
Health Measures
Evaluating the business value of an owned community
Assess member outcomes, support work, customer relationships and full operating costs before deciding what an owned community is worth.
Evaluate an owned community against the member needs it serves, the business outcomes it may contribute to and the work required to run it. Membership and activity counts provide context; they are not a valuation. Start with the investment decision the evaluation must inform.
Set the decision before choosing metrics
State whether the organisation is deciding to continue funding the current approach, change how staff spend time, expand a useful service or consider another home for members. Set the period under review and say what evidence would change the decision.
Write the member promise beside the business aim. If the promise is practical help, include unanswered questions. If it is peer learning, examine what participants could apply. A business benefit that depends on an experience members are not receiving is unlikely to last.
| Value question | Evidence to assemble | Limit to keep visible |
|---|---|---|
| Are members getting help? | Reviewed questions, confirmed outcomes, open cases and private handoffs | A reply or selected answer may not settle the problem. |
| Is support work changing? | Comparable contact volumes, reasons for contact and verified resolutions | A visit without a ticket does not prove a ticket was avoided. |
| Are customer relationships changing? | Customer cohorts, renewal or repeat-purchase outcomes, and relevant member accounts | Customers who participate may differ from those who do not. |
| Is the work sustainable? | Staff hours, platform and content costs, volunteer reliance and unresolved work | A low cash bill can hide substantial effort. |
Build a record that can survive scrutiny
Define who is included in each measure and when an outcome is assessed. Keep customers, member accounts, visitors, support requests and discussion threads as separate units. One customer may have several users; one question may generate several contacts. State how staff and test accounts are handled. Record changes to access, reporting and staffing so a measurement change is not mistaken for a change in value.
Use community, support and customer records together only where the organisation has an appropriate basis.
A public post does not provide general permission to join every dataset. Use the least identifying information that answers the question and involve the organisation’s privacy owner when linking records.
Pair counts with case review. Read successful and unsuccessful exchanges, including quiet threads. Ask whether the original need was met, what staff did and what happened after a handoff. Record unknown outcomes as unknown. Keep a dashboard figure’s definition and channel coverage beside it.
For an organisation covered by the Australian Privacy Principles, personal information may generally be used for its primary collection purpose, or for a secondary purpose where an exception applies. Examples include the individual’s consent; the individual’s reasonable expectation of a related purpose, or, for sensitive information, a directly related purpose; or use required or authorised by Australian law. Record the basis relied on when linking information for evaluation.
Key Metrics for Evaluating an Owned Community
- Member UnitsCustomers, member accounts, visitors, support requests, discussion threads
- Privacy Compliance BasisAustralian Privacy Principles (APPs) – consent, reasonable expectation, directly related purpose, or legal requirement
- Deflection Data CoverageZendesk self-service data from 5 March 2025; events from 4 April 2025 show only help centre origin
- ROI DefinitionReturn on Investment (ROI) = (Gain from investment – Cost of investment) / Cost of investment
Separate an observed outcome from a claimed contribution
An observed difference may justify investigation without proving that the community caused it. A participating customer may already be more committed. A fall in support contacts may reflect a product fix, a change in customer numbers or a new contact route. Consider these explanations before assigning monetary value.
For consequential decisions, choose an evaluation approach suited to the question and state what it can and cannot establish. Detailed comparison methods belong in the supporting article.
Use a financial estimate only where the underlying outcome is credible. Count support work that plausibly disappeared. Do not multiply every answer view by the cost of a ticket. Treat retention as an influenced outcome only to the extent the comparison supports that wording. Use the organisation’s agreed costs and customer economics. Keep unverified benefits separate.
Check the coverage and definitions behind any analytics used in the decision. Zendesk states that its ticket-deflection reports track only tickets submitted through the help centre request form. Its self-service data is available from 5 March 2025. From 4 April 2025, those datasets show only events originating from the help centre. Note those boundaries when comparing periods or interpreting a deflection figure.
Grade the strength of the comparison
Record the evaluation design used and its limitations. Detailed guidance on selecting and applying comparison methods belongs in the supporting article.
Make the funding decision explicit
Bring member outcomes, business outcomes, full operating cost and evidence limits into one decision record. Say what to continue, what to change, who owns the change and when it will be reviewed. A community may provide dependable help even when its return on investment cannot be estimated credibly. It may also be busy while failing its stated purpose.
If questions are answered but private support work is rising, inspect the handoff. If a small group receives clear value but the current space is expensive, examine a redesign or merge. If members cannot get the promised help and the team cannot support a workable change, plan a responsible closure.
In this guide
- Measuring support deflection without hiding unresolved questionsSeparate confirmed self-service resolutions from assumed deflections, support handoffs and unresolved questions when assessing community support value.
- Assessing community influence on customer retentionDefine customer retention, align exposure and follow-up windows, and compare like customers before claiming that a community influenced renewal.
- Comparing community costs with observed member outcomesBuild a full community cost ledger, count confirmed member outcomes and distinguish operating cost from spending a change would actually avoid.
- Deciding whether to merge, redesign or close a communityCompare merge, redesign and closure against member need, operating capacity and transition risks, then give open questions a clear handoff.



