Read and interpret Quantify results
Start with Enterpret Agent to answer a feedback question or prepare a report. Describe the measure, scope, and period you need. For recurring reporting, daily summaries, or weekly digests, use scheduled Agent automations; add connectors for external context or supported actions. Use the Quantify controls below when you need to build or inspect a specific chart or calculation.
Use this guide after building a Quantify analysis. For setup, see Build your first Quantify analysis. To change the query itself, use the Quantify query reference.
Read the query before the chart
Check Plot, Filter, Duration, and Measured as. A count of feedback records, a count of accounts, and the mean of account ARR answer different questions, even when they use the same Plot field.
For Percentage, read the denominator description and filters. Categories can overlap: the same feedback record, user, or account can appear in several groups. Group counts are not necessarily additive, and displayed percentages need not sum to 100%. A percentage can exceed 100% when its denominator describes a smaller or different population.
Choose a compatible view
Bar: compare the measured values for categories.
Trend + Anomalies: inspect a series over time, when supported by the query.
Table: inspect result values and available record links.
Other offered views can include Pie, Metric, Stacked Bar, or Tree Map. The available set depends on the grouping, measured field, calculation, and Breakdown By. Compare uses bar results; some user/account measures do not offer Trend.
Select the values to display
A new analysis normally starts with Top 10 values. The Top menu offers Top 5, Top 10, and Top 15. Templates and saved analyses can retain another selection. “Top” follows the current result ordering and measure; it does not always mean the highest raw feedback count.
You can also select rows in the result table. Bar and Trend charts support up to 15 manually selected values; follow the limit shown for other views.
Changing the selected rows changes what is drawn in the chart. Use Filter to change the analysis population, and inspect the Percentage denominator separately. Hiding a category does not remove its underlying data from that denominator.
Inspect the supporting records
Click a linked result value, or choose View Feedback from a chart point when offered. Check the group, filters, comparison side, and time bucket used by the record view. Some query types open other record types or use a different action label; zeros and disabled values may not offer a link.
Investigate anomaly flags
For a supported grouped Trend + Anomalies analysis, turn on Show Anomalies to inspect flags for the latest completed day, week, or month. For Percentage, detection uses the underlying numerator aggregate rather than the displayed percentage rate; a flag alone does not establish that the rate is unusual. The detection window can differ from the chart's Duration. Use the threshold settings when offered, then choose Apply after changing the threshold. See Anomalies in Quantify: setup and interpretation for details.
Review the observed and expected values and inspect the supporting records. A flag is a prompt to investigate a change, not a finding of its cause or proof that a change is statistically significant.
When saving an anomaly analysis, check whether you selected fixed values or Top values. Fixed selections remain tracked and may no longer be anomalous in later results.
Check comparisons and share the result
Before interpreting a difference, compare like-for-like populations and measures. Check the durations, filters, denominators, and relevant data coverage on both sides. Changes in feedback volume or linked customer data can affect the result.
Use Copy Link to share a saved analysis with teammates who have access. To export, open the more menu and choose an option offered for the result, such as Export Table as CSV or Export Chart as Image. Export Feedback as CSV, when available, exports underlying feedback rather than the summary table.
For saving an unsaved analysis or adding it to a dashboard, follow Build your first Quantify analysis.