Give Sopact Sense a program's page and it reconstructs the theory of change, grades every link by how much evidence backs it, and flags exactly where the logic is strong, thin, or missing — for pressure-testing a theory of change before a funder does.
In short: To audit a theory of change, give the program's page (or its text) to an AI and ask it to reconstruct the causal chain and grade every link by evidence — green where data backs it, amber where it rests on a belief, red where it's missing. Sopact Sense does this in minutes straight from a URL, marking anything it can't verify, so you see exactly which links are strong, weak, or absent.
An audit starts from what the program actually publishes — not what you assume. Point the Assistant at the program or proposal page and tell it to use only what's stated there:
You are the Sopact Sense Assistant. Audit the program/proposal page at [URL] (or I will paste its content). Use only what is stated on the page. Wait for my task.
Ask Sense to rebuild the theory of change from the page and grade every link — naming the assumption under each arrow and separating strong signals from weak links and outright gaps:
Read the program at [URL] and reconstruct its Theory of Change as a causal chain. Name the assumption under each arrow; list strong signals, weak links, and what's missing. Do not infer what the page doesn't say. Grade every element green / amber / red.
Five elements make the audit rigorous: the input (the live [URL] or its text); reconstruct the chain (inputs to impact); strong / weak / missing (evidenced vs assumed vs absent); no inference (it won't invent what the page doesn't say); and the grade (green / amber / red at a glance).
Sense returns the reconstructed chain with each element graded and the assumption named under every arrow — the strong links, the thin ones, and the gaps. The demo audits Bright Futures Initiative, engineered to grade one green, one amber, one red:
Bright Futures Initiative helps ~150 first-generation students each year. We run weekly tutoring, college-application workshops, and a summer bridge camp. Our goal is for students to enroll in college and persist to year two. Last year 128 of 150 students (85%) submitted at least one college application, tracked through our application portal. We believe consistent mentoring builds the confidence students need to persist. We have not yet measured year-two persistence, and we do not state a baseline college-enrollment rate.
GRADE: green | 85% applied | 128 of 150, tracked in the portal; amber | mentoring → confidence | claimed, not measured; red | year-2 persistence | no data, no baseline
The green link is carried by tracked data, the amber link rests on a stated belief, and the red link is missing entirely — no year-two persistence and no enrollment baseline.
The payoff of an audit is the fix. Take the lowest-graded element and make it measurable with one realistic change:
Take the lowest-graded element above and fix it using only what the program could realistically measure. Show the before → after grade and the single indicator/edit that moves it to green.
For Bright Futures, that's a year-two enrollment check against a stated baseline — turning the red persistence gap into something the program can actually show.
Generate a decision-first report in your own brand, then a shareable link:
Create a 'missing & incomplete' report from this analysis in Sopact branding [or paste your website URL / brand guideline to apply your own]. List every element graded amber or red, what is missing, and the one input that fixes each. Lead with the decision this report informs.
Create a shareable link for this report and open it in a new tab.
Audit from the live page. Pointing Sense at the program's URL keeps the audit grounded in what the program actually publishes — and surfaces the claims a funder would read first.
The red links are the payoff. An audit isn't about the tidy boxes; it's about the missing evidence. The red and amber links are exactly what to fix before the next proposal.
Re-audit after each fix. Add the one indicator Sense suggests, then re-run the audit next cycle and watch the grade climb.
Tighten the program page for accuracy. Once you've seen the grades, ask Sense to rewrite the page so its claims match the evidence:
Based on the grades above, suggest edits to the program page so its claims match the evidence. Flag every sentence that overstates what the program can show, and rewrite it to be accurate and specific.
A theory of change audit reconstructs a program's causal chain — inputs → activities → outputs → outcomes → impact — and grades each link by how much evidence supports it, separating what's proven from what's assumed or missing. It tells you whether the logic is credible, and where it isn't.
Check the assumption under every arrow, not just the boxes. A strong theory of change has evidence behind each causal link; a weak one rests on untested beliefs (“we believe mentoring builds confidence”) or leaps to long-term impact with no follow-up data. Grading each element green, amber or red makes the weak spots obvious.
Yes — give Sopact Sense the program's page URL and it reconstructs the theory of change from what's published, grades every link, and flags anything it can't verify, without inventing outcomes the page doesn't state. It takes minutes instead of a manual review.
Open Sopact Sense, paste your program description, and put it to work.
Try in Sopact
