What can our current evidence support?
- Board or funder question
- Annual report or important claim
Evidence synthesis
For foundation leaders
Fund practical support that a grantee chooses and directs, or bring Prova a foundation-owned question about a portfolio, initiative, evaluation, or funding decision. Prova combines frontier AI with proprietary evidence systems and accountable professional judgment. The systems connect important statements to sources and make gaps visible for professional review.
The first question is whose decision the work is meant to serve.
Recognizable situations
A grantee needs senior evidence or measurement support around a live decision.
Reporting burden has grown without proportionate learning value.
You need to decide which grants actually have a live decision before commissioning more evaluation.
Your team needs to learn across programs that use different models, populations, or measures.
A strategy or initiative needs a clearer account of what the record actually supports.
A consequential claim or evaluation needs independent review.
Making a personal or family giving decision? Explore private philanthropy
Establishing how well community investment is going, for the company? Explore corporate
Two ways to work
That choice determines who directs the work, which materials may be used, and who receives the answer.
Grantee decision or foundation decision?
Path 01 · Grantee decision
Result: work that helps the nonprofit make its own program decision.
Path 02 · Foundation decision
Result: a clear answer for a portfolio, initiative, evaluation, or funding decision.
Grantee-directed support
The foundation makes access possible. The nonprofit chooses whether to participate, sets the question and materials, and controls what may be shared. The offer is not a condition of funding.
Evidence synthesis
Theory of change
Measurement design
Program evaluation
A foundation's own question
Start with what you want to learn across the portfolio and how you would use the answer. Then check whether definitions, populations, contexts, and measures can responsibly be compared.
Before comparing programs
| Element | Question before comparison | Expected status |
|---|---|---|
| Decision | What decision will the comparison inform? | Can be common |
| Definition | Do programs mean the same thing by the measure? | Must be checked |
| Population | Are participants and eligibility conditions comparable? | Often differs |
| Context | Do operating and local conditions change interpretation? | Often differs |
| Measurement | Were data collected in sufficiently similar ways? | Must be checked |
See an illustrative Portfolio Learning Map and what you receive
Collaboration or independent review
If Prova helped build or substantially revise the work, Prova should not later present its judgment as independent.
Choose the role the decision requires
Two review types
Question, protocol, measures, sample, comparison, analysis plan, and implementation conditions.
Design ReviewThe claim, materials used, source locations, method, other plausible explanations, and limits.
Evidence ReviewTrust and data boundaries
The engagement defines what may be used and shared before information moves through the work.
A practical first step
A first conversation can clarify the smallest useful piece of work, who should direct it, whether independence is required, and what information may be shared.
Discuss what you need