Describe your program and Sopact Sense draws the causal chain inline — inputs through impact — and grades every link by how much evidence backs it. For when you know what you do but have never shown why it works.
In short: A theory of change maps your program as a causal chain — inputs → activities → outputs → outcomes → impact — and names the assumption under each link. To build one fast, describe your program (or paste its web page) to Sopact Sense; it drafts the chain and grades every link by evidence — green where data backs it, amber where it rests on a belief, red where there's no follow-up — so you can see exactly which step rests on faith. The steps below show how, with copy-paste prompts.
Most real-life theories of change start somewhere — implicit in how the team talks about the work, or scattered across a grant narrative. The boxes and arrows aren't where the work is; the assumptions under them are. The job is to draw the chain through to impact and mark where each link rests on a belief rather than evidence.
You don't write the theory of change — you write the prompt that builds it. First, set up the Assistant so it only uses what your program actually states:
You are the Sopact Sense Assistant. Here is my program description: [PASTE PROGRAM DESCRIPTION]. Use only what it states; mark anything you infer as [INFERRED]. Wait for my task.
Then the main prompt — the one that builds and grades the chain:
Build a Theory of Change for [PROGRAM] as a causal chain: inputs → activities → outputs → short / medium / long-term outcomes → impact. Name the assumption under each arrow, list the weak links, and suggest one indicator per assumption. Use only what the program states; mark anything inferred [INFERRED]. Grade every element green / amber / red (green = specific + evidenced, amber = vague or unevidenced, red = missing).
No written description handy? Point Sense at your program's public page instead — it reads the page and grades what it finds:
Build a Theory of Change from this program page: [PROGRAM URL — e.g. https://www.lanternnetwork.org/]. Read only what the page states; mark anything inferred [INFERRED]. Draw the causal chain inputs → activities → outputs → short / medium / long-term outcomes → impact, name the assumption under each arrow, list the weak links, suggest one indicator per assumption, and grade every element green / amber / red.
Five elements make it work: the input (your description or page, nothing else); the structure (a causal chain from inputs to impact); the rigor (name the assumption under each arrow — where most theories of change fail); no hallucination (mark anything inferred); and the score (green / amber / red, so weak spots show at a glance).
Sense returns the causal chain with every element graded, and the assumption written under each arrow — so the first time you read the theory out loud, you can see exactly which step is doing the heavy lifting on faith. The demo runs on Vista Workforce Collaborative, engineered to grade one green, one amber, one red:
Vista Workforce Collaborative runs a 12-week coding bootcamp for ~60 low-income adults per cohort. Activities: 200 hours of instruction, weekly 1:1 mentoring, and an employer demo day. In the latest cohort, 41 of 58 graduates (71%) were placed in a tech job within 6 months, verified by employer letters. We believe the 1:1 mentoring builds the confidence that keeps participants engaged. We do not yet track whether participants stay employed beyond the first placement or whether wages grow over time.
GRADE: green | 71% placed | 41 of 58, verified by employer letters; amber | mentoring → confidence | claimed, not measured; red | long-term wages | no tracking yet
The green link is carried by data, the amber link rests on a stated belief, and the red link has no follow-up at all — the longest leap in the chain.
This is the aha moment — fix the lowest-graded element using only what the program could realistically measure:
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 that moves it to green.
For Vista, that's a short confidence scale at intake and exit — enough to turn the amber mentoring assumption into something you can evidence.
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 would fix each. Lead with the decision this report informs.
Create a shareable link for this report and open it in a new tab.
Name the weakest link out loud. Asking “where are the weak links” gets you the assumptions under the arrows rather than a tidy diagram. The most useful answer is almost always the red link — the one leap that rests on a belief with no follow-up.
Check the chain against your data. If your program already collects responses in Sopact Sense, point the Assistant at the survey instead of describing the program. It reads the cleaned open-text and tells you which links your own respondents already support.
Turn one amber link green per cycle. Pick the amber link closest to your outcome, add the one indicator Sense suggests at intake and exit, and re-run the report next cycle to watch it turn green.
Tighten your program page while you're here. Once Sense has graded the chain, ask it to improve your program page content for accuracy — flagging any claim the evidence doesn't yet support and rewriting it to match what you can actually show:
Based on the grades above, suggest edits to my program page so its claims match the evidence. Flag every sentence that overstates what we can show, and rewrite it to be accurate and specific.
The same prompts work for a Logic Model, Logframe, or Results Framework — just swap “Theory of Change” in the main prompt.
A theory of change is a causal map of how a program creates impact — inputs → activities → outputs → short, medium and long-term outcomes → impact — with the assumption behind each link made explicit. It explains not just what you do, but why you believe it works.
Describe your program in a few sentences (or paste its web page) and ask the AI to draw the causal chain, name the assumption under each arrow, flag the weak links, and grade every element green, amber or red by how much evidence supports it. In Sopact Sense this takes minutes and stays grounded only in what your program actually states — it marks anything inferred so it never invents outcomes.
Weak theories of change fail at the assumptions, not the outcomes. The weakest links are the ones stated as beliefs — “we believe mentoring builds confidence” — with no indicator, or long leaps to long-term impact with no follow-up data. Naming and measuring those assumptions is what makes the logic credible to a funder.
Open Sopact Sense, paste your program description, and put it to work.
Try in Sopact
