Design a Data Flow Diagram
Design a data flow diagram for a MEL plan.
Draft: Design a Data Flow Diagram
Fill In Your Details (3 Blanks)
Add your information to get a tailored prompt. You can also use it as is.
Sector and Donor Context (Optional)
Your Prompt (With Your Details)
This is the complete prompt, ready to copy and paste.
You are a senior MEAL specialist designing the data flow diagram and accompanying narrative for your program's MEL plan.
The diagram must make explicit who collects what, where it lives, how it moves, and who acts on it.
Requirements:
- Entities and roles. Identify the actors in the data flow (e.g., field staff, partner M&E officers, district focal points, MEL manager, donor focal point).
- Capture stages. Describe each capture stage: point of contact (e.g., service delivery, household survey), tool (paper form, mobile app, registry), and responsible role.
- Aggregation stages. Describe how raw records are aggregated up the chain (site, partner, region, national) and where the system of record sits for each indicator.
- Validation and quality. Show where validation steps occur (auto-checks at entry, supervisor review, monthly DQA, quarterly back-check).
- Storage and access. Identify the platform for each stage (e.g., paper file, Excel, KoBo, DHIS2, partner system) and access permissions.
- Use points. Indicate where data is reviewed and used (program meetings, quarterly reviews, donor reports), tying use back to specific roles.
Output Format:
Produce:
- A 1-paragraph framing.
- A structured description of the diagram, organized as nodes (capture, aggregation, storage, use) and arrows (data movement), so the diagram can be rendered visually.
- A roles and tools table (role, tool, action, frequency).
- A short narrative explaining how the diagram ensures end-to-end data integrity.
- A list of key risks (data loss, version conflict, delayed transmission) and mitigations.
Check: Score It with the MEL Data Flow Diagram Quality Rubric
The Review Prompt
Paste the draft at the bottom of this prompt and run it. It scores each criterion from 1 to 5 and says why.
You are an expert M&E advisor specializing in data management systems. Score the MEL data flow diagram I will provide using the rubric below.
SCORING RUBRIC - MEL Data Flow Diagram Quality Score each dimension 1-5 using these criteria:
DIMENSION 1: Completeness of Flow
- Score 5: Every stage from raw data source to decision-maker is shown. The flow covers collection, entry, storage, cleaning, analysis, reporting, and use. No step is implied; every transition between stages is drawn or described. Both routine and one-off data products (e.g., baseline, endline, ad hoc analyses) are accounted for.
- Score 4: All major stages from source to use are shown. No more than one transition is implied rather than drawn. Routine data products covered; one-off products may be referenced without full detail.
- Score 3: Source-to-use path recognizable but at least two transitions are implied or missing. A reader can follow the general direction but must infer where data goes between named steps. Only routine flows shown.
- Score 2: Half or more transitions are missing or jumpy. The flow stops at storage or reporting rather than reaching the decision-maker, OR collection is shown but analysis and use are not.
- Score 1: No end-to-end flow. Diagram shows a list of tools or steps without connections, or covers only one segment (e.g., collection only).
DIMENSION 2: Role Assignment Clarity
- Score 5: A named position is assigned at every transition point. Every box and every arrow has a clear owner. Handovers between positions are explicit (who passes to whom). Supervision or approval roles named where relevant (e.g., who signs off on cleaned data).
- Score 4: Named positions assigned at no fewer than 80 percent of transition points. The remainder show a unit (e.g., "M&E team") rather than a specific position. Major handovers explicit.
- Score 3: Named positions assigned at half or more transition points. The remainder show unit-level ownership or no owner at all. Handovers between positions implied rather than stated.
- Score 2: Named positions appear at fewer than half of transition points, OR all ownership is assigned at unit level ("the team," "M&E staff") rather than position level.
- Score 1: No roles assigned. The diagram shows tools and arrows but no one owns any step.
DIMENSION 3: System Specification
- Score 5: Every storage and transfer point names a specific tool or platform (e.g., "KoboToolbox Form A1," "PowerBI Dashboard v2," "shared OneDrive folder /MEL/2026/"). No generic placeholders ("project records," "the database"). File formats and transfer methods named where relevant.
- Score 4: Specific tools named at no fewer than 80 percent of storage and transfer points. The remainder use slightly generic labels (e.g., "field tablet") but the tool category is clear.
- Score 3: Specific tools named at half or more storage and transfer points. The remainder use generic labels ("project database," "spreadsheet") with no further specification.
- Score 2: Tools labeled generically for half or more points ("project records," "M&E system"), OR multiple distinct systems are collapsed into a single undifferentiated box.
- Score 1: No specific tools named. Diagram shows abstract stages only.
DIMENSION 4: Quality Gates and Verification
- Score 5: Validation, cleaning, and review steps are drawn as explicit nodes in the flow with named owners and stated triggers. At least one verification step (spot-check, supervisor review, or DQA) is shown before data is used for analysis or reporting. Error-handling path shown for rejected or flagged data.
- Score 4: Validation and cleaning steps are explicit with named owners. Verification step shown before use, though one element (frequency, sample size, or error-handling path) is unspecified.
- Score 3: At least one quality gate is shown (typically cleaning or supervisor review), but it is referenced as a step rather than detailed with owner and trigger. No error-handling path.
- Score 2: Quality gates implied through general statements ("data will be cleaned") rather than drawn into the flow. No named owner or trigger for any check.
- Score 1: No quality gates shown. Data moves from collection to reporting with no verification step.
DIMENSION 5: Feasibility and Context Fit
- Score 5: The flow is realistic given the staffing, geography, connectivity, and reporting cycles described elsewhere in the MEL plan. Field-level steps account for connectivity constraints (e.g., offline data capture with later sync). Frequency of each transfer aligns with the reporting calendar. Workload at each named position is plausible.
- Score 4: Flow is realistic overall. One feasibility concern is unaddressed (e.g., a daily sync requirement in a low-connectivity area), but the rest of the flow accounts for context.
- Score 3: Flow is conceptually sound but feasibility is partially addressed. Two or more steps assume conditions (connectivity, staff time, device access) that are not justified by the broader plan.
- Score 2: Flow is unrealistic at multiple points (e.g., real-time dashboards in a paper-based field operation, or daily entry workload that exceeds named staff capacity).
- Score 1: Flow is detached from operational reality. The diagram describes an idealized system with no relationship to documented staffing or context.
OUTPUT FORMAT: Return your assessment as a table followed by a summary:
| Dimension | Score (1-5) | Evidence | Priority Revision |
|---|---|---|---|
| Completeness of Flow | |||
| Role Assignment Clarity | |||
| System Specification | |||
| Quality Gates and Verification | |||
| Feasibility and Context Fit |
Total: X/25 Band: Strong (22-25) / Adequate (17-21) / Needs Revision (11-16) / Substantial Revision (5-10) Single Most Important Revision: [One specific sentence]
For any dimension scored 1 or 2, add a brief explanation and a concrete revision example.
DATA FLOW DIAGRAM TO SCORE: [Paste your Data flow diagram or section from a MEL plan (narrative + visual, or text description) here]
Fix: Revise What Scored Low
The Revise Prompt
Paste the draft and its scores at the bottom of this prompt. It revises the parts that scored below 4.
You are an expert M&E advisor specializing in data management systems. Review the draft data flow diagram I will provide and fix every problem that falls below the Adequate (3) threshold on the MEL Data Flow Diagram Quality rubric.
WHAT TO LOOK FOR AND FIX:
Completeness of Flow problems: Flows that stop at storage or reporting without reaching the decision-maker. Missing transitions between collection, cleaning, and analysis. One-off data products (baseline, evaluation) not represented.
Role Assignment Clarity problems: Boxes or arrows with no named owner. Ownership assigned at unit level ("the M&E team") rather than position level. Handovers between positions left implicit.
System Specification problems: Generic tool labels ("the database," "project records," "M&E system"). Multiple distinct systems collapsed into one box. No file format or transfer method specified.
Quality Gate problems: No verification step before data is used. Cleaning referenced in general terms but not drawn into the flow. No error-handling path for rejected data.
Feasibility problems: Real-time or daily-frequency steps in a low-connectivity setting. Workload at a named position that exceeds plausible staff time. Mismatch between flow frequency and the reporting calendar.
OUTPUT FORMAT: For each problem found:
- Dimension: [which dimension]
- Location: [which step or transition in the flow]
- Problem: [one sentence describing what is wrong]
- Revised version: [the corrected text or step]
Then provide the complete revised data flow diagram (narrative + role and tool table) incorporating all corrections.
DRAFT DATA FLOW DIAGRAM TO REVISE: [Paste your Data flow diagram or section from a MEL plan (narrative + visual, or text description) here]