A call for investing into information management in aidwork

I think of information management a bit like of logistics: You don’t notice it, if it works smoothly; but it has a massive impact if it doesn’t work properly. But, unlike with logistics, many people are so used to information management being crap, that they think that it cannot be improved. Let me tell you: good information management is possible, it’s not even difficult, and it can do lots to improve humanitarian aid.

The difference between data and information

Files in drawer

Data or information?

In my experience one of the biggest problems is that people don’t understand the difference between collecting data and analysing information. When you talk to a beneficiary, fill out a form and then archive that form somewhere, then you have collected data. However, that data only turns into useful information once you have entered it into a database or spreadsheet so that you can filter, sort, aggregate or calculate data.

If you work for a big retailer, you might think that this is obvious. However, in the humanitarian aid industry it sadly is not.

Three examples:

  1. An organization working with refugees announced during a meeting that they would relocate between 2,000 and 4,000 people from host communities to a refugee camp during the next few weeks. Another organization that is in charge of providing food to that camp, said that they’d need to have a more precise number so that they know how many sacks of rice to buy. However, in the rush of the crisis the first organization had only registered heads of household and was not sure how many people were part of each household.
  2. Different meeting, different actors, similar issue: Somebody announced that 458 people would be relocated to a refugee camp in the coming week. Someone working for a an education agency asked, how many of these would be children for whom schooling would have to be provided. The first organization was not able to provide that information since, while they had captured the age during registration on paper, they had only entered “minor” or “adult” into their database.
  3. A while age I got to work with a big Excel spreadsheet that contained data of people who had been registered. Many of these people had two or three names (Last name, Middle name, First name), all of which were entered into a single cell called “Name” in the spreadsheet. However, people would sometimes give their last name first, and another time say their first name first, making it difficult to find them in the list. This problem was made worse by the fact that many of the people were illiterate so spellings might differ slightly from one time to the next. Once we put each name into a separate field, it was much easier to find even the ones that were misspelled because at least one of the three names would normally be consistently spelled and then you could simply scan the list of remaining names for phonetic matches.

Granularity and structure of information

I realize that collecting data during an emergency is a very challenging task. However, the three issues I described are not about an inability to collect data, they are about a lack of knowledge how to treat data properly so that you can use it to inform your decisions. In other words: they are examples of bad information management.

One frequent answer to this is “True, but we didn’t think about that in the beginning.” Well, that is exactly why you need to have professional, experienced information managers!

Particularly during the first phase of an emergency, organizations should invest in people who can help them structure the information they need in such a way that it is collected and entered in a useful way. This is something that can even be done by consultants on short-term contracts. Then, once the structure is in place, other (less expensive) people can enter data into those spreadsheets or databases and programme people can use the information as a tool.

Data confidence and errors

The second big challenge is to ensure that the data is entered correctly and stays correct. Everyone makes mistakes, particularly when you have a mind-boggling boring job like entering a list of thousands of names or dates of birth into a spreadsheet. One of the most efficient ways to prevent data entry errors is the double entry method. This means that everything is entered twice, by different people. Once all the data is entered, you can automatically compare the two data sets and look at the differences. I know, I know … I can almost hear cash-strapped managers groan at this suggestion, but hiring two data entry clerks might be only marginally more expensive then hiring one more qualified person.

The second way to reduce errors and increase data confidence is to have programme people work with the data themselves. I’m not necessarily talking about the initial massive data entry, but of the subsequent steps. In a lot of situations, updates are simply given to a person, whose job it is to update the database. This person normally has no knowledge of the programmes or the cases and most of the time never leaves the office. So for him, the data has no context and he will happily enter things that are written down, even if they don’t make sense. On the other hand, a programme person will notice inconsistencies or even be able to suggest improvements to the database. However, that means that organizations have to invest into teaching their programme people the basics of information management so that they understand how to do things and why certain things are done a certain way.

Stop reinventing the wheel

Last but not least, I’m still amazed to see people reinvent the wheel over and over again. A lot of the data that has to be captured is the same in different emergencies, with only small adjustments being necessary to fit the specific context. However, in each emergency people seem to start from scratch and build their databases and spreadsheets as they go.

It should be the role of the global cluster leads to provide standardized information management tools that can be reused and adapted. And while some clusters are already doing this, it is still too few and in the cases where the clusters do provide these tools, a lot of organizations are not aware of them – which again calls for an experienced information manager who knows where to find these tools and how to use them.

I know that hiring people with laptops and spreadsheets is not as sexy as buying trucks or as photogenic as handing out food to beneficiaries. But I am convinced that if we are serious about being efficient and about using the money that we have been given to the greatest advantage of the beneficiaries, we need to professionalize and invest into information management. The people who can do that job already exist, you just need to have to the will to find the budget and hire them. You programmes will improve because of it.