How to Use This List
Each example includes:
- The exact question you can type (adjust names, dates, and metrics to match your data)
- What the AI typically returns
- Tips for getting the best results
Copy any question, replace the specific references with your own column names and date ranges, and paste it into the conversational analytics interface.
Sales Questions (1-5)
1. "What was our total revenue for the last 3 months, broken down by month?"
Returns: A bar or line chart showing monthly revenue with exact figures. A written summary with month-over-month growth rates.
Tip: If your data has a currency column, the AI will detect and format it automatically. Specify the date range explicitly for more accurate results.
2. "Which 10 customers generated the most revenue this year?"
Returns: A horizontal bar chart ranking the top 10 customers by revenue. A data table with customer names and revenue figures.
Tip: Replace "customers" with whatever your data calls them (accounts, companies, clients). Add "and what percentage of total revenue do they represent?" for a deeper answer.
3. "Compare our Q1 and Q2 sales performance by product category"
Returns: A grouped bar chart showing Q1 vs. Q2 for each category. Written analysis highlighting which categories grew, shrank, or stayed flat.
Tip: The AI understands quarter references (Q1, Q2, etc.) relative to the current or most recent year in your data.
4. "What is our average deal size, and how has it changed over the last 12 months?"
Returns: A line chart showing average deal size by month. A metric showing the current average and the trend direction.
Tip: If deal size varies by segment, follow up with "Break that down by customer segment" for more detail.
5. "Which sales rep has the highest close rate this quarter?"
Returns: A ranked table of sales reps by close rate (deals won / deals total). May include a bar chart visualization.
Tip: Make sure your data includes both won and total deal counts per rep. If your data tracks stages, the AI can calculate close rates from stage transitions.
Marketing Questions (6-10)
6. "What are our top 5 traffic sources by sessions this month?"
Returns: A bar chart or pie chart of top traffic sources. Session counts and percentage of total for each source.
Tip: Works best with Google Analytics data connected. The AI maps standard GA dimensions automatically.
7. "Show me the conversion rate trend from website visits to signups over the last 6 months"
Returns: A line chart showing monthly conversion rates. Written analysis of whether the rate is improving or declining.
Tip: Define what counts as a "conversion" if your data has multiple event types. Be explicit: "conversion rate from page_view to signup_completed."
8. "Which marketing campaign had the best ROI last quarter?"
Returns: A ranked list of campaigns by ROI (return on investment), showing spend, revenue generated, and ROI percentage.
Tip: The AI needs both spend and revenue data to calculate ROI. If they are in separate data sources, connect both and specify which source has which metric.
9. "What is our cost per acquisition by channel?"
Returns: A bar chart comparing CPA across channels (paid search, social, email, organic, etc.).
Tip: CPA = total spend / number of acquisitions. Ensure your data includes both spend and acquisition counts per channel.
10. "How does our email open rate compare to industry benchmarks?"
Returns: A written comparison with your open rate and typical industry ranges. The AI uses its training knowledge for benchmark data (it does not query external sources).
Tip: The benchmark data comes from the AI's general knowledge and may not reflect your specific industry. Use it as a rough reference point.
Finance Questions (11-15)
11. "What is our monthly burn rate over the last 6 months?"
Returns: A line chart of total expenses per month. A metric showing the average monthly burn and the trend.
Tip: Burn rate typically means total operating expenses. If you want a specific definition (net burn, gross burn), include it in the question.
12. "Break down our operating expenses by category for this quarter"
Returns: A pie chart or treemap showing expense categories and their proportions. A data table with exact amounts.
Tip: If your chart of accounts is granular, add "group into the top 10 categories and combine the rest as Other" for a cleaner visualization.
13. "What is our gross margin percentage, and how has it trended over the last year?"
Returns: A line chart of gross margin percentage by month. A metric showing the current margin and year-over-year change.
Tip: Gross margin = (revenue - cost of goods sold) / revenue. Make sure both revenue and COGS are in your data.
14. "Show me accounts receivable aging -- how much is 30, 60, 90, and 90+ days overdue?"
Returns: A stacked bar chart or table showing AR aging buckets with dollar amounts.
Tip: Your data needs invoice dates and payment status. If the aging calculation is not in your data, the AI can compute it from invoice dates and the current date.
15. "What is our revenue per employee, and how does it compare to last year?"
Returns: A metric showing revenue per employee with year-over-year comparison. May include a bar chart if employee count has changed significantly.
Tip: This requires both revenue data and employee count data. If they are in different sources, specify which source has which.
Operations Questions (16-20)
16. "What is our average order fulfillment time, and which warehouse is fastest?"
Returns: A bar chart comparing average fulfillment time by warehouse. A metric showing the overall average.
Tip: Fulfillment time = ship date - order date. The AI can calculate this if your data has both timestamps.
17. "Show me inventory levels by product category, highlighting any items below reorder point"
Returns: A table of inventory levels with conditional formatting (red for below reorder point). May include a bar chart.
Tip: Your data needs current stock levels and reorder point thresholds. If reorder points are not in the data, specify them in the question: "flag anything below 100 units."
18. "What is our on-time delivery rate by shipping carrier?"
Returns: A bar chart comparing on-time rates by carrier. A written summary identifying the best and worst performers.
Tip: On-time = delivered by estimated delivery date. Ensure your data includes both actual and estimated delivery dates.
19. "How many support tickets did we receive each day last week, and what was the average resolution time?"
Returns: A dual-axis chart showing ticket volume (bars) and average resolution time (line) per day.
Tip: Works well with Jira, Zendesk, or any support system data. Follow up with "Break that down by priority" for more insight.
20. "What is our system uptime percentage for the last 30 days?"
Returns: A metric showing uptime percentage. May include a timeline showing downtime incidents.
Tip: Your data needs timestamps of outage events. If tracking uptime as a percentage directly, a simple metric widget may be more appropriate than conversational analytics.
HR Questions (21-25)
21. "What is our employee turnover rate by department this year?"
Returns: A bar chart comparing turnover rates across departments. Written analysis highlighting departments with unusually high or low turnover.
Tip: Turnover rate = separations / average headcount. Ensure your data includes both termination events and headcount figures.
22. "Show me the headcount trend over the last 24 months"
Returns: A line chart of total headcount by month, showing growth or contraction patterns.
Tip: Follow up with "Break that down by department" or "Show new hires vs. terminations separately."
23. "What is the average time to fill open positions by role type?"
Returns: A bar chart comparing time-to-fill across role categories (engineering, sales, marketing, etc.).
Tip: Time to fill = offer accepted date - job posted date. Your ATS (applicant tracking system) data needs both dates.
24. "How is our training completion rate trending across the organization?"
Returns: A line chart of training completion rates over time. May segment by department if the data supports it.
Tip: Specify what counts as "completion" if your training system tracks partial progress vs. full completion.
25. "What is the gender and tenure distribution across management levels?"
Returns: A grouped bar chart or stacked chart showing the distribution. Written summary of any notable patterns.
Tip: Sensitive demographic data should be handled carefully. Ensure your organization's data privacy policies allow this analysis.
Product Questions (26-30)
26. "What are the most-used features in our product over the last 30 days?"
Returns: A ranked bar chart of features by usage count or unique users. A data table with exact numbers.
Tip: Requires product analytics data (event tracking). The AI works with whatever event schema your data uses.
27. "Show me the daily active users trend and highlight any anomalies"
Returns: A line chart of DAU with anomalous days called out in the written analysis. The AI identifies statistical outliers automatically.
Tip: "Anomaly" means a value that deviates significantly from the recent pattern. The AI uses the surrounding data context, not a fixed threshold.
28. "What is our user retention rate at 7, 14, and 30 days?"
Returns: A cohort-style table or bar chart showing retention at each interval. Written explanation of the retention curve.
Tip: Retention analysis requires user signup dates and activity dates. Specify the time period for the cohort: "for users who signed up in January."
29. "Which features are most correlated with user retention?"
Returns: A written analysis identifying features that retained users tend to use more often than churned users. May include a bar chart of feature usage rates by retention group.
Tip: This is a correlation question, not causation. The AI's response will note this distinction. Useful for generating hypotheses to test.
30. "Compare our mobile and desktop user behavior -- sessions, pages per session, and conversion rate"
Returns: A grouped bar chart comparing mobile vs. desktop across the three metrics. Written analysis of the differences.
Tip: Works especially well with Google Analytics data where device category is a standard dimension.
Customer Support Questions (Bonus)
These work well if you have connected Jira, Zendesk, or another support system.
"What is the average first response time by support tier this month?"
Returns: A bar chart comparing response times across support tiers (L1, L2, L3). Written analysis of bottlenecks.
Tip: Define "first response time" clearly if your data tracks multiple response types. "Time from ticket creation to first agent reply" works well.
"Which support categories have the longest resolution times?"
Returns: A horizontal bar chart ranking categories by average resolution time. A table showing volume and resolution time for each category.
Tip: Follow up with "What percentage of those are escalated?" to dig into the root cause.
"Show me the CSAT trend over the last 12 months and flag any months below 80%"
Returns: A line chart of monthly CSAT scores with a reference line at 80%. The AI highlights months below the threshold in the written analysis.
Tip: If your data includes individual survey responses, the AI can break CSAT down by category, agent, or channel.
Executive-Level Questions (Bonus)
These are designed for high-level dashboards and C-suite reporting.
"Give me a month-over-month summary of our top 5 KPIs"
Returns: A table or multi-metric card showing each KPI with its current value, previous month value, and change percentage.
Tip: List the specific KPIs you care about for more focused results: "revenue, active users, NPS, churn rate, and gross margin."
"What are the three biggest risks in our data right now?"
Returns: A written analysis highlighting anomalies, declining trends, or metrics approaching critical thresholds. May include supporting charts.
Tip: This is an open-ended question, so results depend on what data is connected. The AI looks for negative trends, sudden drops, and values outside normal ranges.
Tips for Getting Better Answers
- Be specific about dates. "Last month" is better than "recently." "January 2025" is better than "a while ago."
- Name your metrics. "Revenue" is clearer than "the numbers" or "how we are doing."
- Ask one question at a time. Multi-part questions sometimes produce partial answers. Ask them separately, then synthesize.
- Use follow-ups. The AI remembers context within a conversation. "Break that down by region" after a revenue question works well.
- Check the confidence score. If it is below 70%, rephrase the question or check that the required data is available.