A portfolio SROI ratio combines compatible estimates of social value and inputs. It is not the average of whatever ratios partners happen to report. Before combining them, establish the reporting boundary, check how each estimate was made and show which part of the portfolio has usable evidence.
This optional lesson is for a portfolio or reporting lead who already has reviewed value maps. You will build a consolidation table and decide whether one combined ratio is justified. If the underlying models cannot be reconciled, the useful result is a transparent comparison of separate analyses.
Define what the portfolio number would describe
Start with the decision and scope. Does the report concern all program resources, a defined group of funded activities or the investor’s own capital? A program-level value estimate cannot be divided by only one funder’s contribution and automatically become that funder’s return.
Record the included entities, activities, period, currency, valuation date and input boundary. Distinguish a forecast from an analysis of observed outcomes. Decide how negative outcomes, missing evidence and overlaps will be handled before looking at which combination produces the most attractive number.
Check compatibility before combining values
| Check | What to inspect |
|---|---|
| Outcome boundary | Which stakeholders and changes are counted, including adverse outcomes? |
| Time and money | Are periods, currencies, price bases, duration and discounting compatible? |
| Inputs | Do the denominators include comparable resources and avoid internal transfers counted twice? |
| Evidence and adjustments | How were quantities, proxies, deadweight, attribution and displacement established? |
| Overlap | Could two partners claim the same outcome for the same people and period? |
Partners need not collect every local question in the same form. Ask for the shared fields required to understand and reconcile the models. Preserve the local value map and its assumptions. A data dictionary can map compatible fields; it cannot make incompatible valuation methods equivalent by changing their labels.
Use the underlying values and inputs
Consider two fictional, compatible analyses with no overlapping outcomes or inputs:
| Activity | Present value of outcomes | Inputs | SROI ratio |
|---|---|---|---|
| A | $300,000 | $100,000 | 3:1 |
| B | $100,000 | $50,000 | 2:1 |
| Combined assessed scope | $400,000 | $150,000 | 2.67:1 |
The combined ratio is $400,000 ÷ $150,000. An equal average of 3 and 2 gives 2.5 and does not represent this combined scope. Weighting each ratio by its corresponding inputs gives the same 2.67 result, but only when the ratios and denominators are compatible and overlap has been resolved.
Keep present value of outcomes separate from net present value after subtracting inputs. In this example the latter is $250,000. Dividing that by inputs produces a different measure. Label the numerator and formula so reviewers know which one the report presents.
Make missing coverage visible
Now add activity C with $50,000 of inputs and no usable outcome valuation. The assessed activities represent $150,000 of the portfolio’s $200,000 inputs: 75% input coverage. That is not 75% participant coverage or 75% evidence quality.
The 2.67:1 ratio still describes A and B only. Calling it the whole portfolio’s ratio would hide C. Record C as unassessed and report its inputs alongside the assessed scope. Do not silently assign C a zero outcome value or invent a ratio.
You may show an explicitly labeled scenario that includes all $200,000 of inputs while using only the $400,000 of assessed value. That yields 2:1, but it is not an established portfolio estimate or necessarily a lower bound: C may have positive or negative outcomes, and other model uncertainty remains.
Resolve overlaps without losing the original reports
Suppose two partners both include the same employment outcome for a participant they jointly supported. Adding both values could double count the change. Identify the overlapping outcome, person or relevant unit, period and valuation basis. Agree how it is represented at the consolidated level and retain a record of the adjustment.
Do not assume a persistent person ID alone resolves this. Two outcomes for the same person may be distinct, while differently identified records may concern the same outcome. Matching rules, purpose, permitted access and human review matter. Where individual linkage is unavailable or inappropriate, document the uncertainty and use a justified aggregate treatment.
Likewise, an investor providing 30% of funding has not thereby proved responsibility for 30% of the social outcome. Funding allocation and causal contribution answer different questions. Avoid turning a financing share into an attribution claim.
Show the components beside the headline
A reviewable output includes each activity’s inputs, outcome value, model period, evidence status and contribution to the assessed total. Add coverage, overlap adjustments and assumptions that most affect the result. Do not turn these into an investee league table when the underlying populations and purposes differ.
Test sensitivity to material assumptions such as duration, proxy choice or contribution adjustments. Keep the report version and calculation inputs fixed for an issued report. Later evidence should create a new reviewed version rather than silently rewriting what the board previously saw.
Organize the recurring workflow
In Sopact, the practical starting point is a structured return attached to the right partner and reporting period, with the supporting model or document. Define the small shared data dictionary, preserve source references and identify missing or incompatible fields before aggregation.
Configured analysis can help extract proposed values and locate explanations for review. The approved calculation should use explicit formulas and reviewed inputs. The platform does not make a partner’s valuation valid simply by receiving it. Assign an owner for reconciliation and approval before sharing the report.
Practice: prepare a consolidation decision
- Select two actual or sample value maps.
- Write their reporting boundaries and list compatibility gaps.
- Build the value-and-input table, including unassessed activities.
- Check possible double counting and document the treatment.
- Conclude either “combine within this scope” or “report separately,” with the reasons.
Use the resulting evidence table when preparing your report. The impact-report writing guide and report examples can help you present the scope, findings and limits clearly.
Frequently asked questions
Can we average the SROI ratios supplied by partners?
Not without examining the models. Use compatible underlying outcome values and inputs, resolve overlaps and show coverage. An input-weighted average is equivalent only under those conditions.
What happens to an unmeasured activity?
Keep it visible as unassessed with its known inputs and evidence gaps. State precisely which activities the reported ratio covers.
Is a combined ratio always the best portfolio report?
No. Separate outcome and value-map summaries can be more useful when methods, populations or reporting boundaries cannot reasonably be reconciled.