Skip links

Type and hit enter

The How To
Published in: How To "AI" Correctly

How To Use AI To Analyze Data

Author
Published on:

Written by Jacob Sipos
Published September 18, 2026

How To Use AI To Analyze Data

AI can find patterns quickly, but a pattern is not automatically an explanation.

AI can inspect datasets, calculate statistics, find trends, identify outliers, build charts, and compare groups.

Strong analysis starts with clean data and a clear question. AI accelerates exploration, but you still judge the method, assumptions, and conclusion.

REVENUE VS. RETENTION
TRENDHigher retention is associated with higher revenue.
OUTLIEROne segment performs below the pattern.
NEXTControl for customer size before concluding why.

First, What Can AI Actually Analyze?

Modern AI tools can inspect structured datasets, summarize columns, calculate statistics, create charts, identify trends and outliers, and run code-backed analysis. ChatGPT can analyze uploaded `.xlsx`, `.xls`, `.csv`, JSON, XML, text, and other supported files and can use Python for calculations, transformations, and statistical analysis.

Copilot in Excel can answer workbook questions and return charts, PivotTables, summaries, trends, and outliers.

The Important Distinction

AI calculates quickly, but it cannot rescue a vague question, biased sample, missing data, or invalid comparison.

Good Analysis Begins Before the First Calculation

Structured data is easier to analyze. OpenAI recommends clear column names, one record per row, and avoiding empty rows, unrelated tables, or images containing needed values.

InspectUnderstand columns and units
CleanFix blanks and duplicates
ValidateCheck impossible values
DefineChoose the question
AnalyzeUse the right method

Before analysis, ask AI to list data types, missing values, duplicates, suspicious ranges, inconsistent categories, and required assumptions.

A Better Question Produces Better Analysis

Weak request

“Analyze this data.”

Stronger request

“Compare monthly revenue and gross margin by region, identify unusual changes, and show which conclusions remain true after excluding the largest customer.”

Useful questions specify a metric, comparison, timeframe, segment, or decision. Ask AI to show its method so the result is easier to audit.

Try this: “Before analyzing the file, propose five questions this dataset can answer reliably and three questions it cannot answer from the available fields.”

Start With Descriptive Analysis Before Looking for Explanations

Before asking why, establish what happened. Counts, averages, medians, ranges, distributions, missing values, totals, and rates reveal the dataset’s shape.

DATASET SUMMARY
ROWS24,812Customer transactions
MEDIAN$84Order value
MISSING2.1%Region field
OUTLIERS37Flagged orders

Median can be more useful than average when a few large values distort the distribution. You decide which statistic fits the question.

Compare Groups Instead of Treating the Dataset as One Average

Overall averages can hide important differences between customer types, locations, products, channels, periods, or cohorts.

Segment the data intentionally.

RegionWhere does performance differ?
Customer typeWho behaves differently?
Acquisition channelWhich source performs best?
ProductWhich offerings drive results?
CohortHow does behavior change over time?
PeriodIs the pattern seasonal?

Ask for raw values and denominators behind percentages. A 40% rate from ten people differs greatly from 40% across ten thousand.

Use AI To Find Trends, Changes, and Outliers

Copilot in Excel can surface trends and outliers, while ChatGPT can use Python to detect changes and unusual values. ·

Find the pattern.

Identify growth, decline, seasonality, clustering, or unusual values.

Quantify it.

Measure the size, timing, and persistence of the change.

Check whether it is data quality.

An outlier may be a real event or a bad record.

Investigate before explaining.

The pattern alone does not prove the cause.

Correlation Is Useful. Causation Requires More Evidence.

AI can calculate correlations, cross-tabs, regressions, and other relationships worth investigating.

OBSERVATION

Customers with higher product usage have higher retention.

→
CAREFUL INTERPRETATION

Usage and retention are associated. Other factors may influence both.

High usage may partly reflect existing satisfaction rather than cause it. Ask AI for plausible confounders and alternative explanations.

Use Visuals To Reveal Structure, Not Decorate the Result

ChatGPT can create charts, while Copilot in Excel can generate charts and PivotTables from plain-language requests. ·

Match the chart to the job: lines for time, bars for comparisons, scatterplots for relationships, histograms for distributions, and tables for exact values.

Useful request: “Recommend the best visualization for this question before creating it. Explain why that chart is more appropriate than two alternatives.”

AI Can Run Statistical Methods, but Method Choice Still Matters

ChatGPT can use Python for statistical analysis, while Copilot in Excel can provide Python-backed analytical answers. ·

Do not ask only, “Is this statistically significant?”

Also ask whether the test fits the data, whether assumptions are satisfied, how large the effect is, how uncertain the estimate is, and whether the result is practically meaningful.

For consequential research, medical, financial, scientific, regulatory, or policy analysis, use qualified statistical or domain review.

Data Analysis Can Include Thousands of Words, Not Just Numbers

AI can turn surveys, reviews, support tickets, and notes into themes, classifications, and sentiment. Copilot in Excel currently supports summarizing text and identifying key themes and sentiment in workbook data.

TEXT → STRUCTURED DATA

Customer comment

“The product works well, but setup took too long and the instructions were confusing.”

Product: Positive
Setup: Negative
Documentation: Negative
Theme: Onboarding

Manually review a sample of classifications so inconsistent labeling does not go unnoticed.

The Final Step Is Translating Analysis Into a Decision

An analysis can be mathematically correct and still fail to guide a decision.

Analysis FindingDecision Question
One channel has the highest conversion rate.Does it also produce enough volume and acceptable acquisition cost?
A product’s revenue grew 20%.Did margin, returns, or customer retention change at the same time?
One region is underperforming.Is the issue demand, pricing, staffing, competition, or data quality?
Customers with one behavior retain longer.Can that behavior be influenced, or is it merely associated with stronger customers?

Separate finding, interpretation, uncertainty, and next test so a pattern does not become an overconfident conclusion.

Things People Often Miss About AI Data Analysis

You can ask AI to audit the dataset before analyzing it.

Have it identify missing data, duplicates, impossible values, and category inconsistencies first.

You can ask for competing explanations.

After finding a pattern, request plausible reasons and the evidence that would distinguish them.

You can request sensitivity analysis.

Ask whether the conclusion changes after removing outliers or changing assumptions.

You can ask AI to show the exact code or calculation method.

Code-backed analysis is easier to inspect when you can see how it was produced.

You can create reusable analytical workflows.

Define a standard process for cleanup, validation, metrics, charts, and quality checks.

You can make AI argue against its own conclusion.

Ask for the strongest reasons the finding may be misleading.

You can combine quantitative and qualitative data.

Numbers show what happened while text may suggest what to investigate next.

Useful AI Data Analysis Prompts

Describe this datasetFind missing valuesCheck duplicatesFlag impossible valuesSummarize distributionsCompare segmentsFind outliersAnalyze trendsCalculate growthShow median and meanFind correlationsList confoundersCreate a scatterplotBuild a cohort tableRun sensitivity analysisExplain the methodShow the codeChallenge the conclusionClassify commentsSummarize themesRecommend next testsSeparate fact from inference

AI Can Make Analysis Faster, but It Can Also Make Bad Analysis Look Convincing

ProsCons or Risks
Can inspect and summarize large datasets quickly.Poor data quality can produce misleading results.
Can run calculations and statistics with code.The chosen method may be inappropriate for the question or data.
Can find trends and outliers that deserve investigation.Patterns can be mistaken for causes.
Can make analysis accessible without advanced coding skills.Users may trust methods they do not understand well enough to audit.
Can combine charts, tables, text, and numbers.Polished outputs can create false confidence.
Can accelerate repeated analysis.Private, regulated, proprietary, or sensitive data requires careful handling.

Microsoft and OpenAI both advise reviewing AI-generated analysis and assumptions before relying on results.

A 7 Day AI Data Analysis Challenge

Day 1: Audit a dataset.

Identify columns, blanks, duplicates, and suspicious values.

Day 2: Describe the data.

Calculate counts, averages, medians, and ranges.

Day 3: Compare segments.

Compare one important metric across meaningful groups.

Day 4: Find one anomaly.

Check whether it is real or a data problem.

Day 5: Explore one relationship.

Create a relationship view and list alternative explanations.

Day 6: Analyze text.

Classify comments and manually review the categories.

Day 7: Challenge your conclusion.

Run a sensitivity check and ask what could prove your interpretation wrong.

Final Thoughts

AI lowers the friction of data analysis. More people can explore datasets, calculate statistics, and build charts without starting from code.

That makes analytical judgment more important, not less.

Use AI to explore faster and notice patterns you might miss. Then verify the data, method, assumptions, uncertainty, and explanation before acting.

Disclaimer: This article provides general educational information about AI-assisted data analysis. AI can produce incorrect calculations, inappropriate methods, or unsupported conclusions. Verify important data, code, formulas, methods, assumptions, and results. Do not rely on AI alone for high-impact decisions. Protect confidential, personal, regulated, and proprietary data and seek qualified professional review when needed.

References:ChatGPT data analysis · ·

Jacob Sipos is an educator, entrepreneur, former paramedic, and technology specialist with more than a decade of experience helping people understand complex subjects and turn that knowledge into practical results. Jacob's career in education began in emergency medical services. After working as a paramedic, he began teaching EMS education, where he developed a passion for taking difficult, high-stakes material and making it easier for students to understand, retain, and apply. Over the years, he refined his teaching methods by paying close attention to where students struggled, how they processed information, and which approaches helped turn complicated concepts into something that finally made sense. His experience in education eventually expanded beyond EMS. Jacob earned a California teaching credential and taught at the high school level, further developing his experience working with different types of learners and finding effective ways to communicate information. He also began applying many of the educational techniques he had developed for EMS students to learners in other fields, including individuals preparing for national music therapy certification examinations. Across his educational work, Jacob has helped thousands of students prepare for and pass national certification exams. Those experiences reinforced a principle that has shaped much of his career: expertise is valuable, but the ability to organize, explain, and teach information in a way that another person can actually use is an entirely different skill. That philosophy eventually expanded beyond formal education. Over the years, Jacob has learned and worked with more than 100 software programs, platforms, and digital tools spanning automation, artificial intelligence, design, video production, productivity, marketing, customer relationship management, business operations, communication, and software development. His experience includes platforms and technology ecosystems such as Zapier, HubSpot, ClickUp, Canva, Microsoft Teams, DaVinci Resolve, Google Workspace, Microsoft 365, WordPress, Shopify, Slack, Zoom, Dropbox, Adobe products, QuickBooks, Calendly, GoDaddy, Cloudflare, GitHub, OpenAI tools, Claude, and numerous other business, creative, educational, and productivity technologies. Rather than simply learning how to use individual programs, Jacob became interested in understanding how technology could be combined to solve real problems. He began developing automations, connecting systems, improving workflows, building digital educational experiences, and finding more efficient ways to accomplish tasks. This work eventually contributed to his creation and operation of businesses focused on education, automation, technology, and software. Throughout these different areas, one pattern continued to emerge. When Jacob spent time learning how to accomplish something, he often developed his own methods for simplifying the process. When he shared those methods with others, he found that the same teaching principles he had spent years refining in education could help people understand far more than academic material. That realization became part of the foundation for The How To Company. Jacob created The How To Company to bring that approach to a much broader range of subjects. His writing explores technology, software, automation, artificial intelligence, personal finance, business, education, productivity, and practical everyday challenges. Whether the subject involves learning a new program, improving a financial habit, using technology more effectively, or figuring out how to accomplish an unfamiliar task, his goal is to make the process easier to understand and act upon. His approach combines research, firsthand experience, continued learning, and more than a decade spent developing effective ways to teach. He focuses on understanding a subject, identifying the information that actually matters, and breaking it into clear, practical steps without assuming the reader already knows the terminology or background. Today, Jacob continues to explore emerging technologies, develop software and automation systems, build businesses, study new subjects, and refine the way information is taught. Through The How To Company, he brings those experiences together with a simple purpose: to take useful knowledge, make it understandable, and help people put it into action.

You may also like