Mission goal: understand what the AI-enhanced version of your role is and why it is emerging.
Step 1 of 5Where are you starting from?
Choose your starting point+10 XP
Where are you today with AI and automation in your coordination work?
Show answer
Every starting point is valid. Follow the mission suggested for your situation.
Step 2 of 5Meet your new role
Today's role
M&E Coordinator
AI-enhanced role
AI-Enabled M&E Coordinator
Level
Specialist · 3–6 years
How the role uses AI
Integrates AI: builds AI-assisted workflows to consolidate data and automate reporting schedules
"AI-Enabled M&E Coordinator" is a proposed career title from the EvalCommunity framework, not a standardised occupational classification or a guaranteed vacancy.
Step 3 of 5What changes most in your job?
Your guess+10 XP
When the chasing, copying and reformatting are automated, where does your time go?
Show answer
In the profile, chasing, copying and reformatting data can now be largely automated. The coordinator's role shifts to designing those workflows, checking that consolidated data is correct and using the time saved to follow up with partners on real implementation issues.
Step 4 of 5AI-Enabled, not AI-Dependent
The goal is not to hand your work to AI. It is to become a stronger M&E coordinator who uses AI and stays in charge of it. Here are the five questions, asked about your own work:
1QualityDo your automated workflows make consolidated data more reliable, and do you check them?Every new indicator mapping and every extracted figure is checked before it enters the tracker.
2AccountabilityWhen a consolidated figure is wrong, can anyone trace which step changed it?A log of every automated step, with its check and its owner.
3IndependenceCan you still tell a template issue from a real implementation problem?You call the partner before you decide, instead of trusting the flag.
4EthicsDo partners' and participants' data move only where they should?Workflows carry aggregated figures, not personal data, and partners know what is automated.
5Professional developmentAre you and your partners learning how the workflows work, or only depending on them?Others can maintain the workflows from your log, and partners are trained on the templates.
The five questions as the series states them
Quality: Does AI improve the quality and credibility of evaluation findings?
Accountability: Who is responsible when AI produces an incorrect interpretation?
Independence: Are evaluators maintaining their ability to think critically and make independent judgments?
Ethics: Are confidentiality, human rights and the interests of affected populations protected?
Professional development: Are we developing stronger evaluators who understand AI, or creating professionals who increasingly depend on it?
Step 5 of 5Your goal for this tutorial
By the end, you will know how to automate multi-partner reporting step by step, with a human check at every step, which skills to build, and you will have a 12-week plan.
Next: see which of your tasks change, and how.
End of Mission 01
MISSION02
Traditional role vs AI-enhanced role
Before and after · Badge: Role Shifter
8 challenges
Mission goal: see which responsibilities become AI-assisted, which need new skills and which stay human.
Step 1 of 4Meet your case study
Case study · Community Resilience Consortium Illustrative
A fictional consortium you will follow through missions 2, 3, 4, 8 and 9.
The consortium
Five projects (water, livelihoods, health, education and nutrition) run by eleven partners from three field offices.
The data
Monthly returns from eleven partners in different templates, three of them as PDF narrative reports, and a shared workplan.
The tools
Excel, Airtable, SharePoint, Power Automate, n8n, an LLM API, Claude, Microsoft 365 Copilot, Power BI and KoboToolbox.
Your objective
Get every return in on time, in one format, and turn it into one checked picture of progress across the five projects.
Step 2 of 4Flip your tasks: today and with AI
Less chasing and copying, more designing and checking. Tap each card to see how it changes.
Step 3 of 4Mini-game: sort the work
For each task, choose: AI-assisted, a new skill for you, or stays human.
Sort it+10 XP
Sending deadline reminders to partners
Show answer
AI-assisted. Automated: reminders and submission tracking (the profile's example).
Sort it+10 XP
Mapping partners' indicator labels to standard codes
Show answer
AI-assisted. An AI step suggests the mapping; you review every new one.
Sort it+10 XP
Extracting beneficiary numbers from PDF reports
Show answer
AI-assisted. AI extracts; you check each figure against the PDF.
Sort it+10 XP
Designing a workflow that collects and consolidates returns
Show answer
New skill. Workflow automation is the first learning priority in the profile.
Sort it+10 XP
Understanding how data moves between tools when a workflow breaks
Show answer
New skill. Data pipelines are one of the profile's learning priorities.
Sort it+10 XP
Deciding whether a mismatch is a template issue or a real problem
Show answer
Human. Listed in the profile as human judgement.
Sort it+10 XP
Prioritising follow-up when several projects are behind at once
Show answer
Human. Listed in the profile as human judgement.
Sort it+10 XP
Building working relationships so partners raise problems early
Show answer
Human. Listed in the profile as human judgement.
Step 4 of 4Your turn: two minutes
Think about last month's reporting round. Which tasks would go in each pile?
Think about it for a moment, then reveal.
A quick way to do it: three columns on a sheet of paper, AI-assisted, new skill, stays human. Keep the list: you will use it in mission 4 (to pick a task to try) and mission 6 (to rate your gaps).
End of Mission 02
MISSION03
What AI actually changes in your daily work
Checks, risks and red flags · Badge: Red Flag Spotter
2 challenges
Mission goal: learn the three checks every automated workflow needs, and practise spotting problems in an AI draft.
Step 1 of 4What can AI really do for you?
Before you look: which of your tasks could AI or automation help with, and which must stay with you?
Think about it for a moment, then reveal.
AI can assist with
Mapping indicator labels to standard codes
Extracting figures from PDF reports
Drafting cross-project summaries
Comparing progress with the workplan
Can be partly automated
Deadline reminders
Recording submissions
Weekly lists of missing reports
Collecting returns into one tracker
New responsibilities
Designing workflows with human checks
Reviewing every new mapping
Logging each automated step
Fixing workflows when they break
Stays with you
Relationships with partners
Template issue or real problem?
Which project to follow up first
Mappings where meanings differ
Step 2 of 4The three checks you must always build in
An automated mistake spreads silently into every consolidated report. Here are the checks, one at a time.
Review every new mapping. No AI-mapped indicator enters the tracker until a person has compared its definition and unit.
Trace every extracted figure. Check each number taken from a PDF against its page.
Log every automated step. What it does, what data it touches, who checks it and who fixes it.
Step 3 of 4Mini-game: spot the red flags
Case study · Community Resilience Consortium Illustrative
An AI assistant drafted the Monday status note from the submission tracker. Here is the source and the draft.
Click every sentence you would not send to the M&E manager.
Submission tracker, Monday
Returns received
8 of 11
Missing
Partners 2, 5 and 7
Partner 5
Extension to Friday agreed last week
Automated reminders
Sent 7 and 2 days before the deadline to all partners
Red flags found: 0 of 3+30 XP
Show answer
Three sentences should not be sent: the warning to Partner 5 (it has an agreed extension), "Partner 7 always reports late" (one month is not a pattern), and the claim that late reporting is caused by weak capacity (no evidence).
Step 4 of 4Mission check
Your call+20 XP
An AI step mapped 40 partner indicator labels to your standard codes overnight. What happens before the data enters the tracker?
Show answer
Review every new mapping, by definition and unit, before the data enters the tracker.
End of Mission 03
MISSION04
Your daily tasks, step by step with AI
Six practical guides · Badge: Workflow Builder
6 challenges
Mission goal: learn how to automate multi-partner reporting in six of your real tasks, with a human check at every step.
Case study · Community Resilience Consortium Illustrative
Every worked example below uses this consortium's data, so you can follow one system from partner returns to the progress summary.
Six real tasks from your working week. Pick one and go through it in six short steps: a task prompt, then a QA prompt that checks the AI's work.
YouAIYour teamAutomated shows who does each step. Examples are illustrative.
Choose a task
Guide 1 of 6 From the career profile
Automate reminders and track partner submissions
From the profile: Power Automate sends partners reminders seven and two days before each deadline, records submissions in SharePoint and emails the coordinator a list of missing reports each Monday (eleven partners in the daily task).
Step 1 of 6Today, and with AI
How it is usually done today Illustrative
Deadlines live in a spreadsheet, and reminders are sent by hand.
Missing reports are discovered on the deadline day.
A deadline moved by a donor is missed by someone.
How it works with AI
Reminders go out automatically, only to partners who haven't submitted.
Every submission is recorded with its date.
Each Monday you get the list of missing reports, check it and call.
Step 2 of 6Step by step
YouBuild one reporting calendarSharePoint · Excel
Every partner, report and deadline in one list, with agreed extensions.
AIAsk AI to draft the workflow stepsClaude
Describe the calendar and the rules; ask for numbered steps you can build (prompt below).
AutomatedSend reminders 7 and 2 days beforePower Automate
Only to partners whose report is not yet recorded.
AutomatedRecord submissions and list the missing onesSharePoint · Power Automate
Each Monday, the coordinator gets the list of missing reports.
YouCheck the list, then call
A report sent by email may not be recorded. Check before you chase.
YouChange deadlines once, at the source
When a donor moves a deadline, update the calendar; the reminders follow.
Step 3 of 6Worked example
What you give the AI Illustrative
Partner
Report
Deadline
Submitted
Partner 1
Monthly return
5 May
3 May
Partner 2
Monthly return
5 May
Partner 5
Monthly return
5 May (extension to 9 May)
What would you ask the AI to do with this?
Think about it for a moment, then reveal.
The task prompt (copy and adapt)
Here is our reporting calendar with these columns: partner, report, deadline, agreed extension, submitted date.
Draft the steps of a scheduled workflow that:
1) sends a reminder 7 and 2 days before each deadline, only if the report has not been submitted, and uses the extension date where there is one;
2) records each submission and its date;
3) every Monday emails me the list of missing reports.
Write numbered steps a Power Automate user can build. Mark each step where I should check the result. Do not include personal data.
What comes back
1. Each morning, read the calendar list.
2. For each report with no submitted date, if the deadline (or extension) is 7 or 2 days away, send the reminder. Check: the first week's reminders.
3. When a report arrives in the submissions folder, write the date into the list. Check: reports sent by email are added by hand.
4. Every Monday, email the coordinator the reports past their deadline or extension. Check: the list, before any call.
Step 4 of 6Check the AI output
Independence first: read the draft and note the problems you see yourself, before you run the QA prompt. Then compare.
Another draft came back like this Illustrative
1. Every day, email all partners a reminder about their next report.
2. On the deadline, mark every report not yet received as late.
3. Every Monday, email the list of late partners to the programme director.
The QA prompt (run it on every draft)
Act as an independent reviewer of an automated reminder workflow.
Compare the steps below with the requirements.
List: every reminder sent to someone who should not get it, every timing that differs from the requirements, every case the steps ignore (such as agreed extensions), and every step that shares information with the wrong person.
Do not rewrite the steps. List the problems only.
Requirements: [paste]
Steps: [paste]
What should the check catch?
Think about it for a moment, then reveal.
What the check finds
Daily reminders go to every partner, including those who have already submitted. The requirement is 7 and 2 days before, to non-submitters only.
Agreed extensions are ignored, so partners with an extension are marked late.
The Monday list goes to the programme director instead of you, before anyone has called the partners.
Corrected output
1. Seven and two days before each deadline, send a reminder only to partners whose report is not yet recorded.
2. Record each submission, with its date, in the SharePoint list.
3. Skip partners with an agreed extension until their new date.
4. Every Monday, email the coordinator the list of missing reports. The coordinator checks it and calls partners before anything is escalated.
Step 5 of 6Your checks
Tick each check as you do it on your own work. Each one earns +5 XP.
Accountability: the workflow sends the reminders, but you still own the deadline list and check it every Monday.
Step 6 of 6Quick check
Quick check+25 XP
The Monday list shows Partner 6 as missing, but they emailed you the report on Friday. What do you do?
Show answer
Record the submission and fix the gap, for example by forwarding emailed reports to the submissions folder.
Try it this week
Put next month's deadlines for three partners in one list with a "submitted" column, and set one reminder seven days before.
Map partners' indicator names to your standard codes
From the profile: an n8n workflow collects monthly returns from eleven partners, an LLM step maps their indicator labels to the programme's codes, and the coordinator reviews every new mapping before data enters Airtable.
Step 1 of 6Today, and with AI
How it is usually done today Illustrative
Each partner names its indicators differently.
You rename them by hand in a master spreadsheet.
A unit mix-up (women or households?) slips into the totals.
How it works with AI
An AI step suggests a standard code for each new label.
You review every new mapping, by definition and unit.
Approved mappings are reused, so only new labels need review.
Step 2 of 6Step by step
YouKeep one standard code listAirtable · Excel
Each code with a definition and a unit (households, women, visits).
AutomatedCollect the returnsn8n
The workflow gathers partner returns into one place.
AISuggest a code for each new labelLLM API
The AI step gets the label and unit only, never beneficiary data (prompt below).
YouReview every new mappingAirtable
Compare definitions and units, not just names. Anything uncertain is marked REVIEW.
Your teamAsk the partner when meaning differs
For example: does "HH visits" count visits or households?
AutomatedReuse approved mappingsAirtable
Next month, only new labels go to review.
Step 3 of 6Worked example
What you give the AI Illustrative
Partner label
Unit
Partner
No. of HH trained in VSLA
Households
Partner 3
Households trained - savings groups
Households
Partner 7
Women trained in savings
Women
Partner 8
HH visits (CHW)
Not stated
Partner 2
Standard codes: L1.2 Households trained in savings groups · L1.3 Women trained in savings groups · H2.1 Home visits by community health workers (visits).
What would you ask the AI to do with this?
Think about it for a moment, then reveal.
The task prompt (copy and adapt)
Map each partner indicator label below to one code from the standard list.
For each, give the code, a one-line reason, and whether the unit matches the code's unit.
If no code fits, or the meaning or unit may differ, write REVIEW instead of a code.
Do not create new codes.
Standard codes: [paste codes with definitions and units]
Labels: [paste labels with units]
What comes back
Partner label
Suggested code
Unit matches?
No. of HH trained in VSLA
L1.2
Yes
Households trained - savings groups
L1.2
Yes
Women trained in savings
L1.3
Yes (women)
HH visits (CHW)
REVIEW
Unclear: visits or households?
Three suggestions to review, and one question for Partner 2.
Step 4 of 6Check the AI output
Independence first: read the draft and note the problems you see yourself, before you run the QA prompt. Then compare.
Another draft came back like this Illustrative
Partner label
AI mapping
Women trained in savings (Partner 8)
L1.2 Households trained in savings groups
HH visits (CHW) (Partner 2)
H2.1 Home visits by community health workers
Savings training, other (Partner 10)
L1.4 Savings training, other (new code)
The QA prompt (run it on every draft)
Act as an independent reviewer of an indicator mapping.
Compare each mapping below with the standard code list and its definitions.
List: every mapping where the unit differs (for example women vs households), every mapping where the meaning may differ, and every code that is not in the standard list.
Do not rewrite the mapping. List the problems only.
Standard codes: [paste]
Mapping: [paste]
What should the check catch?
Think about it for a moment, then reveal.
What the check finds
Partner 8 counts women, but L1.2 counts households. The right code is L1.3.
"HH visits" may count households visited, while H2.1 counts visits. Ask Partner 2 before mapping.
L1.4 is not in the standard list. The AI must not create codes.
Corrected output
Partner label
Reviewed mapping
Women trained in savings (Partner 8)
L1.3 Women trained in savings groups
HH visits (CHW) (Partner 2)
REVIEW: ask Partner 2 whether it counts visits or households
Savings training, other (Partner 10)
REVIEW: ask Partner 10 what is counted; no new code
Step 5 of 6Your checks
Tick each check as you do it on your own work. Each one earns +5 XP.
Quality: every new mapping is reviewed before data enters the tracker. A wrong code spreads to every report.
Step 6 of 6Quick check
Quick check+25 XP
The AI maps Partner 2's "HH visits" to your code for home visits. Partner 2 counts households visited. What do you do?
Show answer
Mark it REVIEW, agree the unit with Partner 2, then map it.
Try it this week
Take ten indicator labels from last month's partner returns. Ask an AI assistant to map them to your codes, then review each mapping by definition and unit.
From the profile: three partners still send PDF narrative reports; Claude extracts beneficiary numbers by activity into a table, and the coordinator checks each figure against the PDF before consolidation.
Step 1 of 6Today, and with AI
How it is usually done today Illustrative
You read each PDF and type the numbers into the tracker.
Figures in the text are easy to miss or double count.
Nobody records which page a figure came from.
How it works with AI
Claude extracts the figures by activity, with the page.
You check each figure against the PDF page.
Unclear figures go back to the partner.
Step 2 of 6Step by step
YouCollect the PDF reports
Check they contain no personal data before you upload them to an AI tool. If they do, remove it or don't upload.
AIExtract the figures with page referencesClaude
Ask for the activity, the number, the page and the exact words (prompt below).
YouCheck every figure against its page
Open the page. Compare the number and what it counts.
YouDon't let the AI add things up
Totals come from the report or from you, not from the AI.
Your teamAsk the partner about unclear figures
For example: is the refresher counted in the total?
AutomatedEnter checked figures into the trackerExcel · Airtable
With the page reference, so anyone can trace them.
Step 3 of 6Worked example
What you give the AI Illustrative
Partner 4, narrative report, April, p. 3 (excerpt): "In April we held 14 training sessions on savings groups, reaching 312 participants (189 women). A further 40 participants attended the refresher in Ward 3, which is not included above."
What would you ask the AI to do with this?
Think about it for a moment, then reveal.
The task prompt (copy and adapt)
From the report text below, extract every beneficiary number by activity into a table with these columns: activity, number, women (if stated), page, exact words from the report.
Do not add numbers together and do not estimate.
If it is unclear what a number counts, write CHECK.
[paste report text with page numbers]
What comes back
Activity
Number
Women
Page
Exact words
Training sessions (14)
312
189
p. 3
"reaching 312 participants (189 women)"
Refresher, Ward 3
40
Not stated
p. 3
"not included above"
Two separate figures. Whether to add them is a decision for you, with the partner.
Step 4 of 6Check the AI output
Independence first: read the draft and note the problems you see yourself, before you run the QA prompt. Then compare.
Another draft came back like this Illustrative
Activity
Participants
Women
Page
Training sessions
352
198
The QA prompt (run it on every draft)
Act as an independent reviewer of data extracted from a report.
Compare the table below with the report text.
List: every figure that differs from the text, every total the AI added up that the report does not state, and every row without a page reference.
Do not rewrite the table. List the problems only.
Report text: [paste]
Table: [paste]
What should the check catch?
Think about it for a moment, then reveal.
What the check finds
352 adds the 40 refresher participants, which the report says are not included in the 312.
Women are 189 in the report, not 198.
The page reference is missing, so nobody can check the figure.
Corrected output
Activity
Participants
Women
Page
Training sessions (14)
312
189
p. 3
Refresher, Ward 3 (reported separately)
40
Not stated
p. 3
Step 5 of 6Your checks
Tick each check as you do it on your own work. Each one earns +5 XP.
Accountability: each extracted figure is checked against its page in the PDF before consolidation.
Step 6 of 6Quick check
Quick check+25 XP
Claude's table says 352 participants. The report says 312, plus 40 at a refresher "not included above". What goes in the tracker?
Show answer
312 for the training, 40 for the refresher separately, each with its page.
Try it this week
Take one partner PDF report. Ask an AI assistant to extract the figures with page references, then check every figure against its page.
From the profile: Copilot drafts a one-page summary of implementation progress across five projects from the consolidated tracker; the coordinator adds context from partner calls and removes speculative explanations.
Step 1 of 6Today, and with AI
How it is usually done today Illustrative
You write the summary from scratch, the day before the meeting.
Explanations are guessed rather than checked.
Some projects get more attention than others.
How it works with AI
Copilot drafts one sentence per project from the tracker.
You add context from partner calls, with the date.
You remove every explanation nobody gave you.
Step 2 of 6Step by step
YouClose the tracker for the periodExcel · Airtable
Note missing returns, so incomplete figures are labelled.
AIAsk Copilot for a draftMicrosoft 365 Copilot
One sentence per project, figures only, with [CONTEXT] where an explanation is needed (prompt below).
YouCheck every figurePower BI
Against the tracker, project by project.
Your teamCall partners for context
Ask what happened, and note who said it and when.
YouReplace [CONTEXT] and remove guesses
Only explanations from partner calls or reports. Delete the rest.
YouSend it to the M&E manager
One page, every project covered, incomplete figures labelled.
Step 3 of 6Worked example
What you give the AI Illustrative
Project
Result
Target
Note
A Water
38 water points
40
B Livelihoods
620 households trained
900
C Health
1,450 home visits
1,600
D Education
2,100 pupils
2,400
Partner 9 return missing
E Nutrition
0 screening campaigns
3
What would you ask the AI to do with this?
Think about it for a moment, then reveal.
The task prompt (copy and adapt)
Using only the tracker rows below, draft a one-page progress summary with one sentence per project: result, target and percentage of target.
Label any figure that is incomplete.
Where a project is more than 15 points behind, write [CONTEXT] instead of explaining why.
Do not add causes, predictions or figures that are not in the table.
[paste tracker rows]
What comes back
A: 38 of 40 water points (95%).
B: 620 of 900 households trained (69%). [CONTEXT]
C: 1,450 of 1,600 home visits (91%).
D: 2,100 of 2,400 pupils (88%); incomplete, Partner 9's return is missing.
E: 0 of 3 screening campaigns (0%). [CONTEXT]
The [CONTEXT] lines are filled from partner calls, not by the AI.
Step 4 of 6Check the AI output
Independence first: read the draft and note the problems you see yourself, before you run the QA prompt. Then compare.
Another draft came back like this Illustrative
Project B is behind (79% of target) because of the rainy season. Project D reached 2,100 pupils. Projects A, C and D are on track.
The QA prompt (run it on every draft)
Act as an independent reviewer of a progress summary.
Compare the summary below with the tracker.
List: every figure that differs from the tracker, every explanation the tracker or partner calls do not give, every project left out, and every "on track" the figures do not support.
Do not rewrite the summary. List the problems only.
Tracker: [paste]
Partner call notes: [paste]
Summary: [paste]
What should the check catch?
Think about it for a moment, then reveal.
What the check finds
Project B is at 69% (620 of 900), not 79%.
"Because of the rainy season" is a guess: no partner call says so.
Project E (no screening campaign started yet) is left out.
"D is on track" ignores that Partner 9's return is missing, so its figure is incomplete.
Corrected output
Project A: 38 of 40 water points (95%). Project B: 620 of 900 households trained (69%); the partner says two trainers left in May (call, 12 June). Project C: 1,450 of 1,600 home visits (91%).
Project D: 2,100 pupils reached so far, but Partner 9's return is missing, so the figure is incomplete. Project E: none of the three screening campaigns planned for this quarter has started; the partner is waiting for supplies (call, 13 June).
Step 5 of 6Your checks
Tick each check as you do it on your own work. Each one earns +5 XP.
Independence: context comes from partner calls, not from AI guesses. Remove speculative explanations.
Step 6 of 6Quick check
Quick check+25 XP
Copilot writes that Project B is behind "because of the rainy season". No partner said so. What do you do?
Show answer
Delete it, and add the reason the partner gives you, with the date of the call.
Try it this week
Before the next progress meeting, ask Copilot or another assistant for a figures-only draft with [CONTEXT] gaps, then fill them from partner calls.
Flag projects behind their workplan and decide what to chase
From the profile: flagging projects whose activity progress lags behind their workplan using AI-assisted comparisons; deciding whether a mismatch is a template issue or a real implementation problem; prioritising follow-up when several projects are behind at once.
Step 1 of 6Today, and with AI
How it is usually done today Illustrative
Delays come up when a project manager mentions them.
Every delay gets the same follow-up, or none.
A missing return looks like a delay.
How it works with AI
AI compares completed and planned activities by project.
You find out whether each lag is real or a template issue.
You decide which project to follow up first.
Step 2 of 6Step by step
YouSet the rule
For example: fewer than 80% of planned activities completed to date.
AutomatedPrepare the comparison tableExcel · Power BI
Planned and completed activities to date, by project.
AIAsk the AI to list, not explainClaude
Projects below the rule, with the percentage (prompt below).
Your teamCall each partner
Is it a real delay, a reporting gap or a template problem?
YouDecide the priority
Risk to people, fixed seasons and deadlines count, not only the percentage.
YouTell the project managers
What is behind, why, and what you agreed.
Step 3 of 6Worked example
What you give the AI Illustrative
Project
Activities planned to date
Completed
A
12
11
B
10
6
C
8
8
D
9
7
E
3
0
What would you ask the AI to do with this?
Think about it for a moment, then reveal.
The task prompt (copy and adapt)
Below are planned and completed activities to date by project.
List every project that has completed fewer than 80% of its planned activities, with the percentage.
Do not explain why and do not rank the projects.
[paste table]
What comes back
Project
Completed / planned
B
6 of 10 (60%)
D
7 of 9 (78%)
E
0 of 3 (0%)
Three projects to call. The priority is your decision.
Step 4 of 6Check the AI output
Independence first: read the draft and note the problems you see yourself, before you run the QA prompt. Then compare.
Another draft came back like this Illustrative
Projects B and E are behind plan. Project C is also slipping. Follow up E first, then B, because E has the lowest percentage.
The QA prompt (run it on every draft)
Act as an independent reviewer of a workplan comparison.
Compare the list below with the table and the rule.
List: every project that meets the rule but is missing, every project listed that does not meet the rule, and every priority that is based only on the percentage.
Do not rewrite the list. List the problems only.
Rule: [paste]
Table: [paste]
List: [paste]
What should the check catch?
Think about it for a moment, then reveal.
What the check finds
Project D (7 of 9, 78%) is below the 80% rule but missing.
Project C (8 of 8) is not slipping.
Ranking by percentage alone is your call to make, not the AI's: risk to people, deadlines and what the partner says also count.
Corrected output
Project
Completed / planned
What the partner said
Your decision
E
0 of 3 (0%)
Waiting for screening supplies
Follow up first: health risk and a fixed season
B
6 of 10 (60%)
Two trainers left in May
Real delay: agree a catch-up plan
D
7 of 9 (78%)
Partner 9's return is missing
Template issue: get the return before judging
Step 5 of 6Your checks
Tick each check as you do it on your own work. Each one earns +5 XP.
Independence: you decide whether a lag is a template issue or a real implementation problem, after talking to the partner.
Step 6 of 6Quick check
Quick check+25 XP
Project D is at 78%, but Partner 9's return for D is missing. What is it?
Show answer
Get the missing return first, then decide whether D is really behind.
Try it this week
Run the 80% rule on this quarter's workplan. Call one partner about a flagged project and record whether it is a real delay or a reporting gap.
From the profile: keeping a log of automated steps and their checks so others can maintain the workflows.
Step 1 of 6Today, and with AI
How it is usually done today Illustrative
Workflows live in one person's head.
When a step breaks, nobody knows what it did.
Nobody is sure what data each tool receives.
How it works with AI
AI drafts the log from your workflow description.
You complete the checks, data and owners.
A colleague can restart a workflow from the log.
Step 2 of 6Step by step
YouList every automated step
Collect, map, extract, remind, consolidate: one line each.
AIAsk AI to draft the logClaude
One row per step, with [TO CONFIRM] where it doesn't know (prompt below).
YouState the data each step touches
Say clearly whether personal data is involved. AI steps get labels or aggregates only.
YouName the human check and the owner
A role who can fix the step, and a back-up.
Your teamTest it with a colleague
Can they restart one workflow using only the log?
YouTell partners what is automatedSharePoint
Which data is processed by which tool, and who checks it.
Step 3 of 6Worked example
What you give the AI Illustrative
Current workflow: n8n collects partner returns from a shared inbox; an LLM step maps indicator labels to codes; approved data goes into Airtable; Power Automate sends deadline reminders; Copilot drafts the monthly summary.
What would you ask the AI to do with this?
Think about it for a moment, then reveal.
The task prompt (copy and adapt)
Draft an automation log for the workflow described below.
One row per automated step, with these columns: what it does, tool, data it touches (and whether personal data is involved), human check, owner, what to do if it fails.
Write [TO CONFIRM] wherever the description does not say.
[paste workflow description]
What comes back
Step
Data it touches
Human check
Owner
Collect returns (n8n)
Partner return files [TO CONFIRM: personal data?]
[TO CONFIRM]
[TO CONFIRM]
Map labels (LLM)
Indicator labels and units
[TO CONFIRM]
[TO CONFIRM]
Reminders (Power Automate)
Partner contact emails
[TO CONFIRM]
[TO CONFIRM]
The [TO CONFIRM] cells are your decisions.
Step 4 of 6Check the AI output
Independence first: read the draft and note the problems you see yourself, before you run the QA prompt. Then compare.
Another draft came back like this Illustrative
Step
Data it touches
Human check
Owner
n8n collects returns from the shared inbox
All partner files
None
IT
LLM maps indicator labels
All partner files
None
IT
The QA prompt (run it on every draft)
Act as an independent reviewer of an automation log.
Check each step below.
List: every step with no human check, every vague description of the data a step touches (say whether personal data is involved), and every owner who could not actually fix the step.
Do not rewrite the log. List the problems only.
Log: [paste]
What should the check catch?
Think about it for a moment, then reveal.
What the check finds
The LLM mapping step has no human check. Every new mapping must be reviewed before data enters the tracker.
"All partner files" is vague. The LLM step should receive indicator labels only, never beneficiary lists.
"IT" is not an owner who knows the workflow. Name a role, with a back-up.
Corrected output
Step
Data it touches
Human check
Owner
n8n collects returns from the shared inbox
Partner return files (aggregated figures; no personal data by agreement)
Coordinator checks the Monday list
M&E Coordinator; back-up: M&E Officer
LLM maps indicator labels
Indicator labels and units only
Coordinator reviews every new mapping
M&E Coordinator; back-up: M&E Officer
Step 5 of 6Your checks
Tick each check as you do it on your own work. Each one earns +5 XP.
Ethics: the log records what data each step touches, so personal data never flows into tools that should not have it.
Step 6 of 6Quick check
Quick check+25 XP
The n8n workflow stops while you are on leave. What makes it fixable?
Show answer
The automation log, with a named back-up owner for each step.
Try it this week
Write down the automated steps you already use, even simple ones. For each, add the data it touches, the check and the owner.
Pick a recurring task that is not covered above, such as preparing the data pack for a quarterly partner review, and use this prompt to draft your own step-by-step guide.
I am an M&E coordinator working with several partners. Here is a recurring task I do: [describe the task, how often, who is involved, which tools and data].
Break it into numbered steps. For each step, say who should do it (me, a partner, a colleague, an automated workflow or an AI assistant), which tool, and what could go wrong.
Mark every step where a person must check the result before it is used, and say what they should check.
Do not suggest putting personal data into an AI tool.
End of Mission 04
MISSION05
The skills you need to develop
Skill inventory · Badge: Skill Builder
1 challenge
Mission goal: find out which of the 12 profile skills you already have and which need new learning.
Step 1 of 5Guess first
Your guess+20 XP
Which skill is the first learning priority for this role in the career profile?
Show answer
Workflow automation, followed by AI document extraction, Airtable, validating AI outputs and data pipelines.
Step 2 of 5Skills you bring with you
Tick the ones you already have. Open a skill to see how to build it.
Your skills match
0 of 12
A. Existing M&E and professional competencies
Transfers from your role
M&E plans, IPTTs, data flow maps and reporting calendars: the backbone every workflow automates.
Build it: Draw the data flow for one monthly reporting round.
Transfers from your role
Workplans, timelines and risk logs. You need them to judge whether a project is really behind.
Build it: Compare one project's workplan with its reported progress.
Transfers from your role
Working relationships with partners, so problems are raised early.
Build it: Call one partner each week for five minutes, without a deadline to chase.
Transfers from your role
Recoding, merging and documenting cleaning decisions across partner templates.
Build it: Write down the cleaning rules you apply to partner returns.
Transfers from your role
Definitions, units and disaggregation. You need them to review AI mappings.
Build it: Add a unit to every code in your standard list.
Transfers from your role
Training partners on templates and data quality.
Build it: Run a 30-minute session for partners on the standard template.
Transfers from your role
Validity, reliability, timeliness, precision and integrity, across partners.
Build it: Trace three consolidated figures back to partner returns.
Transfers from your role
Clear summaries without guesses about causes.
Build it: Remove every unsupported explanation from one AI-drafted summary.
Step 3 of 5New skills for the AI era
These are the ones most coordinators still need to learn.
Your skills match
0 of 12
B. New AI and data competencies
Needs new learning
Triggers, scheduled flows and integrations that remove repetitive steps. The profile's first learning priority.
Build it: Build the reminder workflow from guide 1 for three partners.
Needs new learning
Pulling structured data from reports, PDFs and forms with AI (a learning priority).
Build it: Extract figures from one PDF report and check each one (guide 3).
Needs new learning
Checking AI results against source data. Mistakes in automated mapping spread silently (a learning priority).
Build it: Review ten AI mappings by definition and unit (guide 2).
Needs new learning
How data moves between collection tools, databases and dashboards, so you can fix a broken workflow (a learning priority).
Build it: Draw how one partner figure travels from return to dashboard.
Step 4 of 5Your five learning priorities
In this order, from the career profile:
Workflow automation. Automating reminders and consolidation removes the most repetitive part of the job.
AI document extraction. Many partners still report in PDFs and Word files that need converting to data.
Airtable. A shared tracker with forms and views suits multi-partner coordination better than emailed spreadsheets.
Validating AI outputs. Mistakes in automated mapping spread silently across every consolidated report.
Data pipelines and integration. Understanding how data moves between tools makes workflows easier to fix when they break.
You don't need to be a programmer. Power Automate, n8n and Airtable are built for people who design workflows without code. What matters is that every step has a check and an owner.
End of Mission 05
MISSION06
Interactive skills gap assessment
Know your gaps · Badge: Self-Aware Analyst
1 challenge
Mission goal: measure your readiness in four areas and find your top three gaps.
Step 1 of 3How it works
10 quick statements, about 3 minutes. Rate yourself honestly: this is a learning tool, not a test, and your answers stay in your browser.
Beginner: little or no experience
Developing: I understand it but need practice
Competent: I can apply it independently
Step 2 of 3Questions 1 to 5
Step 3 of 3Questions 6 to 10
Prefer to score it yourself?
Beginner = 0, Developing = 1, Competent = 2. Add them up (maximum 20): 0–7 Getting started · 8–14 Developing · 15–20 Ready to run AI-assisted coordination. For every Beginner answer, start with the matching action:
Build the reminder workflow in mission 4, guide 1.
Use the routine in mission 4, guide 3.
Draw the data flow for one reporting round.
Use the routine in mission 4, guide 2.
Add units and definitions to your code list.
Trace three figures in your last summary.
Use the routine in mission 4, guide 5.
Rank this month's delays and write one line on why.
Mission goal: commit to a realistic 12-week plan and track it week by week.
Step 1 of 5Weeks 1–2: Understand AI
About 3 hours a week (a planning suggestion). Tick steps as you do them: +5 XP each, saved in this browser. The mission completes when every step in all four stages is ticked.
Stage 1 · Weeks 1–20 / 5 done
Understand AI
Map your reporting round and learn what AI gets wrong.
Loot: a map of your reporting round, an error log and a time baseline.
Step 2 of 5Weeks 3–4: Apply AI
Stage 2 · Weeks 3–40 / 4 done
Apply AI
Try one workflow and two AI steps, with checks.
Loot: a working reminder flow, a reviewed mapping and a checked extraction.
Step 3 of 5Weeks 5–8: Integrate AI
Stage 3 · Weeks 5–80 / 5 done
Integrate AI
Make the workflows part of every reporting round.
Loot: an automated reporting round, a shared tracker and an automation log.
Step 4 of 5Weeks 9–12: Demonstrate competence
Stage 4 · Weeks 9–120 / 5 done
Demonstrate competence
Show the workflows are reliable and others can run them.
Loot: a portfolio project, a month of evidence and a development plan.
Step 5 of 5Check
Mission check+20 XP
In which stage do you first build the reminder workflow for a few partners?
Show answer
Stage 2 (weeks 3–4).
End of Mission 07
MISSION08
Your practical learning project
Boss level · Badge: Consolidation Lead
1 challenge + 8 deliverables
Mission goal: produce a portfolio piece that proves you can automate multi-partner reporting responsibly.
Step 1 of 6Your boss-level project
Boss level project
Automate one multi-partner reporting round, with a check at every step
Objective: show you can design AI-assisted workflows that collect, map, extract and consolidate partner data, with a person checking every step and a log others can follow.
Case study · Community Resilience Consortium Illustrative
No partners of your own to use? Build the round for this consortium, using the examples from mission 4.
Context: one monthly or quarterly reporting round with at least three partners. Use aggregated figures only.
Resources: your reporting calendar, partner templates and returns, one PDF report, your standard code list, Power Automate or n8n, Airtable or Excel, and an approved AI assistant.
Step 2 of 6Deliverables, one by one
Deliverables checklist0 / 8 done
8 steps to a portfolio piece
The mission and the certificate unlock only when all 8 deliverables are ticked.
Step 3 of 6Templates to start from
Four ready-made files for your deliverables. Download, adapt and keep them.
Before you share it: what should a person check in your workflows?
Think about it for a moment, then reveal.
Every mapping in the list, by definition and unit.
Every extracted figure, against its page.
The reminders, on the first full cycle.
Any AI-generated text, reviewed and marked as AI-assisted.
No personal data in any AI step or shared file.
Step 5 of 6How it's judged
Criterion
What good looks like
Reliable automation
Reminders and submissions work for a full round without manual fixes.
Checked AI steps
Every new mapping and every extracted figure was reviewed by a person.
Traceable data
Every consolidated figure links back to a partner return and page.
Honest summaries
No explanation that a partner did not give.
Maintainable
A colleague can run the round from the log.
Transparent with partners
Partners know what is automated and what data each tool receives.
Step 6 of 6Check
Mission check+20 XP
Which deliverable lets a colleague keep the workflows running while you are away?
Show answer
The automation log.
End of Mission 08
MISSION09
Responsible AI: what you must never delegate blindly
Guardrails · Badge: Responsible AI Guardian
10 challenges
Mission goal: know exactly which judgements stay with you, and the control for each AI risk.
Step 1 of 6Would you approve this report?
Click every sentence you would not send. There are five problems: an invented figure, an unsupported causal claim, an incorrect percentage, a missing indicator and a misleading interpretation.
Case study · Community Resilience Consortium Illustrative
Copilot drafted the quarterly progress summary from the consolidated tracker. The M&E manager needs it today.
Source (consolidated tracker, Q2)
Result
Partner returns received
9 of 11 (Partners 4 and 9 missing)
Project A: water points rehabilitated
38 of 40 (95%)
Project B: households trained
620 of 900 (69%)
Project C: home visits
1,450 of 1,600 (91%)
Project D: pupils reached
2,100 of 2,400; Partner 9 missing
Project E: screening campaigns
0 of 3 planned this quarter
Household savings
No indicator
Problems found: 0 of 5+40 XP
Show answer
Five problems: the 79% (it is 69%), the invented savings figure, the unsupported claim about health worker motivation, the misleading reading of Project D (a missing return), and the claim that all five projects are covered (Project E is missing).
Step 2 of 6Your decision
Your decision+20 XP
You found the problems. Can you send this summary to the M&E manager?
Show answer
Correct it first. Fix each problem, recheck every figure against the tracker, then send it.
Step 3 of 6Five questions before you send it
The summary is corrected. Before you send it, answer the five questions. Tick each one you can honestly answer yes to.
Step 4 of 6Mini-game: delegate or never?
AI and automation can remind, map, extract and draft. They cannot be accountable. For each task: can AI help (with checks), or must you never delegate it?
Delegate?+10 XP
Sending deadline reminders to partners who have not submitted
Show answer
AI can help. Automation does it well, with you checking the Monday list.
Delegate?+10 XP
Deciding whether a mismatch is a template issue or a real problem
Show answer
Never delegate. Listed in the profile as human judgement.
Delegate?+10 XP
Suggesting standard codes for new indicator labels
Show answer
AI can help. With every new mapping reviewed before data enters the tracker.
Delegate?+10 XP
Approving a mapping where the definitions differ in meaning
Show answer
Never delegate. Listed in the profile as human judgement.
Delegate?+10 XP
Extracting figures from PDF reports, with page references
Show answer
AI can help. With each figure checked against its page.
Delegate?+10 XP
Prioritising follow-up when several projects are behind
Show answer
Never delegate. Listed in the profile as human judgement.
Delegate?+10 XP
Drafting the cross-project summary from the tracker
Show answer
AI can help. Figures only, with context added from partner calls.
Delegate?+10 XP
Raising a sensitive implementation problem with a partner
Show answer
Never delegate. Relationships with partners stay with you.
Step 5 of 6Six risks, six controls
What can go wrong when workflows and AI handle partner data? Name three risks before you look.
Think about it for a moment, then reveal.
Risk
What it looks like
Your control
Wrong mappings
Women counted as households across every report
Review every new mapping by definition and unit
Added-up totals
The AI combines figures the report keeps apart
Check each extracted figure against its page
Silent failures
A workflow stops and nobody notices
The Monday list and an automation log with owners
Invented explanations
"Because of the rains" with no source
Context only from partner calls, with the date
Personal data
Beneficiary lists sent to an AI step
AI steps get labels and aggregates only
Damaged relationships
Automated warnings to partners with agreed extensions
Extensions in the calendar; you call before escalating
Step 6 of 6Tell readers how AI was used
Transparency is a control too. Add a short note to every consolidated summary, and adapt it to what was actually done.
Partner returns were collected and consolidated with automated workflows (Power Automate and n8n). An AI step suggested indicator mappings, and an AI assistant extracted figures from PDF reports and drafted this summary. Every new mapping and every extracted figure was checked by [name], M&E Coordinator, and explanations come from partner calls. No personal data was processed by AI tools.
The AI-Enabled M&E Manager is covered in Tutorial 01, the AI-Enabled M&E Specialist in Tutorial 02, the AI-Assisted M&E Officer in Tutorial 04 and the AI-Enabled MEAL Coordinator in Tutorial 16.
Step 2 of 5What you need to show
Strengths you build on
Keeping many partners reporting on time
Harmonising templates and indicator codes
Following up on delays with partners
Consolidated progress reporting
What you need to show
An automated reporting round that works
AI mappings reviewed by definition and unit
Extracted figures traced to their pages
An automation log others can follow
Summaries without invented explanations
Step 3 of 5Tool: build your LinkedIn headline
Pick the pieces that are true for you. Only claim what you have actually done.
Your headline
Step 4 of 5CV bullets and qualifications
CV bullet templates
Automated reminders and submission tracking for [11] partners with Power Automate and SharePoint, cutting missing returns from [X] to [Y] per round.
Set up an AI-assisted mapping of partner indicator labels to standard codes, with every new mapping reviewed before data entered Airtable.
Extracted figures from [N] partner PDF reports with Claude, checking each figure against its page before consolidation.
Wrote the automation log that let [colleague role] run the monthly reporting round without me.
Replace the brackets with what you have actually done.
Qualifications in the profile
Bachelor's or master's degree in development studies, social sciences, management or a related field.
3–6 years in programme monitoring or coordination, including work with multiple partners or field offices.
Useful certifications: MEAL DPro (PM4NGOs); PMD Pro (PM4NGOs).
Step 5 of 5Check
Mission check+20 XP
What's the best way to use "AI-Enabled M&E Coordinator" on your CV?
Show answer
Use it to describe your capabilities, backed by evidence, unless your employer actually uses the title.
End of Mission 10
MISSION11
Your next steps
Commit and celebrate · Badge: Transition Ready
1 challenge
Mission goal: turn what you learned into a two-week plan and unlock your certificate.
Step 1 of 2Commit to your next two weeks
Pick the actions you commit to for the next two weeks. Your plan builds itself below.
Start learning
Do this
Show this
My plan
Recommended first project: the automated reporting round from mission 8, starting with reminders for three partners.
Step 2 of 2Your completion certificate
Complete all 11 missions to unlock it. Add your name, take a screenshot and share your progress.
AI & M&E Career Transition Series · Tutorial 05
Your name
has completed the learning path
From M&E Coordinator to AI-Enabled M&E Coordinator
This is a learning-progress certificate. It records what you completed in this free tool, in your own browser; it is not independently verified and not a formal qualification. Your mission 8 project is the evidence to show an employer.
End of Mission 11
Questions and answers
What does an AI-Enabled M&E Coordinator do?
Coordinates monitoring across several projects, field offices or implementing partners so that data arrives on schedule, in consistent formats, and is consolidated into one picture of implementation progress. Uses AI-assisted workflows to consolidate data and automate reporting schedules, while the content of indicators and evaluations stays with specialists and managers.
How is AI changing the M&E Coordinator role?
Chasing, copying and reformatting data from many sources, which used to fill much of the coordinator's week, can now be largely automated. The coordinator's role shifts to designing those workflows, checking that consolidated data is correct and using the time saved to follow up with partners on real implementation issues.
Which judgements should an AI-Enabled M&E Coordinator never hand over to AI?
The career profile lists these as human judgement:
Building working relationships with partners so data problems are raised early.
Deciding whether a mismatch is a template issue or a real implementation problem.
Prioritising follow-up when several projects are behind at once.
Checking AI mappings of indicators where definitions differ in meaning.
Which skills should I learn first?
The learning priorities in the career profile, in order:
Workflow automation. Automating reminders and consolidation removes the most repetitive part of the job.
AI document extraction. Many partners still report in PDFs and Word files that need converting to data.
Airtable. A shared tracker with forms and views suits multi-partner coordination better than emailed spreadsheets.
Validating AI outputs. Mistakes in automated mapping spread silently across every consolidated report.
Data pipelines and integration. Understanding how data moves between tools makes workflows easier to fix when they break.
The profile lists 12 skills in all: M&E planning, Project management, Stakeholder engagement, Workflow automation, Data cleaning and wrangling, Data pipelines and integration, AI document extraction, Validating AI outputs, Indicator design, Capacity strengthening, Data quality assessment and Analytical report writing.
Which tools does this role use?
The career profile names: Excel, Airtable, SharePoint, Power Automate, n8n, LLM APIs (OpenAI, Anthropic, Google), Claude, Microsoft 365 Copilot, Power BI and KoboToolbox. The tutorial shows how to check every AI output, whichever tool you use.
What qualifications and experience does the profile list?
Bachelor's or master's degree in development studies, social sciences, management or a related field.
3–6 years in programme monitoring or coordination, including work with multiple partners or field offices.
Useful certification: MEAL DPro (PM4NGOs) and PMD Pro (PM4NGOs).
How long does the transition take?
This tutorial suggests a 12-week plan of about 3 hours a week, in four stages: understand AI (weeks 1–2), apply AI (weeks 3–4), integrate AI (weeks 5–8) and demonstrate competence (weeks 9–12). It is a planning suggestion, not a requirement.
Is "AI-Enabled M&E Coordinator" an official job title?
No. It is a proposed career title from the EvalCommunity framework, not a standardised occupational classification or a guaranteed vacancy. Use it to describe your capabilities, backed by evidence, unless your employer actually uses the title.
