Lightweight presentation book fits US letter-size papers.
- Clear front cover and spine inserts let you add labels or title pages for easy identification.
- Durable plastic covers with non-glare polypropylene sleeves help protect documents from dirt and moisture for everyday presentation and storage.
- Holds standard 8.5" × 11" documents.
This project is useful for demonstrating messy joins, inconsistent categories, missing-data analysis, and careful communication of sensitive findings.
Do these 3 things before closing this tab:
1Repair Windows errors before they cause bigger problems2Fix the driver behind crashes, sound loss and screen glitches3Clear out junk files and repair common Windows errorsData to include
- Exit surveys and employment status
- Department, tenure, and work arrangement
- Reasons for leaving and satisfaction scores
- Demographic fields and compensation bands, where appropriate
Analysis and safeguards
- Calculate the survey response rate.
- Report missingness by field.
- Compare departure reasons by department and tenure band.
- Compare satisfaction with intention to leave.
- Assess whether respondents differ from the overall employee population.
- Normalize inconsistent categories and show before-and-after examples.
- Suppress small groups when results could expose individuals.
Do not identify employees, publish personally identifiable information, or suggest that a demographic group is “the problem.” Separate self-reported survey sentiment from verified employment outcomes, and avoid generalizing a small sample to the entire workforce.
Use Python and pandas for cleaning and joins, SQL for grouped analysis, Excel for a controlled summary, and Power BI or Tableau for an anonymized dashboard. Include a data-quality report, privacy note, recommendations for further investigation, and an explanation of how missing-value treatment could affect the conclusions.
Interview angle
Explain why deleting every incomplete row might bias the findings and how you decided whether to retain, recode, or separately report missing values.
5. Product funnel and cohort analysis
Business question
Where do users abandon a signup or purchase funnel, and how does retention differ by acquisition cohort?
Recommended Free Tools
This project demonstrates event data, SQL joins and window functions, conversion definitions, cohort construction, retention analysis, and product recommendations.
Data to look for
- User ID and event name
- Event timestamp and session or order ID
- Acquisition source, device, and geography
- Subscription or purchase status
Analysis to include
- Conversion at each funnel step
- Time between events
- Conversion by channel and device
- Signup cohorts
- Week-one, week-four, or month-three retention
- Checks for impossible or out-of-order event sequences
Define the denominator for every conversion rate and distinguish unique users from sessions and events. One illustrative SQL pattern is:
WITH user_steps AS (
SELECT
user_id,
MAX(CASE WHEN event_name = 'signup' THEN 1 ELSE 0 END) AS signed_up,
MAX(CASE WHEN event_name = 'trial_started' THEN 1 ELSE 0 END) AS started_trial,
MAX(CASE WHEN event_name = 'purchase' THEN 1 ELSE 0 END) AS purchased
FROM events
GROUP BY user_id
)
SELECT
SUM(signed_up) AS signups,
SUM(started_trial) AS trial_starts,
SUM(purchased) AS purchases,
ROUND(100.0 * SUM(started_trial) / NULLIF(SUM(signed_up), 0), 2) AS signup_to_trial_pct,
ROUND(100.0 * SUM(purchased) / NULLIF(SUM(started_trial), 0), 2) AS trial_to_purchase_pct
FROM user_steps;The event names are examples; adapt them to the dataset’s schema. Publish the event schema, SQL queries, funnel chart, retention heatmap, cohort table, dashboard, and recommendations ranked by estimated impact and confidence.
A funnel drop-off is not automatically a product defect. Users may be allowed to complete steps out of order, and channel performance is not necessarily causal without an experiment or credible causal design.
Free tools Windows power users keep installed
One-click scans. No signup required.
A repeatable workflow for every project
- Define the decision: identify the stakeholder, question, possible action, and success metric.
- Document the data: record the source, access date, geography, time period, license, and limitations.
- Profile the raw files: inspect row counts, types, null rates, duplicate keys, outliers, impossible values, and inconsistent categories.
- Clean reproducibly: preserve raw files and explain every deletion, recode, and imputation. Check row counts after joins.
- Analyze: begin descriptively, use appropriate denominators, segment results, and test sensitivity to important assumptions.
- Validate: reconcile totals, inspect samples, check metric definitions, and verify that visualizations match the underlying query.
- Communicate: lead with the decision, annotate important findings, and state confidence limits.
- Publish: include the repository, scripts, data dictionary, screenshots or live dashboard, reproduction instructions, license, and attribution.
- Prepare: practice a one-minute explanation, the hardest data-quality issue, the most important analytical choice, and the project’s limitation.
How to package the portfolio
Each GitHub repository should make it possible for a reviewer to understand the project in a few minutes. A README can follow this structure:
# Project title
## Business question
What decision does this analysis support?
## Dataset
Source, date accessed, coverage, license, and limitations.
## Tools
SQL, Python/pandas, Excel, Tableau, or Power BI.
## Process
Cleaning, transformations, joins, metric definitions, and validation.
## Key findings
Three to five concise findings.
## Recommendations
Actions, intended users, and caveats.
## Reproduction
Setup steps and file order.
## Limitations
Sampling, missing data, assumptions, and what the analysis cannot prove.Show both the analytical work and the finished communication. A polished dashboard should not conceal the queries, cleaning decisions, or validation behind it.
Choosing tools without overspending
Start with the least expensive stack that supports the target role:
- Free baseline: GitHub, public data, local Python/Jupyter, and SQL practice.
- Public interactive demo: Streamlit Community Cloud describes its Community Cloud as free for public apps. Do not use it for confidential or sensitive work.
- Targeted BI tool: choose Tableau or Power BI based on the job postings you are targeting. Tableau’s pricing page currently advertises a free Desktop Edition for local analysis of Excel, CSV, and database files; paid products and Tableau Next are separate offerings. Verify current Power BI pricing before purchasing.
- Structured education: the Google Data Analytics Professional Certificate can suit learners who need a guided curriculum. Its U.S. and Canada FAQ has listed $49 per month after a seven-day trial, but pricing can change. A certificate does not replace original portfolio work.
Free dataset sources include Data.gov, Kaggle, NYC Open Data, Our World in Data, and World Bank Open Data. Confirm each dataset’s reuse terms before publishing it.
Quick wins for a faster PC:
Clear out junk files and repair common Windows errorsFree Scan →Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →What to avoid
- Using Titanic or Iris as your only portfolio centerpiece
- Copying a popular tutorial without a new question or decision frame
- Publishing data with unclear provenance or no license
- Exposing confidential, personally identifiable, or proprietary data
- Creating a dashboard from fabricated numbers without disclosure
- Adding machine learning merely to appear advanced
Machine learning is appropriate only when it answers a genuine business question. If you build a churn or approval model, define the target, establish a baseline, prevent leakage, separate training and test data, report suitable metrics, explain business costs and thresholds, and discuss fairness and monitoring. For many entry-level analyst roles, a clear segmentation or funnel project is more relevant than an elaborate model no one can use.
How to choose your final projects
Score each candidate from 1 to 5 for business relevance, data realism, skill coverage, originality, explainability, reproducibility, ethical safety, presentation quality, interview value, and time to completion. A familiar business problem with an original dataset or decision frame is often the best compromise between memorability and feasibility.
Review target job postings before committing to a tool stack. SQL, Excel, Python, and a BI tool recur across many analyst roles, but requirements vary by employer, industry, geography, and title. Do not assume that every employer ranks the same skill first.
Prepare an interview story
For each project, practice this sequence:
- Situation: who the stakeholder was and why the question mattered.
- Question: the decision your analysis supported.
- Data problem: the most difficult quality, structure, or sampling issue.
- Method: the queries, transformations, metrics, and validation you used.
- Finding: the result that mattered most.
- Recommendation: the action you would prioritize.
- Limitation: what the data cannot prove.
- Next step: the additional analysis, experiment, or data you would request.
That story is often more valuable than adding another chart. A portfolio can provide evidence of practical ability and useful interview talking points, but hiring also depends on role fit, communication, experience, and the employer’s process.
Quick Recap
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
