Prompt of the Day: Table Analyst -- Read the Story Behind the Numbers in Raw Data
You are sitting in front of a spreadsheet. Revenue figures for the last twelve months. Survey results from 200 customers. Website statistics from Google Analytics. Or simply a list of expenses you finally want to understand.
You see numbers. But you do not see the story. Which month was actually bad -- and why? Which customer segment is the most profitable? What trend is hiding behind the fluctuations? Are there outliers pointing to a problem?
The problem: Data analysis feels like a job for specialists. Pivot tables, VLOOKUP, statistical significance -- for most people, these are foreign concepts. So what usually happens: the data stays in the spreadsheet, and decisions are made on gut feeling.
The solution: Modern AI models can read and analyze tabular data directly. You copy your data into the chat, and AI takes on the role of an experienced analyst -- one who does not just crunch numbers but explains what the numbers mean and what you should do next.
What you can use this for:
- Revenue and financial data -- spotting trends, finding seasonal patterns, understanding growth or decline
- Surveys and feedback -- What are your customers really saying? What patterns hide in the responses?
- Marketing metrics -- Which campaign worked? Where are you burning budget?
- HR statistics -- turnover, sick leave, overtime -- where is action needed?
- Personal finances -- Where is your money going? Where can you save?
- Project data -- analyzing time spent, budget deviations, milestone delays
How to use it:
1. Prepare your data: Export your spreadsheet as CSV or copy it directly from Excel/Google Sheets. Tip: Select all cells, copy, and paste into the chat -- AI recognizes the table structure automatically.
2. Use the prompt: Copy the prompt below and paste your data. The more context you provide (what the columns represent, what the table reports on, what your goal is), the better the analysis.
3. Ask follow-up questions: The first analysis is the starting point. Ask specifically: 'Why is March so low?', 'Compare Q1 with Q2', 'What if we discontinued product C?'
Pro tips:
- Request visualizations: 'Create the Python code for a chart showing the revenue trend over the last 12 months.' -- You can run the code in Google Colab or a Jupyter Notebook
- Run scenarios: 'What happens to our profit if raw material costs increase by 15% and revenue stays the same?'
- Benchmarking: 'Are these numbers good, average, or bad for a company of our size and industry?'
- Use regularly: Create a template with your standard questions and feed in new data every month -- this builds a consistent reporting system
- Mind privacy: Anonymize personal data before uploading. Replace names with codes, remove email addresses and birth dates. For sensitive company data, use API access or business versions with privacy guarantees
You are an experienced data analyst who makes complex data understandable for non-technical people. You do not just look for numbers -- you find the story the data tells and derive concrete, actionable recommendations.
**My data:**
[Paste your table data here. You can use CSV data, a copied Excel table, or even just a list. Briefly describe what the columns mean if it is not obvious.]
**Context:**
- Industry/area: [e.g., 'Online shop for office supplies', 'SaaS startup', 'Freelance graphic design', 'personal finances']
- Time period: [e.g., 'January to May 2026', 'Last 12 months', 'Q1 2026']
- What I want to know: [e.g., 'Why has revenue been dropping since March?', 'Which products should I push?', 'Where can I cut costs?', 'Are there noticeable patterns?']
---
Analyze my data in 6 steps:
**1. Data Overview**
- What did I give you? (Number of records, columns, time period, completeness)
- Are there missing or conspicuous values I should know about?
- First assessment: Does this look like healthy or problematic data?
**2. The 5 Most Important Findings**
Summarize the five most notable patterns, trends, or facts -- in 1-2 sentences each, in clear everyday language. No jargon without explanation.
**3. Trends and Patterns**
- What trends do you see over the time period? (rising, falling, seasonal, cyclical)
- Are there turning points -- months or periods where something changed?
- Correlations: Are certain values connected? (e.g., 'When X rises, Y falls')
**4. Outliers and Warning Signs**
- Which data points deviate significantly from the pattern?
- Possible explanations for each outlier
- What warning signs should I keep an eye on?
**5. Comparisons and Context**
- Compare the best and worst periods/categories -- what sets them apart?
- If possible: How do my numbers stack up against industry benchmarks?
- 80/20 analysis: Which 20% of my data (products, customers, categories) drives 80% of the result?
**6. Action Recommendations**
Give me 3-5 concrete, prioritized actions based on the data:
- **Do now** (this week): [Action with expected impact]
- **Short-term** (next 4 weeks): [Action with expected impact]
- **Medium-term** (next quarter): [Action with expected impact]
**Finally:**
- **Executive summary:** Summarize everything in 3 sentences -- so I can explain it to my boss or partner in 30 seconds
- **Next question:** What single follow-up question should I ask you next to gain even deeper insights?
**Rules:**
- Speak plainly -- no statistics jargon without an immediate explanation in parentheses
- If the data is insufficient for a reliable conclusion, say so honestly instead of guessing
- Always put numbers in context: not just '15% decline' but '15% decline -- that is roughly 3,000 USD less revenue'
- Separate confirmed findings from assumptions ('The data shows...' vs. 'A possible explanation would be...')
- When recommending visualizations, briefly describe which chart type fits best and why