Evidence Inventory Guide

An evidence inventory is a structured record of every piece of supporting documentation for your project's reported results. Maintaining a complete, organised evidence inventory throughout the project — not just at reporting time — is one of the most effective practices in credible reporting.

Why an Evidence Inventory Matters

Without a structured evidence inventory:

• Report writing becomes frantic at deadline time
• Evidence is duplicated or lost
• Claims cannot be verified by auditors
• Programme officers who request evidence have to wait
• Data quality suffers because evidence is collected inconsistently

With a well-maintained inventory:

• Reports are written in hours, not days
• Every claim is backed by a traceable evidence record
• Audit responses are prepared quickly and confidently
• Monitoring visits go smoothly because data is organised
• Evidence gaps are identified and filled proactively

The Evidence Inventory Structure

Each evidence record should capture the following fields:

Required Fields

Field	Description	Example
Evidence ID	Unique identifier for this record	EV-2026-Q2-001
Date Collected	When the data was gathered	15 June 2026
Date Uploaded	When the record was entered into DonorDesk	16 June 2026
Type	Category of evidence	Household Survey
Methodology	How the data was collected	Structured questionnaire, n=312, random sampling
Data Source	Who or what provided the data	Beneficiaries in target villages
Location	Where data was collected	Migori County, Villages A–D
Indicators Supported	Which logframe indicators this evidence supports	IND-1.2a, IND-2.1
Collected By	Name and role of person responsible	J. Okoth, M&E Officer
Reviewed By	Name and role of quality reviewer	S. Amina, M&E Manager
File Reference	Drive link or DonorDesk evidence title	DD-EV-2026-Q2-001
Limitations	Known issues with data quality	Response rate 78%; recall bias possible

Optional Fields (Recommended)

Field	Description	Example
Consent Status	Whether participant consent was obtained	Written consent obtained
Disaggregation Available	What disaggregation levels exist	By sex, age, disability
Sensitivity	Whether data contains sensitive information	Low / Medium / High
Related Evidence	Linked evidence records	EV-2026-Q1-003 (prior period)

Evidence Types and Their Use

Household Surveys

Best for: Outcome-level indicators, beneficiary satisfaction, awareness, behaviour change

Required documentation:

• Questionnaire / survey instrument
• Sampling methodology and sample size justification
• Enumerator training notes
• Raw data (or analysis dataset)
• Consent procedure description
• Analysis plan / report

Minimum standards:

• Sample size stated with margin of error
• Response rate reported
• Limitations acknowledged

Routine Programme Records

Best for: Output-level indicators — attendance, distribution, service delivery counts

Required documentation:

• Activity log / register (countersigned)
• Distribution list with beneficiary signatures (or alternative where literacy is a constraint)
• System-generated reports (e.g., DHIS2 export)
• Aggregation methodology explained

Minimum standards:

• Must be contemporaneous (not reconstructed after the fact)
• Must be signed or system-generated
• Must show date, location, and beneficiary identifiers

Health Facility Data (DHIS2)

Best for: Health programme indicators, service utilisation

Required documentation:

• DHIS2 data export (signed / stamped by health authority where possible)
• Indicator definition used
• Reporting period covered
• Any data quality flags noted

Monitoring Visit Reports

Best for: Verifying that activities happened, spot-checking beneficiary reach

Required documentation:

• Visit report with date, location, observer name
• Methodology used (interview, observation, focus group)
• Sample size (number of beneficiaries interviewed)
• Key findings
• Photos with captions (if relevant)

Official Statistics

Best for: Context indicators, baseline and endline comparisons

Required documentation:

• Source citation (name of dataset, organisation, year)
• Link to source (URL or publication reference)
• Methodology note for the official statistic
• How it was used in the project context

Partner Reports

Best for: Joint output indicators, sub-recipient activities

Required documentation:

• Original partner report (submitted to you)
• Your review notes
• How you aggregated or validated the data

Case Studies and Testimonials

Best for: Qualitative outcomes, human impact stories

Required documentation:

• Written consent from the individual (or parent/guardian for children)
• Date and location of interview
• Context of the story
• Confirmation that the story is representative (not cherry-picked)
• How the story relates to programme outputs

Photos and Video

Best for: Visibility, participation documentation, infrastructure

Required documentation:

• Date, time, location
• Subject description
• Photographer's name
• Consent form for any identifiable individuals
• Caption (context, names optional per consent)
• File metadata intact (do not strip EXIF data before storage)

Organising Evidence in DonorDesk

DonorDesk's Evidence library fits the inventory structure. For each piece of evidence:

1. Upload the file (or link it from your Google Drive) as soon as it exists, not at reporting time
2. Set the title, evidence type, reporting period and confidentiality
3. Accept or edit the AI tag suggestions
4. Link the evidence to the indicators and activities it supports
5. Verify it (a reviewer confirms it)
6. Review and update records after each monitoring visit or data collection exercise

Evidence File Naming Convention

Use a consistent naming convention for all uploaded files:

EV-[Year]-[Period]-[SequenceNumber]_[EvidenceType]_[ShortDescription]

Examples:

• EV-2026-Q2-001_Survey_HH_Nutrition_Migori.pdf
• EV-2026-Q2-002_DistributionLog_WASH_Kits_Kisumu.pdf
• EV-2026-Q2-003_Photo_WaterPoint_Rehabilitation_Consent.pdf
• EV-2026-Q2-004_DHIS2_Health_Facility_Data_Export.xlsx

Evidence Quality Assessment

For each evidence record, assess quality before using it to support a reported result:

The CRAFT Test

C — Credible: Is the source trustworthy? Is the methodology sound? R — Relevant: Does it directly support the indicator claim? A — Adequate: Is the sample size sufficient? Is the coverage representative? F — Fresh: Is it current enough to support the reporting period? T — Traceable: Can you trace it back to its source?

If any element fails, note the limitation in the evidence record and the narrative.

Evidence Gap Management

At each reporting period, before writing the narrative:

1. Run an evidence gap report — List every indicator in the narrative and confirm evidence exists
2. Identify gaps — Note which indicators lack supporting evidence
3. Prioritise gaps — Focus evidence collection on the most significant gaps
4. Plan collection — Schedule data collection activities to fill gaps before the next reporting period

Evidence for Different Donors

Different donors have different evidence standards:

Donor	Specific Evidence Requirements
UNHCR	Protection monitoring data must be from approved UNHCR methodologies
DG ECHO	Outputs must be verified through field monitoring or partner reports
USAID	Data quality assessments (DQA) required for all custom indicators
Global Fund	LFA will verify SR data; documentation must be available for audit
GCF	Emission reductions require tiered methodology documentation; ESS documentation mandatory
FCDO	VfM requires cost-effectiveness data; evidence must support value claims
EU	Audit trail must be complete; procurement documentation required above thresholds
Gates Foundation	Evidence should support scalability claims; methodology must be replicable

Evidence Retention

Retain all evidence records for the duration specified in your grant agreement, plus an additional 5 years (or the statute of limitations in your operating country, whichever is longer).

Evidence should be stored:

• In a secure, backed-up digital system (DonorDesk, approved cloud storage)
• With access controls (who can view vs. who can edit)
• In a format that will remain accessible (avoid proprietary formats that become obsolete)

---

Next: Consult the Report Writing Glossary for definitions of key terms used throughout this course.