Do you need MEAL software?
A spreadsheet is enough for a single small project with one person entering data. Software starts to pay off when you have several projects, several donors, many data collectors, or a need to see results across the organization. At that point, scattered files cause errors, delay and arguments about which number is right.
The aim of software is not to hold more data. It is to make the data trustworthy and easy to use.
What should you check first?
Write down your answers to these questions before you look at any product:
- How many projects, donors and users do we have?
- Who collects data, where and how? Do they have a reliable connection?
- What do donors ask us to report, and how often?
- Who must approve data before it counts?
- Who may see what?
- What must we be able to export?
Your answers become your checklist.
What features matter?
| Area | Questions to ask |
|---|---|
| Results frameworks | Can you build a framework and link indicators to results? |
| Indicators | Is there a library? Can you define units, baselines, targets, periods and disaggregation? |
| Data collection | Can field staff enter data easily? Does it work offline, or only online? |
| Review | Can a reviewer approve or reject data before it updates results? |
| Reporting | Can you produce donor-style reports and exports? Can you choose periods? |
| Feedback | Does it hold complaints and feedback, with safe handling of sensitive cases? |
| Access | Can you set roles and permissions, and give donors limited access? |
| Languages | Which languages are supported, and how fully? |
| Exports | Can you take your data out in common formats? |
How do you check data quality features?
Weak data is the main risk in MEAL. Look for:
- Approval before counting. Data that is pending or rejected should not change results.
- Clear definitions. The system should show the definition next to each indicator.
- Checks on entry. Required fields, valid ranges and dates.
- History. You should be able to see who entered or changed a value, and when.
- Honest totals. Ask whether totals are people or records. Many systems count records, which can double count the same person. See reach totals and double counting.
What about security and data protection?
MEAL data often includes sensitive information about vulnerable people. Ask where data is stored, how it is protected in transit and at rest, how access is controlled and whether you can remove data. Ask for these answers in writing. Do not accept general claims. Check local law and donor rules on where data can be kept.
How do you test and compare?
- Run a pilot. Load one real project, with your own indicators and a sample of data.
- Run a real report. Produce a donor report from it. Note what you had to fix by hand.
- Ask for the limits. List what the product does not do, and decide whether you can live with them.
- Include field staff. If the people who collect data find it hard, quality will drop.
- Count the full cost. Include licenses, set-up, training, support and the time to clean existing data.
- Check the exit. Make sure you can export your data if you leave.
You can also read what MEAL is if your team needs a shared starting point.
What questions should you put to any supplier?
Put these in an email and keep the written answers:
- Which of our requirements do you meet today, and which do you not meet?
- Does field data entry work without a connection?
- Do you connect to other collection tools, or do we enter data by hand?
- How is data deleted if we ask?
- What happens to our data and our access if we stop paying?
- How many languages are fully translated, and which screens remain in English?
- How do you handle support, and in which time zone?
A supplier that answers plainly, including with “no”, is easier to trust than one that answers “yes” to everything. Compare the answers against your checklist, and weigh the gaps against what your team can work around.
Where EMPOBase fits
EMPOBase includes a MEAL module with results frameworks tied to Programs, a library of 235 indicators in eight sectors, Monitoring, 3W Coverage and Complaints & Feedback (CFM). Collection links stage field submissions for review, and only approved submissions update indicator actuals. Known limits: field reporting is online only, there is no connector to KoBo or ODK, and totals count records, not unique people. See the indicator library or ask for a walk-through.