So you want a new M&E software

This post is for you — if you are juggling a dozen different reports and spreadsheets for every project.This post is for you — if you bought a monitoring and evaluation database, only to find it disappointing.This post is for you — if you feel like your workday would be half as long if you didn’t do boring, repetitive tasks.

This post is for everyone who wants digital tools to make monitoring and evaluation better, faster, and more insightful.

This post is here to tell you not only that that’s possible but how you can get there.

Here are four points to help you along the way:

1) Start with the data

Specifically, start with the concrete use cases and purposes of your data. It is tempting to jump from your collection of templates and processes straight to a software solution. But skipping this step means you just digitise what you have. You digitise the same data, the same processes, the same inefficiencies, and the same frustrations.

Instead, look at what your organisation wants to use data for, in an ideal world. Define the exact occasions on which certain people need certain information. Define in detail what you need to design a new project, to convince a donor, to write a report. Then map all these data points and decide where you will find that data in the real world. Only then are you ready to begin looking for software tools.

2) Search widely, choose wisely

This is fun — a world of tech possibilities! It’s exciting to look at fancy tools, and imagine all your M&E worries being automated away. However, don’t just believe a consultant (no, not me either!) that a specific solution is the right one for you. Make sure you ask the right questions and keep an eye on your organisation’s priorities in this phase. Below is the current spectrum of available options.

Patch-working office software

This is the cheapest option by far, and the most commonly used one. It means using a combination of SAAS products to fulfil all your requirements. For example, you have KoboToolbox or Surveymonkey for data collection, Google drive/sheets for data storage, and then R, Power BI and Tableau for visualisation. Perhaps you automate sending data between those. These kinds of solutions will be appropriate for very small organisations. However, you risk having data that is not accurate, data that is not analysable, and data that gets lost.

Data software off the shelf

If your organisation has the budget for a comprehensive solution such as Toladata or Salesforce, you can buy one tool for all your M&E needs. Most of these tools can be configured to fit your organisation. However, the ongoing costs are high, depending on your number of users. Toladata costs more than 3k USD per year even for organisations under 5 users. Salesforce additionally requires you to use a consultant for configuration, so you won’t get that implemented for under 40k. Besides, there will likely be limitations on features. So some of your processes may have to stay offline.

Custom-built

If your organisation has both budget and risk appetite, you can contract a software developer or an agency. They can build you a tool from scratch, with every feature you can think of. They can make it look exactly like you dreamed.

However, software developers are incredibly expensive. That goes both for building the initial version and for making changes afterwards. And you will need lots of changes afterwards, see section 3. If you hire an agency, do not budget less than 100k USD for the first year, and 50k for the second. You could alternatively hire a software engineer as staff, that will be less expensive. Just bear in mind that in both cases, you will need them to maintain your tool in the long run.

Custom-built, but no-code

If your organisation has a limited budget but is keen to innovate, no-code is your paradise. No-code means building applications for web/phone, but not using traditional coding languages (such as javascript, python, CSS). Instead, no-code is a visual coding language. This makes building apps much faster and much cheaper than a software development agency. Examples include Knack, Bubble, or Airtable. Ongoing costs can be kept to under 2k USD per year, independently of how many users you have.

(This is my speciality — reach out if you are keen to explore this further!)

3) Commit and make it work

None of the solutions above will be implemented once and immediately solve all your M&E issues. It is impossible to know from the start what the user needs are and what they will be in six months. It is entirely normal for new user needs and issues to become visible only after you have launched a software tool.

The important part is not to give up at this point. Far too often, new software tools go from an exciting launch to the dustbin of abandoned processes within months. Instead, expect the first version to not be perfect, likely not even good. Then you can budget time and money to make the necessary changes. And then you keep making changes. The good thing is that most solutions can be used in their imperfect version to a degree. So you can still collect and manage data while working on improvements.

… and finally

I hope these steps help you find your way towards an M&E software that works for you. Start with the data, search widely and wisely, and commit to making it work. And also — don’t get discouraged. Just like in other aspects of M&E, failures are good, because they help you learn.

I’ve had some spectacular fails, that’s why I’m confident in giving this advice. :)