Data analysis prompts

 

Okay, here are 3 examples of data analysis prompts, ranging in scope and domain, along with explanations for each:

Example 1: E-commerce Sales Performance Analysis

• Prompt: "Analyze the provided quarterly sales data (transactions including product ID, category, price, quantity, customer ID, region, date) for our online retail store. Identify the top 3 performing product categories by revenue and the top 3 performing regions by number of unique customers. Are there any significant trends in overall sales volume during the quarter? Provide a brief summary of findings and one recommendation for the marketing team."

• Explanation:

  • Goal: To understand recent sales performance by identifying key drivers (top categories, active regions) and spotting temporal patterns (trends).

  • Data: Transactional sales data, likely in a tabular format (CSV, database table).

  • Analysis Tasks: Requires data aggregation (summing revenue by category, counting unique customers by region), potentially time-series analysis (plotting sales over time), filtering, and sorting.

  • Output: A summary report highlighting the best-performing segments and any observed trends, culminating in a specific, data-backed recommendation (e.g., "Focus marketing spend on Region X as it has the most engaged customer base," or "Promote Category Y as it's driving the most revenue"). This prompt is typical for business intelligence or marketing analytics roles.

Example 2: Website User Engagement & Conversion Funnel

• Prompt: "Using the website analytics data from the past 30 days (including session duration, pages per session, bounce rate, page URLs visited, and conversion events like 'sign-up'), analyze the user journey for visitors who land on our main product feature page. What percentage of these visitors eventually sign up for a free trial? Identify the top 3 pages users navigate to after the feature page but before signing up, and determine the drop-off rate at each step. Suggest one hypothesis for why users might be dropping off at the point of highest friction."

• Explanation:

  • Goal: To understand user behavior on the website, specifically focusing on the path towards a key conversion (free trial sign-up) and identifying bottlenecks.

  • Data: Web analytics data (e.g., from Google Analytics, Adobe Analytics), often involving user sessions and page flows.

  • Analysis Tasks: Requires filtering data for specific user segments (those visiting the feature page), calculating conversion rates, performing funnel analysis (tracking users step-by-step through a defined path), calculating drop-off percentages, and interpreting user behavior to form hypotheses.

  • Output: A quantitative analysis of the conversion funnel, highlighting key navigation paths and points where users abandon the process. The output includes an educated guess (hypothesis) about the cause of the biggest drop-off, which could inform future A/B testing or website design changes. This is common in web analytics, product analytics, or UX research.

Example 3: Customer Support Ticket Analysis

• Prompt: "Analyze the customer support ticket dataset covering the last six months (fields include ticket ID, creation date, resolution date, issue category, priority level, customer satisfaction score (CSAT)). Calculate the average resolution time overall and break it down by issue category and priority level. Is there a correlation between resolution time and the final CSAT score? Identify the issue category with the lowest average CSAT score and suggest one potential area for investigation."

• Explanation:

  • Goal: To evaluate the efficiency and effectiveness of the customer support process and identify areas needing improvement.

  • Data: Customer support ticketing system data.

  • Analysis Tasks: Requires calculating time differences (resolution time), calculating average metrics (resolution time, CSAT), segmenting data (by category, priority), performing correlation analysis (between resolution time and CSAT), and identifying outliers or underperforming segments.

  • Output: Key performance indicators (KPIs) for the support team, insights into how different factors (issue type, priority) affect performance and satisfaction, and identification of the most problematic area based on customer feedback (CSAT). The final suggestion points towards further, more specific investigation (e.g., "Investigate training for support agents handling 'Billing Issues' as this category has the lowest CSAT"). This type of prompt is relevant for operations analysis or customer success roles.