Data Analysis Prompts

Ready-made prompts to interpret numbers in tables, turn them into presentation lines and learn Excel formulas.

Ready-made prompt templates

• Top 3 trends in data: Explain the 3 most notable trends in the following data, in plain language: [data or table]
  Sample case: Data (example): Jan 120, Feb 135, Mar 150, Apr 148, May 190, Jun 205 (monthly orders).
  1. Overall rise: orders went from 120 to 205 in six months, roughly a 71% increase.
  2. There is a small dip in April (150 to 148); the gap is tiny and not a trend by itself.
  3. The biggest jump was April to May (148 to 190, about 28%). Check whether a campaign or a seasonal effect happened that month.
  
  Note: six months of data cannot show cause and effect; it only supports observation.
  This sample output was produced with Claude; your result may differ.
• Sales data checklist: Prepare a checklist of questions I should ask when interpreting this sales data: [data summary].
  Sample case: Data summary: store-level sales for the last 3 months, product categories and discount periods.
  Checklist:
  ☐ Does the data cover the same period? Are the same months compared with last year?
  ☐ Unit sales, revenue or profit? Which one am I looking at?
  ☐ Could discount periods have inflated sales?
  ☐ Did the number of stores or opening hours change?
  ☐ Is there a shift between categories (is one eating another's sales)?
  ☐ Are returns and cancellations excluded?
  ☐ Are there missing or duplicate records?
  ☐ Is there an outside factor (holiday, weather, competitor campaign)?
  This sample output was produced with Claude; your result may differ.
• Turn numbers into 3 sentences: Turn the following numbers into 3 short, striking sentences for use in a presentation: [numbers and context]
  Sample case: Numbers: Website visits rose from 40,000 to 52,000; the conversion rate rose from 2% to 2.5%.
  1. Our visitors rose from 40,000 to 52,000: a 30% increase.
  2. The share of visitors who buy rose from 2% to 2.5%.
  3. So more people came, and a larger part of them bought.
  
  Check: 40,000 × 2% = 800 sales; 52,000 × 2.5% = 1,300 sales. With this data sales would be up by about 62%.
  This sample output was produced with Claude; your result may differ.
• Outlier check questions: Suggest questions that would help me assess whether there's an inconsistency or outlier in this dataset: [data summary].
  Sample case: Data summary: order amounts per customer; most between 100 and 500, one record is 45,000.
  Questions:
  1. Is the 45,000 record a real order or a wrongly entered value (for example an extra zero)?
  2. Are there similar large amounts from the same customer; could it be a bulk purchase?
  3. How much does the average change when this record is removed? Is the median different from the mean?
  4. Could currency or unit (piece/case) have been mixed up?
  5. On what date and through which channel did the record come?
  6. Does this amount belong to a refund or cancellation?
  This sample output was produced with Claude; your result may differ.
• Explain an A/B test result: Write how I'd explain the following A/B test result to a business partner without using statistical jargon: [test results].
  Sample case: Test: Version A (old button) 5,000 visitors, 150 clicks. Version B (new button) 5,000 visitors, 190 clicks.
  “The new button was clicked more than the old one: out of 100 people about 3 clicked in A and about 4 in B. So roughly a quarter more people clicked. But this is the result of a single test; to be sure the difference is not down to chance it is a good idea to run it a bit longer or repeat it. For now there is a lean towards the new button, but it is too early to say for certain.”
  
  Maths: A 3%, B 3.8% (190/5000). Relative increase about 27%. A statistical test is needed to confirm significance.
  This sample output was produced with Claude; your result may differ.
• Trends and likely causes: Looking at the following data table, describe 3 notable trends and a possible explanation for each. State clearly where you are unsure: [data]
  Sample case: Data: Weekly visitors: week 1 900, week 2 950, week 3 1,800, week 4 1,750, week 5 1,700.
  1. In week 3 visitors almost doubled (950 to 1,800). Possible explanations: a post went viral, ads started or a site published a link. I cannot know which from the data; check the traffic source report.
  2. The rise looks lasting: weeks 4 and 5 stay around 1,700-1,750. If it is lasting, a new baseline may have been reached.
  3. A slight dip (1,800 to 1,700): small, perhaps a settling effect. But with this little data it is too early to call it a trend.
  
  Where I am unsure: why week 3 rose and whether the dip will continue.
  This sample output was produced with Claude; your result may differ.
• 5 beginner Excel formulas: Suggest 5 beginner-level spreadsheet formulas I can use for [data area] and explain each with an example.
  Sample case: Data area: product sales table (A: product, B: quantity, C: unit price, D: region).
  1. Total: =SUM(B2:B100) — adds up all quantities.
  2. Revenue per product: =B2*C2 — multiplies quantity by price; drag it down the rows.
  3. Conditional total: =SUMIF(D2:D100,"Istanbul",B2:B100) — adds only the quantities in Istanbul.
  4. Average: =AVERAGE(C2:C100) — gives the average unit price.
  5. Highest value: =MAX(B2:B100) — finds the largest quantity sold.
  
  Note: depending on Excel's language setting the separator may be a semicolon (;) or a comma (,).
  This sample output was produced with Claude; your result may differ.