# The decision brief Write the brief as plain text in the form below, in this order, with the headings spelt as shown. This is the form `site/js/modules/decision-brief.js` draws. Use no bold and no markdown tables. The two-column rows under a reason use a bar, as shown. Use only the headings in the form. Leave out a block when its note says so. Replace each word in capitals or braces with what the note says. ## The form ``` Implementation Readiness Framework for NGOs Decision brief{ for ORGANISATION} {Prepared for: JOB TITLE} {Organisation: TYPE, SIZE, uses USES} {Use case: NAME} Date: D Month YYYY {Reference: IRF-YYYY-MM-DD-NNNN} {only if the person described the problem in a sentence:} {The problem, in your words} {“THEIR SENTENCE”} VERDICT LABEL DECISION SENTENCE {only if more than one problem has a line, or the single line carries a caution:} {Your problems, in order of priority <- this heading only if more than one} {1. LINE} { CAUTION} Why{, for NAME} 1. TAG: REASON {You were asked | You said} {QUESTION | ANSWER} 2. TAG: REASON 3. TAG: REASON {only if the reasons are for yes, yes with conditions or yes once these change, and readiness was scored:} Readiness, 0 to 5 Where this problem needs you to be, against where you are. - AREA: X of 5{, needs Y} Next step{, for NAME} HEADLINE - {[ ] }ITEM {only for yes, yes with conditions and yes once these change:} {What any solution must meet} {- REQUIREMENT} {only if you hold any of the five lines:} {What you told us} {- TYPE | SIZE | Uses USES | COUNTRY, one per line | SECTOR} {only if there are gaps:} {What this brief leaves out} {- GAP} {only if there is a verdict:} {QUESTION LINE} {SUB-LINE} {CONTEXT LINES, one per line} Prepared with IRF v2, in association with the Nordic Blockchain Association D Month YYYY ``` Notes. The reference is `IRF-`, the date as YYYY-MM-DD, and four digits worked out from the lead problem's id (see the header lines below). The organisation appears in the title only if the person gave one. The word for is followed by NAME in the two headings only when more than one problem has a line. With one problem that has a verdict, write the Why and Next step blocks once. With more than one, write the pair of blocks for each of the first three problems that have a verdict, in the same order as the lines, each with its own NAME (the short name of that problem), and give the other problems their line only. Then add as the first line under "What this brief leaves out": `Reasons and next steps are given for the first three problems. The others have a decision line only.` When more than one problem has a block, give the need reason one quote, not two, leave the quote of the parties answer out of the fit reason, and show at most three conditions, then `And N more.`. If there are no context lines, drop that block and its blank line. The date is today's date, written as 16 October 2026. ## The header lines and the verdict label The lines under the title are a short header block, one line each, with no blank line between them. Leave out a line when you hold nothing for it. `Prepared for:` is the job title from `irf_act0.jobTitle`. `Organisation:` is the organisation type, then the size, then `uses USES` from `irf_act0.uses`, joined with commas, with a capital first letter; when there is no type and no size it reads `Uses USES`. `Use case:` is the lead problem named as in the lines below the decision sentence (the short name with a capital first letter, or `The AREA problem` when the short name holds a technology word); with more than one problem that has a line, add ` and N more`, N in words. The lead problem here is the one the script ranks first among those with a verdict. When none has a verdict it is the first that ranks high or medium, and in the case where nothing ranks high or medium it is the highest ranked; when nothing could be ranked, leave out `Use case:` and `Reference:`. `Date:` is the date at the foot of the brief. `Reference:` is `IRF-`, the date as YYYY-MM-DD, a hyphen and four digits from the lead problem's id in the catalogue, for example `IRF-2026-10-18-5845`. Work the digits out as `refCode` does in `site/js/modules/decision-brief.js`: start with h = 7; for each character of the id, set h to (h times 31 plus the character's code) modulo 10007; then take h modulo 10000 and pad it with zeros to four digits. The line above the decision sentence is the verdict label. With one problem that has a verdict it is the verdict as a short label: `Yes`, `Yes, with conditions`, `Yes, once these change` or `No, use something simpler`. With more than one it is `K of N can go ahead`, K the count of yes and yes with conditions, N the count with a verdict. The cases with no verdict at all use `No verdict yet`, `Not enough information` (nothing could be ranked), `Not much to change` or `No viable use case`, in the order of the cases in the section on a brief with no verdict. ## Which problems, and in what order List the use cases that have a verdict, ordered by the rank in the script's output, then by problem magnitude. The lead is the first that has a verdict other than No verdict yet. If none has one, the lead is the first of them. Verdict words. From the script's outcome and rule: PROCEED is Yes. PROCEED_WITH_CONDITIONS is Yes, with conditions. NOT_NOW is Yes, once these change. EXPLORE_ALTERNATIVES is No, use something simpler. NOT_ASSESSED, and the rule no-part-a or no-part-b, are No verdict yet. ## Naming a problem Take the use case's name from the catalogue. Drop a trailing part in brackets: that is the short name. Begin it with a lower-case letter unless it opens with a capitalised word that is not the first word of a sentence (an acronym, or a proper name); in practice, lower-case the first letter when the second letter is also lower-case. Use that as LABEL in the sentences below. If the short name contains a technology word from the list at the end, do not use it in the opening. Use "the AREA problem" instead, with AREA from this table: - A: financial operations and payments - B: identity and credentials - C: data management and privacy - D: supply chain and verification - E: governance and coordination - F: fundraising and donor relations - G: resilience and risk management - H: land, property and natural resources - I: energy and environment - J: workforce and operations Each line under the decision sentence reads: the short name with a capital first letter (or LABEL when it was replaced), a colon, the verdict in lower case, a full stop, and the short reason. For example: "Cross-border beneficiary payments: yes, with conditions. Two conditions to meet first." Short reasons. Yes: Ready to write the tender. Yes with conditions: N condition(s) to meet first (N in words). Yes, once these change: Build AREA first (the readiness area with the biggest gap, named below). No, use something simpler: Try a ALTERNATIVE first (see Next step). No verdict: The questions about your organisation were not answered. (rule no-part-a), or The questions about this problem were not answered. (anything else). Add a caution under a line in two cases. When the fit gate failed and the verdict is yes or yes with conditions: The fit check scored this below the bar, so treat the yes with care. When the use case is not-enough-information but has a verdict: No answer described this problem, so its size could not be checked. Put both, in that order, if both apply. ## The decision sentence Exactly one problem has a verdict: - Yes: `Yes: go ahead with LABEL, starting with a small pilot.` - Yes with conditions: `Yes, with conditions: go ahead with LABEL once N condition(s) is/are met.` N in words (no, one, two ... ten). One condition: "one condition is met". More: "two conditions are met". - Yes, once these change: `Yes, once these change: close the gaps in your readiness for LABEL first.` - No: `No: use something simpler for LABEL.` - No verdict: `No decision yet for LABEL: readiness was not judged.` If other problems have no verdict yet, add: ` N more problem(s) has/have no decision yet.` ("One more problem has no decision yet." or "Two more problems have no decision yet.") More than one problem has a verdict (call that number n; go is the count of yes and yes with conditions; cond is the count of yes with conditions): - All can go (go equals n): `Both problems can go ahead` if n is 2, otherwise `All N problems can go ahead`. Then `, with conditions` if cond equals n, or `, K of them with conditions` if cond is between 1 and n. End with a full stop. - None can go (go is 0): `Neither problem is ready to go ahead.` if n is 2, otherwise `None of the N problems is ready to go ahead.` Then: if all are yes once these change: `Both should wait while your organisation prepares.` or `All N should wait while your organisation prepares.` If all are no: `Both are better served by something simpler.` or `All N are better served by something simpler.` If mixed: `K should wait while your organisation prepares.` and `K is/are better served by something simpler.` (count in words, capitalised). - Some can go: `K of N problems can go ahead` then `, with conditions` if every one that can go has conditions, or `, J with conditions` if some do. Full stop. Then `The other one should wait or use something simpler.` or `The other K should wait or use something simpler.` - If some have no verdict: add ` K more problem(s) has/have no decision yet.` at the end. No problem has a verdict yet, but readiness was attempted for at least one: `No decision yet: readiness was not judged for LABEL.` or `... for K problems.` ## When there is no verdict at all The script gave no readiness result for any use case. Choose the first case that fits. Use the same form as above, but the three blocks as stated here. - Nothing could be ranked (no use case has a rank). Decision sentence: `There is not enough here to give you a decision.` Reason 1 (rule): if no category was chosen, `No problem area was chosen, so no use case could be ranked.` Otherwise `Too few of your answers touched the problems in the areas you chose, so no use case could be ranked.` Next step headline: `Describe the problem you want to solve, in a sentence or two, then run this again.` - Nothing ranks high or medium. Decision sentence: `As you described them, these problems are too small for a new system to pay back.` Reason 1: `The problem that ranked highest for you, LABEL, scored X of 5. That is low.` (say minimal if its tier is minimal). Reason 2: the need reason below. Next step headline: `Keep what you do now.` Item: `Run this again if the problem grows or your situation changes.` - Some rank high or medium, but all fail the fit gate. Decision sentence: `No: none of the uses this tool compares against fits your problem well enough.` Reason 1: the need reason for the highest ranked. Reason 2: `On how well a shared ledger suits them, the best of your problems scored X of 5, below the Y it needed.` (leave out the part after the comma if you have no gate). Next step: `Try a ALTERNATIVE first.` Items: `It WHAT.` and `If it does not solve the problem, describe it in other words and run this again.` - Some rank high or medium and pass the fit gate, but none was checked for readiness. List up to three, each as `Name: no verdict yet. Readiness has not been checked.` Decision sentence: `Not decided yet: LABEL looks worth a closer look, and it has been checked for readiness.` for one, or `Not decided yet: N of your problems look worth a closer look, and none has been checked for readiness.` for several. Reason 1: `Readiness has not been checked, so there is nothing to decide on yet.` Reason 2: the need reason. Next step headline: `Check readiness for LABEL first.` No items. In each of these, the gaps block says how many use cases could not be judged for lack of answers: `N use case(s) could not be judged for lack of answers.` (N in words), taken from the script's notEnoughInformation, ignoring those whose reason says their category was not selected. ## The reasons, by verdict A use case with a verdict gets three reasons, numbered 1 to 3. The kinds depend on the verdict. Readiness is never a reason for No, use something simpler: it has nothing to do with whether a simpler tool does the job. - Yes: the need reason, the fit reason, the strong-areas reason. - Yes, with conditions: the need reason, the fit reason, the gaps reason. - Yes, once these change: the need reason, the fit reason, the first-gap reason. - No, use something simpler: the failed-fit reason, the need reason, the simpler-tool reason. - No verdict yet: the rule line, the need reason, the readiness line (see the last part of this section). Each reason starts with a tag and a colon: `Problem:` for the need reason, `Fit:` for the fit reason and for the failed-fit reason (single-party, low-parties-frequency-impact, evaluation-explore-alternatives), `Readiness:` for the gaps reason, the first-gap reason, the strong-areas reason, the readiness line and the rules about readiness. The simpler-tool reason has no tag. A rule line takes the tag of what it is about. After the tag come the conclusion only, with the quotes taken out of it (see Quoting the person). The quotes follow as a small table under the reason. ### The rule lines Use the line for the rule the script reports for the use case the block is about. DIMNAME is the name of the readiness area with the biggest gap, from the table at the end. SEV is the highest deep-dive rating. The failed-fit reason is the line for single-party or low-parties-frequency-impact. The gaps reason uses the line for large-gap-severity-override, total-gap-severity-override and evaluation-softens-not-now. The other rules are not used as a reason, except for a use case with no verdict. - gaps-cleared: `Taken together, your answers about your organisation come close enough to what this problem calls for.` If the biggest gap is over 0.3, add ` One area falls short. It is named below.` - light-gaps: `Small gaps remain between your answers about your organisation and what this problem calls for. They can be closed while you run a pilot.` - heavier-gaps: `Several gaps remain between your answers about your organisation and what this problem calls for. Close them before or during a pilot.` - modest-gaps-severe-problem: `The gaps in your readiness are modest and you described a severe problem, so the answer is yes.` - large-gap-severity-override: `Your readiness in DIMNAME is well below what this problem calls for, which would normally mean waiting. You described a severe problem (rated SEV of 5), so the answer is yes, with conditions.` - total-gap-severity-override: `Your readiness gaps add up to a lot, which would normally mean waiting. You described a severe problem (rated SEV of 5), so the answer is yes, with conditions.` - large-single-gap: `One readiness gap is large enough that your organisation should close it before anything else. It is named below.` - total-gap-too-large: `Taken together, the gaps in your readiness are too large to start now.` - single-party: `With only your organisation involved, a shared ledger adds little.` - low-parties-frequency-impact: `Few parties are involved, the problem is rare and the effect is small, so a shared ledger adds little.` - no-part-a: `The questions about your organisation were not answered, so readiness could not be judged.` - no-part-b: `The questions about this problem were not answered, so no decision was reached.` If you cannot quote the parties answer in single-party, write `Your answers describe a problem that involves only your organisation, so a shared ledger adds little.` instead. Leave out "(rated SEV of 5)" when you have no rating. ### The failed-fit reason (No, use something simpler) The rule line for the rule that fired. For single-party and low-parties-frequency-impact, put the person's own answers about how many parties and how often in front, in the quoting form below, when you can quote them. For evaluation-explore-alternatives, take the script's sentence, drop its first sentence (`The detailed evaluation found a simpler system would do the job.`) and use the rest. If nothing is left: `The detailed questions you answered describe a problem that a simpler system can solve.` ### The need reason: how big the problem is Begin `You described ADJ.` where ADJ is `a serious problem` if the tier is high, `a real problem` if medium, `a mild problem` if low or minimal. Follow it with up to two quotes (one when more than one problem has a block) from the script's evidence list, need answers first, then scale, then fit boost, highest rating times weight first, leaving out answers such as unknown or not sure, and any answer the script rates below 3 of 5 for severity: a quote must show an answer that pushed the score up, never a low one. If there is no usable quote: `Your answers rank this as TIERWORD (number N among the problems you described).` where TIERWORD is `a high-priority problem`, `a medium-priority problem`, `a low-priority problem` or `a minor problem`. If the use case has no signal at all: `No answer you gave describes how serious this problem is, so its size is not part of this decision.` ### The fit reason (Yes, Yes with conditions, Yes once these change) Say why a shared record has a use, in the first case that fits: - The answer to how many parties are involved is 2 or more: the quote of that answer (leave it out when more than one problem has a block), then `Several parties are involved, so a record they can all check has a use.` for 3 or more, or `More than one party is involved, so a shared record can help.` for 2. - The script's rule is evaluation-softens-not-now, or the detailed evaluation reached an outcome other than explore alternatives: `The detailed questions you answered found that a shared ledger fits this problem.` - The fit gate passed: `A shared ledger fits this kind of problem: it passed the fit check.` - Otherwise: `Your answers describe a problem where a shared record between several parties has a use.` Then add ` On the fit check this scored below the bar, so treat the fit with care.` if the fit gate failed, or ` On the fit check this came close to the bar.` if it was borderline. ### The strong-areas reason (Yes) Take the readiness areas, those furthest above their bar first (bar minus score, smallest first; if no bar, 3 minus score). If none has a score: `Your answers about your organisation meet what this problem calls for.` Otherwise `Your organisation is ready for this.` (`Your organisation is close to ready for this.` when the rule is modest-gaps-severe-problem) then ` Your strongest area is DIMNAME (X of 5).` for one, or ` Your strongest areas are DIMNAME (X of 5) and DIMNAME (Y of 5).` for the first two. For modest-gaps-severe-problem add ` The remaining gaps are modest and the problem is severe, so the answer is yes.` The nearest gap is not a reason in a Yes. It is the first item in the next step (see Next step). ### The gaps reason (Yes, with conditions) For the rules large-gap-severity-override, total-gap-severity-override and evaluation-softens-not-now, use the rule line. Otherwise take the readiness area with the biggest gap (bar minus score; if no bar, 3 minus score) and write `Small gaps remain in your readiness. ` first for light-gaps, or `Several gaps remain in your readiness. ` first for heavier-gaps, then `Your weakest readiness area is DIMNAME: X of 5, where this problem needs Y.` If the area meets its bar: `Your weakest readiness area is DIMNAME, and it still meets what this problem needs (X of 5 against Y).` With no bar: `Your lowest readiness area is DIMNAME (X of 5).` Follow with the quote of the person's lowest answer in that area, when you can quote it and that answer is 3 or lower. If none of the five areas has a score, use the readiness line. ### The first-gap reason (Yes, once these change) Take the readiness area with the biggest gap and write one of these, with `, where this problem needs Y` left out when there is no bar: - total-gap-too-large: `Taken together, the gaps in your readiness are too large to start now. The largest is DIMNAME: X of 5, where this problem needs Y.` - evaluation-not-now: `The detailed questions you answered found readiness gaps that need closing first.` and, when the gap is over 0.05, ` The largest is DIMNAME: X of 5, where this problem needs Y.` - any other rule: `The gap to close first is DIMNAME: X of 5, where this problem needs Y.` Follow with the quote of the person's lowest answer in that area, when you can quote it and that answer is 3 or lower. If none of the five areas has a score, use the readiness line. ### The simpler-tool reason (No, use something simpler) `Something simpler would do the job: a ALTERNATIVE. It WHAT.` with the ALTERNATIVE of the lowest effort number in the table row for the use case's category. Do not mention readiness in this brief. ### The readiness line (No verdict yet, and a result with no area scored) If there is no verdict: `Readiness has not been judged for this problem, so nothing here rests on it.` If none of the five areas has a score: `None of the five readiness areas was answered, so readiness was not scored.` Otherwise take the readiness area with the biggest gap and write `Your weakest readiness area is DIMNAME: X of 5, where this problem needs Y.` If the area meets its bar: `Your weakest readiness area is DIMNAME, and it still meets what this problem needs (X of 5 against Y).` Follow with the quote of the person's lowest answer in that area, when you can quote it and that answer is 3 or lower. ## Quoting the person Quote only a question that is not sensitive. Name the question by its Ask line and the answer by its spoken label. Write a quote as a row. Under the reason, indented three spaces, write the line `You were asked | You said` once, then one row for each quote: `QUESTION | ANSWER`, with no quotation marks. For a readiness question the QUESTION is the Ask line of that question. Where this template says to write `Asked "QUESTION", you answered "ANSWER".`, `On "QUESTION", you answered "ANSWER".` or `On AREA, you chose "ANSWER".`, take the quote out of the sentence and write it as a row; the sentence keeps its conclusion and nothing else. A reason with no quote has no table. Do not quote an answer longer than 150 characters or a stored code. Never use a dash: write a comma. ## Next step - Yes: headline `Build the tender for LABEL.` Items: `Make a small pilot the first thing a supplier delivers.` and `Write down what the pilot must show before you scale it.` When the readiness area with the biggest gap is below its bar by more than 0.05, put this item first: `Close the nearest readiness gap during the pilot: DIMNAME, X of 5 where this problem needs Y.` followed by the action text for that area from the conditions table (the moderate text when the gap is over 0.6, the minor text otherwise; the small-organisation text when the organisation is small), with a capital first letter and a full stop. - Yes with conditions: headline `Meet this condition, then build the tender for LABEL.` (one) or `Meet these conditions, then build the tender for LABEL.` (several). Items below. - Yes, yes with conditions and yes once these change: the lines that say what any solution must meet do not go in the list of things to do. They go under their own heading, `What any solution must meet`, after Next step (see below). The providers line, for a yes or a yes with conditions, is worked out as follows: a use case whose required parts fall under two or more supplier families (the `f` entry of the use case in `data/brief-index.json`, which lists the `family` of each required part from `data/component-labels.json`, once each): makes a line that names them. Two families: `This needs A and B, and someone to join them up.` with A and B from this list: payments `a payments provider`, custody `a custody provider`, identity `an identity provider`, data platform `a data platform provider`, devices `a device supplier`, integration `an integration provider`, services `a services provider`. Three or more: `This needs providers for X, Y and Z, and someone to join them up.` with the topics payments, custody, identity, data, devices, integration, services. One family: no line. - A use case marked sensitive (`Sensitive: yes` in USE-CASES.md, or `s` in the use case's entry in `data/brief-index.json`), for a yes, a yes with conditions or a yes once these change: make the line `This use case handles sensitive personal data. Design the pilot with your data-protection lead from the first day.` It is the first line under `What any solution must meet`. A no, use something simpler, does not get it, because no pilot follows. - Yes, once these change: headline `Close this gap, then build the tender for LABEL.` (one gap) or `Close these gaps, then build the tender for LABEL.` (several). The items are the gaps, listed as conditions are: the script's conditions for the use case, most serious first, each as `DIMNAME: ACTION.` from the conditions table, with the same limit, the same `And N more.` and the same rule for the lines about the connection. For the area with the biggest gap, take the line from the roadmap table instead: `DIMNAME: ACTION (EFFORT, led by OWNER). First milestone: MILESTONE.` with ACTION as written in the table. Use the deep row (D4_deep or D5_deep) when that area scored under 2 and the table has one. If the script gives no conditions, the one line for the biggest gap is the whole list. - No, use something simpler: headline `Try a ALTERNATIVE first.` Items: `It WHAT.` and `If it does not solve the problem, run this again with what you learned.` ALTERNATIVE is the one with the lowest effort number in the table row for the use case's category. - No verdict: headline `Answer the readiness questions for LABEL, then run this again.` No items. Conditions. Take the script's conditions for the use case the block is about. Keep the significant and moderate ones; if there are none, keep them all. Most serious first. For a condition on one of the five readiness areas, write `DIMNAME: ACTION.` with DIMNAME's first letter capitalised and ACTION from the table below for the area and level (use the second wording if the organisation size is small). For any other condition, use the script's sentence, without the part that begins "Recommended first step". Show at most five (three when more than one problem has a block), then `And N more.` (N in words). If the script gives a stepping stone, add `Consider starting with NAME first. It builds what you need for this one.` NAME in lower case. A condition that reads `Connectivity gaps were reported. ...` or `You mostly work offline. ...` is a requirement. Nothing is there to meet. Do not count it in N (in the decision sentence, the headline, the short reasons or `And N more.`), never cut it with `And N more.`, and write it under `What any solution must meet` after the countries line. If it is the only condition, it stays a condition: keep it in the list and count it. ### The next step as a checklist and the requirements In a yes, a yes with conditions and a yes once these change, write each thing to do as `- [ ] ITEM`: a dash, a space, an empty box in square brackets, a space and the item. Do not put a box on `And N more.` or on `Consider starting with NAME first.`. The steps of a no, use something simpler, have no box. Then, for a yes, a yes with conditions and a yes once these change, write the heading `What any solution must meet` and one `- ` line each for, in this order and only those that apply: the sensitive-data line; the providers line; `Rules of COUNTRY, COUNTRY and COUNTRY apply.` from the countries in `irf_act0.countries` (put this line in the first block only when there is more than one block; leave it out when no country is known); the connectivity line from the conditions. Say nothing twice: a line that is under this heading is not in the list above it. If no line applies, leave out the heading. ### What you told us After the last block, write the heading `What you told us` and one `- ` line each for what you hold, in this order: the organisation type with a capital first letter; the size with a capital first letter; `Uses USES.` without the full stop; each country on its own line; the sector with a capital first letter. Leave out the heading when you hold none of them. These are the same values as the context lines at the foot, which stay. ### The readiness list For a yes, a yes with conditions and a yes once these change, when at least one of the five readiness areas has a score, write after the reasons the heading `Readiness, 0 to 5`, the line `Where this problem needs you to be, against where you are.` and one line for each scored area in the order of the table at the end: `- AREA: X of 5, needs Y.` with AREA's first letter capitalised and `, needs Y` left out when you have no bar. X and Y have one decimal. With more than one problem that has a block, give each block its own list, after its reasons. ### The question at the end After `What this brief leaves out` and before the context lines, one question that follows the verdict of the lead problem. The brief is advice, not a gate, so it is the same question after every answer that names a problem: `Take this forward to a tender?` then, on the next line, a sub-line. Yes, and yes with conditions: `The tender builder starts from this brief.` Yes, once these change: `The gaps above go into the tender as preconditions for the pilot.` No, use something simpler, and not much to change: `The brief says a simpler tool would do. If you go on, the tender will say so.` No viable use case: `Keep what you use today.` alone, left out when the Next step headline already says exactly that. In every other case, including no verdict yet and not enough information, write no question. ### Condition actions governance and authority (D1) - significant, standard: Secure a named executive sponsor who will champion this and allocate staff time to it - significant, small organisation: Ensure your director or lead decision-maker is actively engaged and supportive - moderate, standard: Brief senior leadership on the benefits and the resources this will need - moderate, small organisation: Make sure your director understands what a pilot involves and the time it takes - minor, standard: Document how this fits your current strategic plan, for donor reporting and internal approval - minor, small organisation: Note down how this supports your organisation's goals, for donor conversations financial resources (D2) - significant, standard: Develop a pilot budget estimate covering costs, staff time and any outside support - significant, small organisation: Estimate the total pilot cost, including subscriptions, training time and outside support - moderate, standard: Draft a budget for the pilot and confirm which budget line it sits under - moderate, small organisation: Estimate the pilot cost and confirm you can cover it from existing funds or a small grant - minor, standard: Document expected costs, so providers can prepare realistic proposals - minor, small organisation: Note your available budget range, so providers can tailor their proposals operational readiness (D3) - significant, standard: Map which teams and workflows this will affect and who needs to be involved - significant, small organisation: Map which of your current processes will change and who on your team is affected - moderate, standard: Identify where this will meet your current processes - moderate, small organisation: Walk through your current process and note where a new system would change how you work - minor, standard: Document your current workflow, so a provider can design a smooth fit - minor, small organisation: Write down your current process step by step, so a provider understands how you work today technical capability (D4) - significant, standard: Run a basic IT infrastructure check: internet reliability, devices and system access - significant, small organisation: Check that your team has reliable internet access and suitable devices - moderate, standard: Review which existing systems the new one must connect to and document the access it needs - moderate, small organisation: List the tools and systems you use now, so a provider knows what the new one must work with - minor, standard: Document your current setup (systems, devices, connectivity) for a provider's reference - minor, small organisation: Note what devices and internet access your team has today data readiness (D5) - significant, standard: Carry out a data protection impact assessment for the data this will handle - significant, small organisation: Review what personal or sensitive data this involves and confirm you can handle it responsibly - moderate, standard: Map the data flows this will create: what is collected, where it is stored, who has access - moderate, small organisation: List what data will be collected and from whom, and confirm your consent process covers it - minor, standard: Document your current data handling and privacy procedures for a provider's reference - minor, small organisation: Note your current approach to data protection, so a provider can fit the design to it ### Roadmap: the first step for each area - D1: milestone: a decision authority matrix agreed by leadership; action: Map who approves what for a new system, how long approval takes and where it stalls; effort: 2 to 4 weeks; owner: project lead and senior management - D2: milestone: a cost analysis showing the annual impact of the problem; action: Work out what the problem costs you each year, using simple estimates where exact figures are missing; effort: 2 to 4 weeks; owner: finance team and operations - D3: milestone: a named project owner with agreed time and authority; action: Name one person to own the project and give them time and authority; effort: 2 to 4 weeks; owner: HR or operations director - D4: milestone: a system inventory and data flow diagram; action: Make an inventory of your current systems and how data moves between them; effort: 2 to 3 weeks; owner: the most technical person on your team - D4_deep: milestone: an audit report with the infrastructure fixes put in order; action: Audit all systems, check how reliable your infrastructure is and find single points of failure; effort: 4 to 6 weeks; owner: the most technical person on your team - D5: milestone: a data quality report with error rates and priorities; action: Check the quality of the data most relevant to the problem; effort: 2 to 3 weeks; owner: data owner or monitoring and evaluation team - D5_deep: milestone: a data quality report with error rates and priorities; action: Audit data across all relevant systems for accuracy, completeness and consistency; effort: 3 to 6 weeks; owner: data owner or monitoring and evaluation team ### Simpler things to try, by area Each is a name, what it does, and an effort number from 1 (least) to 5. Use the lowest. The row dflt serves any area not listed. A (financial operations and payments): - cloud accounting system, effort 1: handles several currencies, donor fund tracking and automatic matching of payments to records - payment processing service, effort 1: handles transfers, mobile money and several currencies with compliance checks built in - financial management system, effort 5: covers finance, purchasing and reporting across several offices B (identity and credentials): - biometric registration system, effort 3: built for humanitarian settings and works offline - identity management system, effort 3: one place for identity checks, finding duplicates and controlling who can see what - case management system, effort 3: registers people, tracks their cases and reports across programmes C (data management and privacy): - data connection tool, effort 2: links your different systems and standardises the data without replacing them - master data system, effort 5: sets one set of data quality rules, finds duplicates and fixes the format - cloud data warehouse, effort 3: brings data from several sources into one place for reporting D (supply chain and verification): - supply chain management system, effort 4: tracks goods, stock and suppliers from end to end - QR and barcode tracking, effort 1: tracks assets and shipments with phones, QR codes and a shared dashboard - last-mile delivery tracking, effort 3: confirms deliveries to the final recipient with location, photos and digital signatures E (governance and coordination): - shared workspace, effort 1: documents, tasks and decision logs for several organisations in one place - governance management system, effort 2: records board decisions, votes and compliance checks - shared reporting dashboard, effort 3: one reporting view for several stakeholders, with automatic data collection F (fundraising and donor relations): - donor management system, effort 2: tracks donors and campaigns and automates reporting - grant management system, effort 3: follows each grant from application to financial report - public transparency page, effort 3: shows donors how funds were allocated and spent G (resilience and risk management): - tamper-evident evidence capture, effort 2: secure photo, video and document capture for field use, with a record of who handled it - early warning and early action system, effort 3: monitors hazards and triggers funding and action ahead of a crisis - parametric insurance or a risk pool, effort 4: pays out when an objective measure is crossed, faster than loss-adjusted insurance H (land, property and natural resources): - community land mapping, effort 2: community-led mapping and tenure records on mobile, offline-first tools that governments and donors recognise - case management for land claims, effort 3: tracks claims, disputes and resettlement with documents and records - cloud land registry, effort 5: formal registration, titling and transactions run by the land authority I (energy and environment): - remote-sensing monitoring, effort 3: satellite and ground checks of forests, emissions and land use - established carbon credit registry, effort 3: issuance and retirement of credits that buyers and regulators already trust - renewable energy certificate scheme, effort 4: tracks and trades renewable energy claims across borders with audited claims J (workforce and operations): - volunteer and workforce system, effort 1: recruits, schedules and tracks the hours of staff and volunteers - learning system with open badges, effort 2: delivers training and gives portable proof of skills - duty of care system, effort 3: handles travel risk, incident reporting and worker welfare for field staff any other area: - better use of your current tools, effort 1: simplify the steps you follow today and automate what you can, before buying anything - ready-made cloud service, effort 2: a specialist service for this problem, with low set-up cost - custom-built application, effort 5: built to your needs when nothing off the shelf fits ## What this brief leaves out Up to four lines, after the line about the first three problems when that applies. Say only what is true of this run: - If fewer than five readiness areas were answered for the lead (and at least one was): `Only N of five readiness areas were answered for LABEL.` - For each other problem without a verdict because the organisation questions were unanswered: `No decision for NAMES: the questions about your organisation were not answered, so readiness could not be judged.` (names joined with semicolons) - For each without a verdict because its own questions were unanswered: `No decision for NAMES: the questions about this problem were not answered.` (or `these problems were not answered`) - If the best-ranked use cases have no readiness check: `Also ranked high, but not yet checked for readiness: NAME; NAME; and N more.` (two names at most) - If you could not run the script: `The ranking and verdict were worked out without the scoring script, so treat them as approximate. Run the script to check them.` - If a category ended below its floor: `Ranking in CATEGORY NAME is based on N of M questions.` ## The context lines One line each, only for what you hold, in this order: `Prepared for JOB TITLE, ORGANISATION.` (either part may be missing); `Organisation: TYPE, SIZE.` (when `irf_act0.uses` holds an answer, add ` You use USES.` to this line, for example `You use spreadsheets.`); `Works in COUNTRY, COUNTRY and COUNTRY.` (list with commas and `and`; with more than five, the first four and `and N more`); `Sector: SECTOR.` Sector names: - humanitarian: humanitarian - environmental: environmental - healthcare: health - microfinance: microfinance - education: education - agriculture: agriculture - human_rights: human rights - wash: water and sanitation - community: community development ## Technology words These must not appear before the Why heading (the decision sentence and the lines under it): blockchains?, ledgers?, dlt, smart contracts?, tokens?, tokeni[sz]\w*, crypto\w*, stablecoins?, wallets?, on-chain, off-chain, decentrali[sz]\w*, immutab\w*, hash(?:es|ed|ing)?, consensus, nodes?, web3, nfts?, daos?, bitcoin, ethereum, digital assets?, zero-knowledge, technology, technologies, technical, software, algorithms?. ## Words you must not use in a brief No dash as punctuation, no exclamation mark, no product or company name, no precedent, no score offered as a prediction, no category letter and no question id.