play icon for videos
← Academy
Module 2 of 5 · Case intelligence

Read notes and check-ins without losing the person’s meaning

Make notes useful by preserving who said or observed what, when it happened and what remains uncertain. Review evidence against the question at hand rather than turning every phrase into a score. AI can organize material and suggest themes, but staff must verify interpretations and own consequential decisions.

<!DOCTYPE html> <html lang="en"> <head> <meta charset="utf-8"> <meta name="viewport" content="width=device-width, initial-scale=1"> <style> @keyframes pulseDot{0%,100%{opacity:1}50%{opacity:.35}} *{box-sizing:border-box} body{margin:0;font-family:'Hanken Grotesk',-apple-system,BlinkMacSystemFont,sans-serif;background:#EEF2F9;padding:26px;display:flex;justify-content:center} .card{width:600px;max-width:100%;background:#FFFFFF;border:1px solid #E3E9F4;border-radius:20px;box-shadow:0 26px 64px -32px rgba(15,30,70,.34),0 2px 4px rgba(15,30,70,.05);overflow:hidden} .hdr{display:flex;align-items:center;gap:11px;padding:16px 20px;border-bottom:1px solid #EEF1F7;background:#F5F8FE} .hdr .ic{display:inline-flex;width:30px;height:30px;align-items:center;justify-content:center;border:2px solid #155DFC;border-radius:50%;color:#155DFC;font-size:15px;font-weight:800;flex-shrink:0} .hdr .t{min-width:0} .hdr .t .n{font-weight:800;font-size:14.5px;color:#0F172A;letter-spacing:-.2px} .hdr .t .s{font-size:12px;color:#64748B} .hdr .pill{margin-left:auto;display:inline-flex;align-items:center;gap:6px;font-size:11.5px;font-weight:700;color:#155DFC;background:#E6EEFF;padding:5px 10px;border-radius:20px;white-space:nowrap} .hdr .pill .d{width:6px;height:6px;border-radius:50%;background:#155DFC;animation:pulseDot 2s infinite} .tabs{display:flex;gap:0;padding:12px 16px 0;background:#F5F8FE} .tab{flex:1;display:flex;flex-direction:column;align-items:center;gap:5px;cursor:pointer;font-family:inherit;font-size:11.5px;font-weight:700;color:#64748B;background:transparent;border:none;border-bottom:2px solid transparent;padding:8px 4px 10px} .tab.on{font-weight:800;color:#155DFC;border-bottom-color:#155DFC} .tab .num{display:inline-flex;width:20px;height:20px;align-items:center;justify-content:center;border-radius:50%;font-size:11px;font-weight:800;color:#64748B;background:#EEF1F7} .tab.on .num{color:#fff;background:#155DFC} .divider{height:1px;background:#EEF1F7} .body{padding:20px} .eyebrow{font-size:10.5px;font-weight:800;letter-spacing:1.2px;text-transform:uppercase;color:#155DFC;margin-bottom:5px} .eyebrow.g{color:#15884F} .headline{font-family:'Inter',Georgia,serif;font-size:18.5px;font-weight:600;line-height:1.25;color:#0F172A;margin-bottom:12px} .lede{font-size:12.5px;line-height:1.55;color:#475569;margin:0 0 14px} .rows{display:flex;flex-direction:column;gap:8px;margin-bottom:14px} .row{display:flex;align-items:center;gap:11px;background:#F5F8FE;border:1px solid #EEF1F7;border-radius:11px;padding:11px 13px} .row .emoji{font-size:17px} .row .rt{flex:1;min-width:0} .row .rt .n{font-size:13px;font-weight:700;color:#0F172A} .row .rt .s{font-size:12px;color:#64748B} .pillsm{font-size:11px;font-weight:800;color:#155DFC;background:#E6EEFF;padding:4px 9px;border-radius:6px;flex-shrink:0} .pillsm.gray{font-weight:700;color:#64748B;background:#EEF1F7} /* chain / timeline */ .chainwrap{position:relative;padding-left:26px} .chainline{position:absolute;left:8px;top:8px;bottom:8px;width:2px;background:#E3E9F4} .cnode{position:relative;margin-bottom:14px} .cdot{position:absolute;left:-26px;top:4px;width:16px;height:16px;border-radius:50%;background:#fff;border:3px solid #94A3B8} .clabel{font-size:9.5px;font-weight:800;letter-spacing:.5px;text-transform:uppercase;color:#64748B;margin-bottom:4px} .cbox{background:#F5F8FE;border:1px solid #EEF1F7;border-radius:11px;padding:10px 12px;font-size:12px;line-height:1.5;color:#334155} .cbox.graded{display:flex;align-items:flex-start;gap:9px;border-left-width:3px;border-left-style:solid} .cnote{margin:0 0 14px 4px;font-size:11px;line-height:1.5} .cnote .plus{display:inline-flex;width:16px;height:16px;align-items:center;justify-content:center;background:#155DFC;color:#fff;border-radius:4px;font-size:10px;font-weight:800;flex-shrink:0;margin-right:6px} .csummary{display:flex;align-items:center;gap:12px;background:#0A1B3D;border-radius:13px;padding:11px 15px;margin-top:4px} .csummary .t{flex:1;font-size:11.5px;color:#CBD8F5} .csummary .t b{color:#F1F5FF} /* matrix */ .mxwrap{border:1px solid #EEF1F7;border-radius:12px;overflow:hidden;margin-bottom:10px} .mxhead{display:grid;grid-template-columns:118px 1fr 1fr;background:#F5F8FE;border-bottom:1px solid #EEF1F7} .mxhead div{font-size:9px;font-weight:800;letter-spacing:.5px;text-transform:uppercase;color:#64748B;padding:7px 9px} .mxrow{display:grid;grid-template-columns:118px 1fr 1fr;border-bottom:1px solid #EEF1F7} .mxrow:last-child{border-bottom:none} .mxrow>div{padding:9px;font-size:11.5px;line-height:1.4;color:#334155;display:flex;align-items:center;flex-wrap:wrap;gap:5px} .mxrow .mxlabel{font-weight:700;color:#0F172A;background:#FAFBFE} .mxchip{font-size:9px;font-weight:800;letter-spacing:.3px;text-transform:uppercase;padding:3px 6px;border-radius:5px;white-space:nowrap} .promptwrap{padding:0 20px 18px} .promptlbl{font-size:10px;font-weight:800;letter-spacing:1px;text-transform:uppercase;color:#94A3B8;margin-bottom:7px} .prompt{display:flex;align-items:flex-start;gap:9px;font-family:ui-monospace,'SF Mono',Menlo,monospace;font-size:11px;line-height:1.5;color:#1E293B;background:#F5F8FE;border:1px solid #EEF1F7;border-left:3px solid #155DFC;border-radius:9px;padding:11px 12px} .prompt .car{color:#155DFC;font-weight:800;flex-shrink:0} .flagrow{display:flex;align-items:flex-start;gap:10px;background:#F5F8FE;border:1px solid #EEF1F7;border-radius:11px;padding:10px 12px;border-left-width:3px;border-left-style:solid;margin-bottom:8px} .flagtag{font-size:9.5px;font-weight:800;letter-spacing:.4px;text-transform:uppercase;padding:4px 8px;border-radius:5px;white-space:nowrap;flex-shrink:0} .flagtxt{font-size:12px;line-height:1.45;color:#334155} .lp{display:flex;align-items:center;gap:12px;background:#0A1B3D;border-radius:13px;padding:14px 16px;margin-top:6px} .lp .ic{display:inline-flex;width:34px;height:34px;align-items:center;justify-content:center;background:#14295C;border:1px solid #26407A;border-radius:9px;color:#7FA8FF;font-size:16px;flex-shrink:0} .lp .rt{flex:1;min-width:0} .lp .rt .n{font-size:13px;font-weight:800;color:#F1F5FF} .lp .rt .s{font-size:11.5px;color:#9DB2D9} .lp .open{font-size:11px;font-weight:800;color:#0A1B3D;background:#7FA8FF;padding:6px 11px;border-radius:7px;white-space:nowrap} </style> </head> <body> <div class="card"> <div class="hdr"> <span class="ic">&#128203;</span> <div class="t"> <div class="n">Case-management program (synthetic) · Structure Case Note</div> <div class="s">DEMO-08 · Sopact Sense · Case Management</div> </div> <span class="pill"><span class="d"></span>Structuring</span> </div> <div class="tabs" id="tabs"></div> <div class="divider"></div> <div class="body" id="body"></div> <div class="promptwrap"> <div class="promptlbl">The prompt behind this step</div> <div class="prompt"><span class="car">&gt;</span><span id="promptTxt"></span></div> </div> </div> <script> (function(){ var TABS = ["Setup", "Structure", "Fix", "Report"]; var CONTENT = [{"eyebrow": "Step 1 · Load the Decision Brief", "headline": "Case notes are unstructured gold — if you extract them carefully.", "lede": "The Assistant works only from the Caseload dataset (DEMO-08), a synthetic case-management program.", "prompt": "You are the Sopact Sense Assistant working over the [DEMO] dataset (clean data + persistent contact IDs). Load my Decision Brief (decision, audience, outcomes, indicators, evidence standard) first, then wait for my task."}, {"eyebrow": "Step 2 · Extract structured fields from the case note", "headline": "Describe only what's written.", "lede": "Needs, progress, services, risks, and next steps are extracted and graded <strong style=\"color:#15884F\">green</strong>, <strong style=\"color:#C9820A\">amber</strong>, <strong style=\"color:#C63E2B\">red</strong> on whether they're explicit or missing.", "prompt": "From the case note for [CLIENT_ID], extract needs, progress, services, risks, next steps, stage-of-change (if derivable). Describe only what's written. Grade green/amber/red."}, {"eyebrow": "Step 3 · One-fix follow-up", "headline": "Capture next steps.", "lede": "Next steps sit at AMBER — no plan recorded. The Assistant names the single data point that moves it to green.", "prompt": "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."}, {"eyebrow": "Step 4 · Missing & incomplete report", "headline": "Is the case note usable as data? Mostly — stage inferred, next steps missing.", "lede": "Every amber or red item becomes one named ask — the report leads with the decision it informs.", "prompt": "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."}]; var BODIES = { setup: "<div class=\"row\"><span class=\"emoji\">&#128221;</span><div class=\"rt\"><div class=\"n\">Decision Brief</div><div class=\"s\">Decision, audience, outcomes, indicators, evidence standard loaded first</div></div><span class=\"pillsm\">Pasted</span></div><div class=\"row\"><span class=\"emoji\">&#128203;</span><div class=\"rt\"><div class=\"n\">Five elements</div><div class=\"s\">Dataset · the fields · only what&#x27;s written · mark missing · grade G/A/R</div></div><span class=\"pillsm gray\">Structure</span></div><div class=\"row\"><span class=\"emoji\">&#128274;</span><div class=\"rt\"><div class=\"n\">No hallucination rule</div><div class=\"s\">Describe only what&#x27;s written; infer = flagged</div></div><span class=\"pillsm gray\">Required</span></div>", build: "<div class=\"mxwrap\"><div class=\"mxhead\"><div></div><div>Extracted</div><div>Note</div></div><div class=\"mxrow\"><div class=\"mxlabel\">Presenting needs</div><div>housing, income<span class=\"mxchip\" style=\"background:#15884F;color:#fff\">GREEN</span></div><div>explicit</div></div><div class=\"mxrow\"><div class=\"mxlabel\">Services</div><div>case management</div><div>explicit</div></div><div class=\"mxrow\"><div class=\"mxlabel\">Stage of change</div><div>contemplation</div><div>inferred<span class=\"mxchip\" style=\"background:#C9820A;color:#fff\">AMBER</span></div></div><div class=\"mxrow\"><div class=\"mxlabel\">Next steps</div><div>not stated<span class=\"mxchip\" style=\"background:#C63E2B;color:#fff\">RED</span></div><div></div></div></div>", fix: "<div class=\"mxwrap\"><div class=\"mxhead\"><div></div><div>Extracted</div><div>Note</div></div><div class=\"mxrow\"><div class=\"mxlabel\">Presenting needs</div><div>housing, income<span class=\"mxchip\" style=\"background:#15884F;color:#fff\">GREEN</span></div><div>explicit</div></div><div class=\"mxrow\"><div class=\"mxlabel\">Services</div><div>case management</div><div>explicit</div></div><div class=\"mxrow\"><div class=\"mxlabel\">Stage of change</div><div>contemplation</div><div>inferred<span class=\"mxchip\" style=\"background:#C9820A;color:#fff\">AMBER</span></div></div><div class=\"mxrow\"><div class=\"mxlabel\">Next steps</div><div>Require a next-step field at note close<span class=\"mxchip\" style=\"background:#15884F;color:#fff\">GREEN</span></div><div></div></div></div><div class=\"cnote\" style=\"color:#15884F;margin-left:0\"><span class=\"plus\">+</span><strong>Fix applied:</strong> Require a next-step field at note close — moves next steps from AMBER (no plan recorded) to GREEN (captured).</div>", reportFlags: "<div class=\"flagrow\" style=\"border-left-color:#C9820A\"><span class=\"flagtag\" style=\"color:#C9820A;background:#FBF0DA\">AMBER</span><div class=\"flagtxt\">Stage of change: inferred, not stated. Fix: add a stage dropdown to the note.</div></div><div class=\"flagrow\" style=\"border-left-color:#C63E2B\"><span class=\"flagtag\" style=\"color:#C63E2B;background:#FCE9E7\">RED</span><div class=\"flagtxt\">Next steps: not recorded. Fix: require a next-step field.</div></div>", lpTitle: "Missing & incomplete report · shareable link", lpSub: "Decision-first · Sopact branding · Case-management program (synthetic) · Structure Case Note", summaryBefore: "Grade: 1 green &middot; 1 amber &middot; 1 red. next steps not recorded", summaryAfter: "Grade: 2 green &middot; 1 amber &middot; 0 red. next-step field required, captured" }; var phase = 1; var tabsEl = document.getElementById('tabs'); var bodyEl = document.getElementById('body'); var promptEl = document.getElementById('promptTxt'); function renderTabs(){ tabsEl.innerHTML = ''; TABS.forEach(function(label, i){ var b = document.createElement('button'); b.type = 'button'; b.className = 'tab' + (i === phase ? ' on' : ''); b.innerHTML = '<span class="num">' + (i+1) + '</span><span>' + label + '</span>'; b.addEventListener('click', function(){ phase = i; render(); }); tabsEl.appendChild(b); }); } function renderBody(){ var c = CONTENT[phase]; var eyebrowClass = (phase===2) ? 'eyebrow g' : 'eyebrow'; var out = '<div class="' + eyebrowClass + '">' + c.eyebrow + '</div><div class="headline">' + c.headline + '</div><p class="lede">' + c.lede + '</p>'; if (phase === 0) { out += '<div class="rows">' + BODIES.setup + '</div>'; } else if (phase === 1) { out += BODIES.build; out += '<div class="csummary"><div class="t"><b>' + BODIES.summaryBefore + '</b></div></div>'; } else if (phase === 2) { out += BODIES.fix; out += '<div class="csummary"><div class="t"><b>' + BODIES.summaryAfter + '</b></div></div>'; } else { out += BODIES.reportFlags; out += '<div class="lp"><span class="ic">&#128196;</span><div class="rt"><div class="n">' + BODIES.lpTitle + '</div><div class="s">' + BODIES.lpSub + '</div></div><span class="open">Open &rarr;</span></div>'; } bodyEl.innerHTML = out; promptEl.textContent = c.prompt; } function render(){ renderTabs(); renderBody(); } render(); })(); </script> </body> </html>
Sopact Academy · Case intelligence
Module 2 of 5 · Case intelligence

Read notes and check-ins without losing the person’s meaning

Make notes useful by preserving who said or observed what, when it happened and what remains uncertain. Review evidence against the question at hand rather than turning every phrase into a score. AI can organize material and suggest themes, but staff must verify interpretations and own consequential decisions.

Read the lesson ↓

Separate what happened from what you infer

P-24 reports that transport is difficult. Two session records show absences. It is reasonable to ask whether transport affected attendance; the available evidence does not establish that it caused both absences. A useful record keeps that distinction visible.

Separate participant statements, practitioner observations, interpretations and agreed actions. These can appear together, but a later reader should be able to tell them apart. Do not convert a tentative observation into a diagnosis or a settled explanation.

A structured note can still preserve narrative. Use a small set of headings for source, event, context, agreed action and next check, then retain the original wording where appropriate. A rigid template that removes the person’s meaning defeats the purpose.

Give analysis a clear question and evidence boundary

Ask the assistant or reviewer a bounded question: what evidence in the authorized record may help prepare the next conversation with P-24? Require source references and unresolved questions. A general request to “score this person” lacks both purpose and a defensible interpretation rule.

If your program uses an observation rubric, write the anchors and intended use first. Test it on varied examples. Do not treat sentiment in a short comment as a validated measure of wellbeing, safety or willingness to participate.

Distinguish concerns explicitly raised by a participant or practitioner from AI-suggested themes. A concern that requires a response needs an appropriate human process. The course does not assume a real-time alerting or risk-prediction engine is available.

Review and correct with the source attached

Keep the original note next to any summary or coded interpretation. If the reviewer corrects the interpretation, record the reason and review status. Where an audio transcript or document extract is used, preserve the relationship to the source and identify uncertainty in extraction.

The advantage over repeatedly opening stored attachments is that the team has already defined how new evidence will be read and checked. The result remains connected to the person’s history and can be revisited when new information arrives.

For the pilot, ask a colleague to challenge one proposed finding. Can they find the passage, distinguish interpretation from fact and record a correction? That is a more useful test than whether a generated summary sounds confident.

Work through the Pathways example

This is a fictional teaching example. Adapt the fields and rules to the question your own program needs to answer.

EvidenceInterpretation to avoidUseful next step
Transport difficulty reported at intakeAll later absences were caused by transportAsk whether the difficulty affected the missed sessions
Two absences recordedThe participant is disengagedCheck context and speak with the participant
Practitioner suggests supportThe intervention will solve the problemRecord the agreed action and plan a follow-up

A note template staff can actually use

Keep a small shared core: participant and episode, observation date, contributor and role, the person’s account, observed facts, interpretation, agreed action, owner and review date. Add program-specific fields only when they serve a decision. An identical template alone does not guarantee comparable judgment.

FieldFictional P-24 exampleReview
Source statement“The evening bus no longer stops near my work.”Retain the quote and date.
Observed factTwo missed sessions in the attendance record.Confirm the dates; do not infer the reason.
InterpretationTransport may affect attendance.Unconfirmed until discussed with P-24.
Agreed actionPractitioner will discuss available session times at the next check-in.Record owner, date and whether P-24 agreed.

Missing information should remain visibly missing. Do not invent an action to satisfy a required field. Allow “not yet agreed” with an owner to resolve it. A revised template improves future collection; it cannot repair earlier notes without new evidence.

Calibrate interpretation before scaling

Ask two colleagues to review the same fictional notes independently. Compare which passages support each category and where they disagree. Refine definitions, retain uncertainty and record the review rule. If AI prepares fields, inspect its output against the original and test ambiguous, incomplete and contradictory notes. A traffic-light label describes your review status only when its meaning is defined; it does not validate a clinical or behavioral assessment.

Build this part of your plan in more detail

These lessons address the next practical questions in this module. Choose the detail your workflow needs, then return to complete the exercise.

More practical questions in this module (6)
Detailed practical guide: How Do You Standardize Case Management Notes?

The full guide is preserved here. Use this detail when it helps your plan, then continue with the exercise.

Standardize case management notes by agreeing the small set of information needed for continuity, review and appropriate reporting. Keep the original account, identify whose perspective it represents, record dates and distinguish a proposed action from one that happened. Different teams can retain useful local note formats while sharing definitions for the fields they genuinely need to compare.

This lesson produces a note structure and a review example. It applies to participant services, coaching and program support. Adapt it to your professional responsibilities and the purpose of the record; there is no single field list that suits every setting.

Start with the next person who needs the record

Ask what an authorized colleague needs to understand at the next contact. Usually that includes who the note concerns, when the interaction occurred, what was reported or observed and whether an action needs follow-up. It does not require documenting every personal detail discussed.

Use a continuing person identifier and, where relevant, the program or service episode. Keep the author and the date of entry distinct from the date of the interaction. A late-entered note should not appear to describe a new event.

Separate the account from the interpretation

Consider this fictional note: “P-24 said the afternoon session conflicts with a work shift. We discussed asking about a morning place. I will contact the coordinator on Friday.”

FieldSupported entryDo not add
Participant accountP-24 reported a work-shift conflict with the afternoon sessionAn assumption about motivation or willingness
Option discussedAsk about a morning placeA statement that a morning place was offered or accepted
Next actionThe note author will contact the coordinator on FridayA statement that the contact already happened
OutcomeNot yet established in this noteA claim that the attendance difficulty is resolved

A staff interpretation can be useful when its basis is clear and the role permits it. Label it as an interpretation, retain the supporting evidence and distinguish it from what the participant said. Do not make a guessed stage, emotion or risk state look like a recorded fact.

Agree a small shared core across locations

A mentoring team and a coaching team may need different local questions. They can still agree common fields such as person or episode identifier, interaction date, source role, action status and review owner. Additional fields should serve the local work rather than satisfy a universal template.

For shared reporting, define the field before combining it. One team’s “completed” may mean that a session occurred; another’s may mean that a support action was finished. Those values should not be added together as the same outcome.

Use the data dictionary to record the common meaning, allowed states and valid mappings. Keep source wording and local detail. If a note does not establish whether an action happened, a mapping cannot supply that missing observation.

Use missing states that reflect the work

A next step is useful when follow-up is needed, but not every contact requires a new task. Allow an appropriate state such as “no further action agreed,” “awaiting a decision” or “not recorded.” These states have different meanings.

Do not make staff invent an owner or a date merely to save a note. Decide which information must be present immediately and what can enter a review queue, based on the workflow. Repeated gaps may indicate an unclear template, training need, workload problem or a field that is not appropriate; investigate before choosing the fix.

Extract structured fields without rewriting the source

In an AI-assisted workflow, the original note remains the evidence. Suggested structured fields should point to the relevant text, use the agreed definitions and show uncertainty. Review ambiguous or consequential interpretations rather than approving a value because it received a color or score.

In Sopact, this can connect qualitative notes to a continuing record and shared analysis context. The operational test is whether a reviewer can inspect the source, correct the proposed field and retain the reason. A prompt alone does not guarantee that every interpretation is accurate or that an access rule is enforced.

Check a varied sample, including ordinary notes without flags, short notes, corrected entries and notes from different locations. If the same phrase is interpreted differently, refine the definition or leave the uncertainty visible.

Keep actions and corrections in the history

When the coordinator is contacted, record that as a later event with its own date and result. If a morning place is offered, record the offer. If the participant accepts, record that separately. This makes the path from discussion to action understandable without claiming an outcome too early.

For a correction, retain the original or an appropriate audit history, the corrected information, the responsible person and the reason. Do not silently alter an earlier note so it appears the team knew something at the time that it learned later.

Reusing a template is different from copying an old observation into a new note. A repeated concern may still be active, but its current status needs evidence. Keep historical statements from becoming fresh observations by accident.

Test both individual review and aggregate reporting

First, ask a colleague to reconstruct the fictional interaction and its next step. Can they tell what was said, what was discussed and what remains to be done?

Then test a group question, such as how many current actions are awaiting review. Define the population, period and status rule. Count the relevant actions or people explicitly; several notes about one action should not automatically become several unresolved actions.

Check the audience for the result. An aggregate report may not need personal quotations. A source note can contain information that should not appear in a shared dashboard or export. Test the relevant access and sharing arrangements using fictional records before relying on them.

Your exercise

  1. Choose a note purpose and the authorized reader.
  2. Define the small shared core and any local fields.
  3. Use the fictional note to separate account, interpretation, option and action.
  4. Mark what is missing without filling it in.
  5. Record a later event without overwriting the earlier note.
  6. Test one aggregate question using the field definitions.
  7. Review what each role can see in the note, analysis and export.

Frequently asked questions

Must every caseworker use the same note template?

Not necessarily. Agree the fields and definitions needed for shared continuity or reporting, while allowing local detail suited to the work. Compare only fields with compatible meanings.

Should AI infer a field that the note does not state?

Do not present an inferred value as a recorded fact. Where interpretation is appropriate, label it, keep its evidence and require suitable review. Leave unsupported fields unresolved.

What is the best fix for incomplete notes?

Investigate the gap and its cause. A clearer field, legitimate missing state, training or workflow change may help. Making every field mandatory can produce invented or misleading entries if the information is not available.

Add this part to your plan

On workbook page 3, write one source statement, one observation, one provisional interpretation and one question for P-24. Add the responsible reviewer.

Download workbook (fillable PDF)
Compare with a suggested answer

The transport statement is participant-reported; the absences are recorded events. Their relationship remains a question. The reviewer checks dates and prepares a conversation rather than asserting a cause. Any agreed support and later response become new evidence.

Self-check: Can a colleague challenge the interpretation without having to reconstruct its source?

Questions you may have

Can AI decide which person is at risk?

This course does not establish a validated risk model. Use source-backed review to support the appropriate human process; do not treat generated labels as validated predictions.

Should standardization remove free text?

No. Standardize useful context and review fields while preserving relevant narrative and who supplied it.

Apply the method to your work

Use the five-part plan to assess the sources, analysis and permissions your team needs. The solution page shows where a connected platform can support that workflow.

Explore the case intelligence solution →

Put this guide into practice.

Start with a decision your team needs to make. Decide what information belongs together, how you will check it, and who is responsible for acting.

See context in action →
Prepare data governance before using AI
Prepare your data governance
nonprofit-data-governance-before-ai
Feedback
Foundation
Prepare an evidence register and a tested governance baseline before applying AI to program data.
Make learning part of the work you already do
The Loop — the method in one read
the-loop
Loop
The method
0
One continuous method for reliable, traceable AI reporting across case, application, grant, and program workflows: collect clean, analyze on arrival, and improve in time.
Teams running cases, applications, grants, or programs that need AI-assisted reporting they can reproduce, trace to source evidence, and act on before the cycle ends.
What Is Case Intelligence?
What Is Case Intelligence?
what-is-case-intelligence
Case
Foundation
1
One current, traceable record for each person—connecting intake, services, notes, surveys, documents, outcomes, decisions, and follow-up.
Workforce and training · Youth and mentoring · Case management · Scholarships · Accelerators · Education · Nonprofit programs
What Is Grant Intelligence?
What Is Grant Intelligence?
what-is-grant-intelligence
Grant
Foundation
1
One connected evidence record from application and committee review through the awarded grant, grantee reporting, renewal, and board accountability.
Foundations and grantmakers · Public grant programs · Scholarships and fellowships · Accelerators
What Is Portfolio Intelligence?
What Is Portfolio Intelligence?
what-is-portfolio-intelligence
Portfolio
Strategy
1
A source-linked portfolio view that connects each investee or grantee's agreed plan, reporting cadence, evidence, risks, and results.
Impact funds and investors · Foundations with grant portfolios · Family offices · Blended-finance vehicles
What Is Connected Data Intelligence?
What Is Connected Data Intelligence?
connected-data-intelligence
Feedback
Foundation
1
Keep evidence from surveys, files, notes, documents, systems, sites, and reporting periods connected to one continuing record.
Multi-program nonprofits · Member and chapter networks · Research associations · Training and coaching teams · Organizations with scattered evidence
Build an impact report from evidence you can explain
What Is Impact Measurement and Reporting?
embedded-impact-measurement
Reporting
Align
1
Define intended change, align organization and funder context, govern measures, interpret evidence, and produce traceable reports for decisions.
Program and impact teams · MEL leads · Funders and donors · Foundations · Impact funds · Organizations building rigorous impact reports
Build a repeatable collect, review and improve cycle
Methodology — continuous, not annual
loop-methodology
Loop
The method
1
The continuous collect–analyze–improve cycle, adopted as an experiment: start with the step that already pays and add one data-collection step at a time.
Teams tired of rebuilding spreadsheets and forms who want a measurement system that compounds instead of resetting.
How to Build a Theory of Change with AI: Prompts and Examples
Build a Theory of Change You Can Test
how-to-build-a-theory-of-change
Reporting
Align
2
How Do You Onboard a Portfolio and Track Results?
Agree the Portfolio Reporting Plan
onboard-portfolio-lock-impact-agreement-track-results
Portfolio
Data Dictionary
2
Test whether an AI-assisted result is repeatable and correct
Reliability — the same answer twice
loop-reliability
Loop
The method
2
Determinism as a feature: the same question over the same data returns the same answer every run — the opposite of a generic AI chat that drifts.
Anyone who has watched a general AI tool give two different numbers for the same question and needs results they can stand behind.
Design application intake around a defensible decision
Design an Application Process
how-to-design-an-application-process
Grant
Foundation
2
How to Structure Stakeholder Data: Four Common Patterns
Choose your record structure
which-shape-is-your-data
Feedback
Foundation
2
Map the people, observations and relationships your workflow needs before collecting data.
How to Build a Logic Model: Steps, Example and AI Prompt
Build a Logic Model You Can Use
how-to-build-a-logic-model
Reporting
Align
3
How Do You Onboard a Grant or RFP Program?
Onboard a Grant or RFP Program
how-to-onboard-a-grant-rfp-program
Grant
Foundation
3
Turn your theory of change into a data-collection plan
Turn a Theory of Change into a Data-Collection Workflow
theory-of-change-to-data-collection-workflow
Case
Foundation
3
Build a portfolio data dictionary without forcing false comparisons
Map Portfolio Data to Reporting Standards
portfolio-data-dictionary-standards-mapping
Portfolio
Chapters
3
Keep a clear trail from a finding to its evidence
Traceability & Transparency
loop-traceability
Loop
The method
3
Every figure links back to the exact response, note, or document it came from — a full audit trail from headline result to raw evidence.
Teams whose numbers get scrutinized — by funders, boards, auditors, or standards — and who need to answer where did this come from on the spot.
How to Change Survey Questions Without Losing Comparability
Change questions with a clear history
change-questions-without-breaking-the-record
Feedback
Control
3
Create a question-change log and decide how old and new versions should appear in reports.
Programme & MEL leads · Teams whose questionnaire has ossified · Anyone evaluating a platform where configuration is a purchased service
Five Dimensions of Impact: How to Review Your Evidence
Use the Five Dimensions to Test the Evidence
five-dimensions-of-impact
Reporting
Align
4
How Do You Design a Grant Rubric and Eligibility Rules?
Design Your Rubric & Eligibility Rules
grant-rubric-eligibility-rules
Grant
Foundation
4
How to Measure Outcomes Across an Investment or Grant Portfolio
Frame Outcomes Over Outputs at Portfolio Level
frame-outcomes-portfolio-level
Portfolio
Chapters
4
Adapt the learning cycle to your workflow
Flexibility — one method, four workflows
loop-flexibility
Loop
The method
4
The same collect–analyze–improve cycle, shaped to four kinds of impact work — case, grant, portfolio, and feedback — each shown end to end.
Anyone deciding where the Loop fits their work, who wants to see the full path from messy input to a report they can defend.
How to Collect Feedback Offline and Keep Records Connected
Collect offline and reconcile the batch
collect-feedback-offline
Feedback
Connect
4
Build a field protocol and reconcile a test batch across devices, visits and delayed uploads.
Field & multi-site programs · Low-connectivity contexts · Nonprofits collecting in person
How Do You Design an Intake Form for a Baseline?
Design an Intake Form That Captures a Usable Baseline
intake-form-usable-baseline
Case
Nonprofit Track
5
Turn a proposed outcome into a reporting definition
Outcomes vs Outputs
frame-outcomes-over-outputs
Grant
Foundation
5
Impact Due Diligence: Review Evidence Before Investment
Pre-Investment Due Diligence & Screening
pre-investment-due-diligence-screening
Portfolio
Chapters
5
Plan and review your first workflow pilot
The Guarantee — first workflow in 2 months
loop-guarantee
Loop
The method
5
How to Build an Organization Evidence Model
Build the Organization Evidence Model
build-organization-evidence-model
Reporting
Align
5
How to Analyze Documents as Evidence: Sources, Context and Review
Read documents as traceable evidence
read-documents-as-evidence
Feedback
Connect
5
Create a document register and reviewed findings with source locations, context and explicit exceptions.
How to Clean Open-Ended Survey Responses Without Losing Meaning
Clean responses and define the denominator
clean-open-ended-survey-responses
Feedback
Clean
6
Create a cleaning log, response-status table and reproducible report statement.
How to Review Participant Support Needs Mid-Program
Spot At-Risk Participants Mid-Program
spot-at-risk-participants-mid-program
Case
Nonprofit Track
6
How to Write a Nonprofit Grant Application: Template and Example
Grant Application for Nonprofits
grant-application-for-nonprofit-organizations
Grant
Foundation
6
Design quarterly portfolio reporting that investees can complete
Collect Investee Reporting Without Repeated Rework
collect-investee-reporting-without-burden
Portfolio
Chapters
6
How to Build a Funder Context Profile: Research to Reporting
Build a Sourced Funder Context Profile
build-funder-context-profile
Reporting
Align
6
How Do You Measure Change at Exit?
Measure Change at Exit (Not Just Completion)
measure-change-at-exit
Case
Nonprofit Track
7
How Do You Collect Applications Clean at the Source?
Collect Applications Clean at the Source
collect-applications-clean-at-source
Grant
Collect
7
How to Follow Up on Missing Investee Data
How to Follow Up on Missing Investee Data
chase-missing-investee-data
Portfolio
Chapters
7
How to Analyze Multilingual Feedback and Evaluate Software
Analyze and review multilingual feedback
analyze-multilingual-feedback
Feedback
Clean
7
Build a language review sheet and test software on original responses, translations, codes and reporting bases.
Multi-country programs · Multilingual survey data · Global networks & chapters
How to Define Impact Metrics Your Team and Funder Can Use
Define Measures the Organization and Funder Can Both Use
define-impact-metrics-funders-want
Reporting
Align
7
Survey Attrition in Longitudinal Studies: Track Missing Waves
Track missing waves and matched outcomes
survey-attrition-longitudinal-studies
Feedback
Read
8
Build a wave-status register, compare response groups and report paired change with coverage and limitations.
How to Use Mentor Notes to Review Participant Support
Use mentor notes for support review
mentor-notes-early-warning
Case
Nonprofit Track
8
How Do You Reduce Applicant Burden?
Reduce Applicant Burden
reduce-applicant-burden-auto-clarification
Grant
Collect
8
Find inconsistencies in portfolio returns before reporting
Analyze Investee Reports Across Sources
read-investee-reports-multi-signal
Portfolio
Chapters
8
Impact Metric Definitions: A Practical Worksheet and Example
Give Every Number One Definition
one-definition-for-every-number
Reporting
Define
8
How to Connect Quantitative and Qualitative Survey Data
Connect scores and comments
connect-quantitative-qualitative-survey-data
Feedback
Read
9
Build a linked analysis view and joint display, with clear groups, reporting bases and evidence limits.
How to Calculate SROI as New Evidence Arrives
Calculate SROI — Live, Sourced, and Honest
calculate-sroi-live
Case
Nonprofit Track
9
How to Collect Grantee Reports with Less Burden
Collect Grantee Reports Without Burden
collect-grantee-reporting-without-burden
Grant
Collect
9
Keep company, investment and reporting history connected
Track Results Against the Impact Agreement
track-investees-impact-agreement-variance
Portfolio
Chapters
9
Turn Reporting Requirements into Evidence: A Practical Mapping Guide
Turn Requirements Into Collectable Evidence
turn-reporting-requirements-into-evidence
Reporting
Define
9
Connect baseline, follow-up and different rater perspectives
Analyze pre, mid and post surveys
analyze-pre-mid-post-survey-data
Feedback
Read
10
Build a matched pre/mid/post analysis, interpret score movement and retain clear rules for missing waves.
How to Report a Job-Training Program to Grant Funders
Turn a Cohort into a Funder Impact Report
job-training-grant-impact-report
Case
Nonprofit Track
10
How to Follow Up on Missing Grantee Data
Chase Missing Grantee Data
chase-missing-grantee-data
Grant
Collect
10
How to Build Useful Portfolio Impact Monitoring Alerts
Build Useful Portfolio Impact Alerts
portfolio-risk-monitoring-alerts
Portfolio
Chapters
10
How to Collect Clean Data Inside Your Workflow
Collect Clean Evidence Inside the Workflow
collect-clean-data-at-the-source
Reporting
Embed
10
How to Analyze Longitudinal Survey Data
Analyze longitudinal survey data
analyze-longitudinal-survey-data
Feedback
Read
11
Build a continuing analysis record with clear time scales, observed trajectories and limits.
How to Turn a Job Description into a Requirements Checklist
Clarify employer requirements
job-description-requirements-checklist
Case
Social Enterprise Track
11
Review applications with evidence, clear criteria and human judgment
Review Without Reviewer Bias
review-applications-without-reviewer-bias
Grant
Analyze
11
How to Ask AI Questions About Your Portfolio Data
Ask AI Questions About Portfolio Data
ask-your-portfolio-anything
Portfolio
Chapters
11
How to Keep Impact Reporting Numbers Consistent
Get Stable Results From Governed Data
same-numbers-every-time
Reporting
Read
11
How Do You Analyze a Batch of Grant Applications?
Analyze a Whole Round
how-to-analyze-a-batch-of-grant-applications
Grant
Analyze
12
How to Measure Outcome Duration and Drop-Off
Measure outcome duration and drop-off
measure-outcome-duration-drop-off
Feedback
Read
12
Build a dated outcome claim, distinguish missingness from outcome loss and test forecast assumptions.
How to Score Candidate–Role Matches with a Clear Rubric
Review candidate–role evidence
score-candidate-role-matches-without-bias
Case
Social Enterprise Track
12
Evidence Traceability: Link Every Report Claim to Its Source
Trace Every Result Back to Its Evidence
where-every-number-came-from
Reporting
Read
12
How Do You Roll Up a Grant Portfolio?
Aggregate Outcomes Across a Grant Portfolio
how-to-roll-up-a-grant-portfolio
Portfolio
Chapters
13
How to Report Job Placements to Impact Investors
Turn a Cohort into a Social-Enterprise Investor Report
job-placement-investor-impact-report
Case
Social Enterprise Track
13
How Do You Track Reviewer Conflicts of Interest?
Track Reviewer Conflicts of Interest
track-conflicts-of-interest-audit
Grant
Analyze
13
How to Write a Donor Report: Format, Evidence and Example
Design a Report for a Real Funding Decision
donor-report-funders-trust
Reporting
Decide
13
Build a report brief and claim-and-evidence table before drafting. Explain delivery, outcomes, spending, limitations and next actions.
Program managers, grant leads and reporting teams
AI Data Access Controls: What Your Assistant May See
Control what the assistant can access
what-the-assistant-may-see
Feedback
Prove
13
Define task-specific access, test synthetic records and verify report-sharing boundaries.
How to Write an Evidence-Based Impact Narrative
Write a cited impact narrative
impact-narrative-funder-report-cited
Feedback
Prove
14
Build and check a report paragraph using a claim-and-source table, appropriate quotations and clear limitations.
How Do You Read a Grantee Report?
Read a Grantee Report
read-grantee-report-multi-signal
Grant
Analyze
14
How to Use SROI Across a Portfolio Without Double Counting
Use SROI Across a Portfolio
monetize-impact-sroi-across-levels
Portfolio
Chapters
14
How to Write a Funder Report with AI—and Check It
Generate the Audience-Specific Report From Evidence
assistant-writes-the-funder-report
Reporting
Decide
14
Draft from approved sources, check the claims, and save an accountable report version. Bring the brief from the previous lesson.
Program managers, grant leads and reporting teams
How Do You Compute Grantee Variance?
Compute Grantee Variance
how-to-compute-grantee-variance
Grant
Analyze
15
When Is a Monetary Value on Social Impact Credible?
Add a Credible Dollar Value With SROI
credible-dollar-value-on-impact
Reporting
Optional method
15
Prepare a valuation brief. Decide what the evidence supports, what needs more work, and when an outcome account is enough.
Program, evaluation and investment teams considering social-value estimates
Keep a person’s history connected across programs and staff changes
Follow one person over time
one-person-followed-for-years
Feedback
Shapes
15
Build a participant record that preserves episodes, dates, versions and missingness across repeated collection.
How Do You Build an SROI Value Map?
Build an SROI Value Map
how-to-build-an-sroi-value-map
Reporting
Optional method
16
Build a first value map, keep missing evidence visible, and give each unresolved outcome a next action.
Evaluation, program and investment teams preparing an SROI analysis
How Do You Track Budget and Actual Spend?
Track Budget vs Actual Spend
how-to-track-budget-invoices-actual-spend
Grant
Analyze
16
Multi-Rater Feedback: Connect Perspectives and Protect Context
Connect several perspectives on one person
several-people-describing-one-person
Feedback
Shapes
16
Design subject-rater relationships, reporting rules and a tested multi-perspective feedback record.
How Do You Pick a Financial Proxy for SROI?
Pick a Defensible Financial Proxy
how-to-pick-a-financial-proxy-for-sroi
Reporting
Optional method
17
Compare candidate valuation sources and document why one fits your outcome, stakeholder and reporting period.
Evaluation and reporting teams selecting financial proxies
How Do You Analyze Grantee Reporting Longitudinally?
Analyze Grantee Reporting Over Time
analyze-grantee-reporting-longitudinal
Grant
Analyze
17
How Do You Compare Investees When Each One Defines Its Metrics Differently?
Compare & Benchmark Investees
compare-benchmark-investees
Portfolio
Chapters
17
Cross-Program Reporting: Combine Results Without Losing Meaning
Combine evidence across programs
many-programs-one-picture
Feedback
Shapes
17
Build a defensible cross-program result with comparable measures, correct denominators and documented exclusions.
How Do You Calculate the SROI Ratio?
Calculate the SROI Ratio With a Range
how-to-calculate-the-sroi-ratio
Reporting
Optional method
18
Build a reproducible SROI calculation, test its assumptions and explain the result in a reviewed report.
Evaluation and reporting teams reviewing an SROI calculation
How to Read Form 990 for a Grant Review
Read a 990 for Compliance
how-to-read-a-990-for-compliance
Grant
Analyze
18
How Do You Build Dashboards and Compliance Reports?
Build Dashboards and Reviewed Reports
dashboards-sroi-compliance-reports
Portfolio
Chapters
18
Plan evidence collection across your network
Run a member-network survey
member-network-survey
Feedback
Shapes
18
Design and test a member reporting cycle with continuing records, coverage checks and authorized results.
Ask Your Whole Grant Round Anything (Assistant + MCP)
Ask Your Whole Grant Round Anything
ask-your-grant-round-anything
Grant
Analyze
19
Produce portfolio reports that trace back to approved evidence
Produce the LP and Board Impact Report
portfolio-lp-board-impact-report
Portfolio
Chapters
19
How Do You Build a Grant Audit Trail?
Build a Grant Audit Trail
grant-audit-compliance-trail
Grant
Communicate
20
How Do You Connect Your Stack Without Lock-In?
Connect Systems and Test Data Portability
portfolio-connect-your-stack
Portfolio
Chapters
20
How Do You Produce Grant Compliance Reports?
Produce Compliance Reports
grant-compliance-regulatory-reports
Grant
Communicate
21
How Do You Roll Grantees Into a Board Report?
Roll Grantees Into a Board Report
roll-grantees-funder-board-report
Grant
Communicate
22
How Do You Build Grant Dashboards and Maps?
Grant Dashboards & Maps
grant-dashboards-geographic-mapping
Grant
Communicate
23