Calculations on Quantify
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.
Choose a calculation in Measured as to determine what a Quantify result counts or calculates. Then choose the object or numeric field it applies to. For the query controls, see the Quantify query reference.
Plot determines the result groups. Filter and, where applicable, Duration determine the included data. The calculation and measured object determine the meaning of each number.
Count
Count counts distinct IDs of the selected object within each result group. Feedback records, users, and accounts are different measured objects; choose the one that answers your question.
For example, three feedback records associated with two accounts can produce a feedback Count of 3 and an account Count of 2 in the same group.
Groups can overlap. A record or account associated with more than one topic can appear in several groups, so adding the displayed group counts may exceed the distinct total.
Percentage
Percentage divides a group's count by the configured denominator for the measured object and expresses the result as a percentage. Read the denominator description and its filters; do not assume it always means all feedback matching the main Filter.
Use the denominator's edit control when offered. Review its population, object, and time range alongside the numerator, especially when comparing two queries.
For example, 20 matching feedback records divided by a denominator of 100 feedback records is 20%. That interpretation depends on the denominator actually describing those 100 records.
Groups need not be mutually exclusive, so displayed percentages need not add to 100%. A separately configured denominator can also describe a smaller or different population; check it before interpreting an unusually large percentage.
Sum and Mean
| Calculation | Meaning |
|---|---|
| Sum | Adds the included numeric values of the selected field. |
| Mean | Averages the available numeric values of the selected field. A missing value does not contribute a zero or an extra value to the average. |
For a single-valued numeric field with values 10, 20, and one missing value, Sum is 30 and Mean is 15. Dividing by all three feedback records would give 10, which is not the mean of the available values.
A field with multiple numeric values on one record does not necessarily represent one numeric input per feedback record. Confirm the field's structure before reconciling it to a response-level report.
Calculations on users and accounts
When linked user/account fields are available for the query, select the appropriate object or numeric attribute in Measured as. For Sum and Mean of a single-valued user/account numeric attribute, each entity contributes once within each result group.
For example, if several feedback records from the same account mention one topic, that account's single ARR value contributes once to the topic's Sum of ARR. The account can still contribute to another topic's group.
Use a single intended numeric value per user or account for these comparisons. If an attribute contains several numeric values per entity, confirm the field mapping with your Enterpret team before interpreting the aggregate.
NPS, CSAT, and DSAT
These score calculations require a suitable numeric field and the expected rating scale. The formulas below describe correctly populated, single-valued rating responses on that scale.
| Calculation | Expected scale | Formula |
|---|---|---|
| Net Promoter Score (NPS) | 0–10 | Percentage of promoters (9–10) minus percentage of detractors (0–6). Ratings 7–8 are passives and remain in the response population. |
| Customer Satisfaction Score (CSAT) | 1–5 | 100 × ratings of 4 or 5 ÷ rated responses. |
| Customer Dissatisfaction Score (DSAT) | 1–5 | 100 × ratings of 1 or 2 ÷ rated responses. |
For valid 1–5 responses of 5, 4, 3, 2, and 1, CSAT is 40% and DSAT is 40%. These use rating responses, not every feedback record in the query.
Check missing values, out-of-range values, and the number of rating responses before interpreting a score. Do not assume an available calculation validates or cleans every source value. A score of zero alone does not establish that rating responses were present.
When a calculation is unavailable
The available choices depend on the Plot field, measured object, numeric type, expected score scale, and other query settings. A numeric-looking value stored as text can prevent a numeric calculation from appearing. Query features such as Breakdown By can also restrict the available measured objects.
Review the message shown by the control. If a field is missing, has the wrong type, or uses an unexpected scale, contact your Enterpret team with the query and field name.
For chart selections, record inspection, and comparisons, see Read and interpret Quantify results. For a complete setup walkthrough, see Build your first Quantify analysis.