On this pageUnderstand the individual survey

Client Admin

Analytics

Client Admin access is required for these organization-facing workflows. Each signed-in user completes their own survey; Client Admin role and analytics views use survey-derived and aggregate organization signals to support enablement, not to make a judgment about an individual.

AvailableLast reviewed: Oct 21, 2026Audience: Client Admin

Understand the individual survey

The survey is an individual signed-in flow. Each user completes their own survey when the product directs them there; Client Admins do not complete responses for other people.

  1. Ask the user to sign in and open their own survey when the product directs them there.
  2. Before the user starts, review the purpose and response context shown in the survey.
  3. Explain what the survey is for without promising a result or visibility that the product does not state.
  4. The user completes and submits their own responses; do not submit or edit responses for someone else.
  5. Use Admin Analytics for aggregate organization signals after data is available; analytics does not replace the individual's survey.

Note: Do not promise a level of anonymity or visibility that the product's current survey context does not state.

Read reporting states

Admin Analytics combines survey-derived and aggregate organization signals. Its current page states that sections other than AI Usage aggregate all-time; visible time controls apply only where the product presents them, such as AI Usage. Withheld means a qualifying result is intentionally not shown for privacy protection. No data means there are no qualifying records. Neither label means zero, and prompt content is not included in reporting.

  • Start with the least detailed view that answers the question.
  • Read the metric definition and any control that is visibly shown before comparing two results.
  • Treat an aggregate pattern as a prompt for enablement or content improvement, not as a diagnosis of a person.
  1. Open Admin Analytics and read the organization-level KPI, funnel, segments, and activity cards.
  2. Treat the main analytics sections as all-time unless the page visibly identifies a metric-specific control, such as the AI Usage time range.
  3. Read Withheld and No data exactly as labeled; do not replace either state with a zero.
  4. Use the least detailed result that supports the decision you need to make.

Note: AI Usage is the one exception to the Withheld protection above: it additionally shows a log of the most recent individual requests, each identified by user name, with no minimum-group suppression applied to that log. Treat it with the same care as any other personally identifiable activity record.

Respect privacy choices

Make your organization's privacy expectations clear and direct people to Profile → Privacy Controls when they need to review their own choices. Each control applies to the identity or contribution it names; do not infer that one choice changes every surface.

  • Profile visibility controls whether peers can view a profile.
  • Work-email opt-in controls whether a work email is shown in supported peer contexts.
  • Anonymity and per-item attribution are separate choices where the product supports them.
  • Manager-scoped reporting still follows active direct-report assignment and the visibility rules of the surface.

When a result is empty or unavailable

An empty result can mean that no qualifying records are available. An unavailable source is a separate failure state and should not be summarized as No data.

  1. If a card says No data, communicate that no qualifying data is available rather than converting it to zero.
  2. If the source is unavailable, choose Retry and wait for a confirmed result.
  3. If the result remains Withheld, keep the privacy-preserving label and avoid reconstructing the suppressed detail; use the aggregate view the product provides.

Tip: A confirmed result is one the product has loaded for the organization-level view or visibly selected metric control. Do not fill a blank card with an estimated value.

Use insight responsibly

Use surveys and reporting to choose useful learning support, content, or communication. Keep the decision at the same level of detail as the evidence and avoid turning an engagement pattern into a claim about ability or wellbeing.

  1. Name the enablement question the result is meant to answer.
  2. Use aggregate evidence and the smallest necessary audience for the next action.
  3. Communicate the action and its purpose without exposing a person's private response or identity choice.

79 public articles available