00:00:04 Introduction and poll

So glad to see everyone joining us for our first webinar of the year. We're very excited for our webinar series for the year with, I believe, quite a relevant topic for many of us here working in the development space. Before we dive into the actual webinar, we just wanted to start off with a quick poll just to get a feel for where you guys are coming from as you're joining us today.

The first question is, do you currently have an information management system in place for the project you're currently working on? The options are: yes, you have a full system in place; or you only have a partial system in place; or you don't have a system in place at all. The second question is, if yes, you do have a system in place, how confident are you in the system that you're using? The responses range from completely confident all the way to not at all confident.

While you're at it, related to if you are using an information management system, I'm curious to hear if you're using any specific tools to manage your information management system, or is there any specific software you're using? Feel free to type that into the chat as well.

It looks like for the first question, the majority do have a system in place, either partial or full. That's great that you do have something already. Between full and partial systems, it looks like more of you only have a partial system in place. Today we'll take a look at what are the components of an actual system that we can consider as a full system, and we'll see how that squares with your own understanding. For those of you who do not have a system in place yet, I hope that you'll pick some things up from today's webinar that will give you some motivation to actually pursue putting a system in place for your own project.

In terms of confidence, this one is more distributed. Only 16% are completely confident, and I think that's fine; I don't think there exists a perfect system. Perhaps most are somewhere in the middle, either fairly or somewhat confident. There are a few of you who are slightly confident or not at all confident. In the course of today's webinar, we'll explore some of the things that help us to explain what to expect in an information management system.

I'm seeing some people sharing about what tools they're currently using. I can see a lot of the common data collection tools like Kobo and Google Forms. Lots of people using spreadsheet-based software like Excel or Smartsheet or Google. Those are all the common suspects for sure. Today we'll see how each of those software tools actually fit into what we understand as an information management system.

00:04:50 Why have an information management system?

Let's dive into our first section answering the question: why we should have a system in the first place. Before we proceed, I just want to make a quick note that in the course of this presentation, we will use the term information management system or the acronym IMS generally speaking, but you may in other contexts have come across other similar terms such as MIS for management information systems, or more simply information systems. These are all indeed related terms, but we'll go through the finer nuances and terminology in today's discussion so that we can become more aware about what each of these terms means.

Let's start with a discussion on the keyword in all of this, which is information. What exactly is information to begin with? To help us define this term, let's consider an example. Imagine that you're working on a nutrition project in Chennai, India that's involved with distributing nutritious meals among school children to reduce the prevalence of stunting.

Consider this statement: "10% of children under 5 are stunted." Is this what we would consider information? I would say not yet. This is just a single data point. Now let's compare this with this next statement: "10% of children under five are stunted at our midline study compared to 20% at baseline. The target for the end of our project is 0%."

Do you notice the difference? On the one hand, we're looking at a single data point, but on the other hand, we're starting to synthesize multiple data points and starting to generate some information. We're starting to get a sense that this project that we're implementing is actually moving the needle. We are now at about halfway through to our target at this point in time.

00:09:50 Data vs. information

Let's talk about data versus information. These two terms are indeed very closely related, and many of us might use these terms interchangeably. However, they're not actually the same. Just because you've collected some data doesn't mean that you have information. Data is raw and unstructured, whereas information has context, it is processed, and it helps us draw some conclusions and communicate about what is actually happening. So while data tells us what events are happening, information helps us to tell the story about our project.

In our example, this information that we have indicates that we're actually making some progress in our project. So now the next thing to think about is what do we do with this information? In our example, we can take this information and use it as the basis for continuing our project. We don't just generate information for the sake of having the information. Information should serve a higher purpose, and that is to serve as an input to decision making.

To transform data into information, there are a couple of things that we need to do. First, we need to contextualize it. The context helps us to make sense of data. Context provides the background against which we can understand the data. Then we need to organize it. Organization makes it actually useful. It helps us to more easily connect related data and piece them together. Thirdly, we need to analyze the data. Analysis helps us to interpret and draw conclusions from the data that we have. Analysis helps us to center our focus on what exactly we need to know.

These processes turn data into something useful, and that is information. And as we said before, what makes information useful is that it ultimately helps us to make decisions. Now this is where an information management system comes in. It is the information management system that transforms the data into information and makes that information available to all of the people who actually need to act on that information.

00:15:15 Benefits of an information management system

Let's talk about the benefits of actually having an information management system. Ultimately, an information management system or an IMS helps to ensure that you have timely access to the information you need about your project. Such information can help you do a number of things.

In the context of our nutrition project, this information might help us to optimize our project implementation. It might tell us about the specific ways that we might deliver our meals to schools. This can inform the way that we allocate resources in our project and uncover some ways that we might go about that more efficiently. Some of this information might help us decide whether we need to pour in more efforts into specific locations that are harder hit by stunting.

Thirdly, we can use this information to communicate to stakeholders. This is also, in its own right, quite an important thing to do because some stakeholders have the power to affect the way that your project is being implemented, for example, donors or partners who are directly involved in the implementation. Finally, this information will help to inform evaluations that we would do at the end of our project. Of course, we'll need to know whether the project overall was successful. We'll need to generate some lessons that we can then apply to the broader programming around nutrition.

Ultimately, having access to this information and acting on this information, we hope that this will lead to better project outcomes.

00:18:10 Components of an information management system

Now that we have a better sense of why we need an information management system, let's then take a look at how these systems look like in the real world. Is an Excel file an information management system? Many of you mentioned Excel or other spreadsheet-type tools. Indeed, these would form part of an information management system. These Excel files might be used to contain registries or beneficiaries or some other data that you're collecting. So this is part, but it's not quite the full system. There are a few more things that need to be in place to be considered a full information management system.

Beyond just a single tool, an IMS really is an integrated group of the people, processes, and indeed the tools that encompass all the activities that are needed to generate information. These activities include data collection, data processing, data organization, data utilization, and governance, performance, and maintenance. Although IMSs can come in different shapes and sizes, you'll generally need to have all of these activities in place in some shape or form to have what we can consider a full system.

00:20:15 Component 1: Data collection

Data collection is, simply put, bringing data into your system. Let's take a look at this in the framework of our system of people, processes, and tools.

On the people side, we'll need to have some enumerators for actually collecting data on the field and of course any other field staff who are working to implement your project on the ground. These are the people who are actually interacting with your beneficiaries or going door to door collecting the primary data that you'll be working with.

Under processes, of course, it's the whole data collection process. You might be doing surveys as part of that. You might also be doing some imports if data is being collected through some other way, through some other system by some other partners that you're working with. In that case, we need to have an integration process in place.

Some of the tools that you'll need to work with are data collection tools. Many of you mentioned things like Kobo; that's definitely relevant to this component. And you might need to use APIs as well, again in the context of if you need to integrate with other systems where data needs to flow from and to other systems.

When thinking about designing for this component, we need to think about efficiency. We don't have the luxury of all of the time in the world or resources to hire as many enumerators as we can. We need to think about the quality of the collected data. This is the starting point for ensuring that the information that you get out of your system is actually useful. We've all heard the phrase "garbage in, garbage out." The third thing we need to think about is alignment with other systems.

00:23:30 Component 2: Data processing

For this component, we're talking about making the data useful. On the people side, there are a couple of different people who might be involved. Generally, you will see information management officers, monitoring and evaluation officers, and in some cases data engineers who would be involved with this component.

These people will be running processes related to validating the data that comes in, making sure that it conforms with the structure that you expect, cleaning the data if it doesn't conform, making sure that the formats are all aligned to your expected structure, and addressing any missing data. Deduplication is another big process. It's really important for your system to not have any duplicate records. Also, we're often collecting sensitive information about people, so anonymization will definitely be one of the important processes that we implement.

In terms of tools, of course, we'll need to use some data cleaning tools to do these processes. In some cases, you might need to run some scripts that transform the raw unstructured data into data that's more structured. One thing to keep in mind is to make sure that we apply any business rules that we set in a consistent way. Another thing is around automation. Wherever we can invest in automation, that will go a long way in making this component more seamless.

00:27:08 Component 3: Data organization

In this component, we're talking about arranging the data for actual use. Under people, we're dealing with database administrators. Specifically when we look at processes, the first crucial component of that is data modeling and then, based on that data model, designing the database. This is very important to spend some time on upfront when you're thinking about creating a new information system because that sets the scene for how your data will actually be used and transformed into information.

The main tool that we're talking about in this component is a relational database. I would say that as much as we've got spreadsheet tools like Excel or Google Sheets that can serve as "databases" or registries, that misses the key benefit that having a relational database offers, which is that it helps you to establish the relationships or the connections between the different kinds of data that you're working with.

There are two considerations I would highlight here: data integrity and minimizing redundancy. If you've got a well-organized database, that means that all of your data comes together in a way that tells you the complete story of what's actually happening. Having redundant data slows down the performance of any database and makes it more confusing for your end users.

00:30:05 Component 4: Data utilization

This is all about actually transforming the data into actionable information. Under people, we've got some of the usual suspects: the information management officers and the M&E officers. Of course, you want to have people who are involved in data analysis. So in some organizations, you might have an actual role for data analysts who are involved in this component.

Some of the processes related to this are data transformation and reporting. This is highly relevant if you're working with lots of different data sets and data that's coming in from many different places. Reporting involves putting all of that data into a format that allows you to disseminate your findings and your learnings to your stakeholders.

Tools that we might need to use as part of this include business intelligence tools or other tools that allow us to create some dashboards. You might also need to use some APIs in this case if you have some other reporting systems outside of your data collection tool or your database tool that you need to push data into.

In order for our system users to be able to fully utilize this data, we need to design the system in such a way that it optimizes workflow. Think about how this fits in with the day-to-day of, say, a case manager who needs to work with service delivery for individuals that they're supporting.

00:33:05 Component 5: Governance, performance, and maintenance

This component is all about ensuring that the system is compliant and performant over time. We want to make sure that our system is sustainable, that it continues to generate the information that we need consistently, and that we are working within the bounds of what we are allowed to do within the jurisdictions that we operate in.

Under people, we're probably needing to involve some new actors into the mix. In some organizations, you might have some people who hold a data steward role. You will likely need to involve your IT staff who are managing the hardware that is involved with storing this data. Then again, you're going to need to involve database administrators.

Data governance is a big process here. There are probably a number of different procedures or policies around data governance that you need to have in place. We also need to think about data security and retention, making sure that only the people who legitimately need to have access to data do have that access. Identifying what roles you have in your organization and building processes in line with what's expected of each role is crucial (role-based access control).

Finally, performance monitoring involves looking at how our system is able to generate the information that we need. Is it working slowly? Do we need to wait a few weeks for a report? All of those questions should be answered by having a performance monitoring process in place.

00:38:10 Users of information management systems

Who are the users of the information systems? Of course, you'll have the project staff who are in their day-to-day making sure that the project activities are being implemented as scheduled. You'll have the M&E staff who are concerned with collecting the monitoring data that's needed to assess whether the project is meeting its intended outcomes. Information management staff are most likely the people responsible for building the system in the first place and maintaining that system over time. Then you might have senior leadership who are interested in seeing the ultimate, higher-level insights.

There are also some people who are maybe outside of your organization who will need to access this. We can think about partners; you might be implementing a project in collaboration with external partners or as part of a consortium. We can also think about donors who need to get information that tells them how well their funds are being used. And finally, of course, the communities that we're working in. Community members have an interest in information that tells them if this project is actually helping their community.

00:40:55 Qualities of an effective system

How can we tell whether our IMS is actually effective? An effective information management system is a system that generates information that is actually used. There's no point in having a system if people don't actually use that information in the end. But in order for that information to be used, three qualities need to be in place: the system needs to be compliant, it needs to generate reliable information, and it needs to generate actionable information.

Compliance means that the usage of the system aligns with the relevant policies. We'll be more likely to have this in place if we have established data governance policies and are actually enforcing those policies.

Reliability means that the system can be trusted to generate complete and credible information over time. We can ensure this by enforcing data quality measures at the start of our data collection process and thinking about the capacity of the users who are involved with handling data. Another thing is having a sound data model. The system will be reliable if we have defined and repeatable processes that are consistently applied.

Actionability means that the system enables users to take action in a timely manner. Information that is generated should be relevant to the stakeholders' requirements. Again, having a sound data model is important for actionability so that the information is easily accessible for us to take action on. Having a system that's deeply embedded into your existing workflows will go a long way to making sure that the information is actually being used.

00:45:50 About ActivityInfo and key takeaways

Just want to leave you with some high-level practices that we can keep in mind to make our system more effective. First, understanding what information you actually need. Second, having a designated champion who can serve to put all of this together. Third, building capacity across all of those different user groups. And finally, leveraging technology.

Speaking of leveraging technology, we ask you to consider investing in one piece of technology called ActivityInfo, which is an all-in-one information management software used by dozens of organizations around the world. You can see here some of the things that ActivityInfo can do, like tracking your project data from activities all the way to outcomes, managing data about your beneficiaries, and collecting survey data. You can do all of these tasks online and even offline.

Just a very brief look at how ActivityInfo can support your own project by looking at this case study from one of our customers, Conservation International. One of their programs is the Amazonia Verde Initiative, which adopted an integrated system using ActivityInfo to centrally manage their project information. This led to enhanced data accessibility across multiple levels and facilitated the collection and consolidation of key program data across different teams.

That takes us to the end of our presentation today. Some key takeaways:

00:48:40 Q&A session

Question: Between validation and cleaning, which activity comes first after the data collection activity?

Answer: That's a good question. In some cases, it might depend on the actual tools that you use. You might have access to some data collection tools that already have built-in validation functionalities. So in that case, the data that comes in already has some validation that's already happened. But in other cases where that's not possible, the data might need to be cleaned first in order to conform to the structure that you expect. Then you want to have a validation process around just reviewing whether the data actually makes sense according to your business needs.

Question: What is the difference between data use and data utilization?

Answer: I personally like the term utilization because that connotes that we're making the most out of the data that we have. We might have some data use, but are we using all of the data that we have available and in the way that is most valuable to us? Are we transforming the data in the way that's most useful? Are we combining all of the data that we could combine to get the insights that we need? If we answer all of those components, I think that goes more into utilization.

Question: Distinction between information management system and data management system.

Answer: I do think that there is a distinction. Data is more to do with single data points that are often raw or unstructured. Whereas information is data transformed, contextualized, organized, and analyzed. So a system that simply has data is different from a system that is able to generate information. I would argue that you would want to build an information management system over and above a data management system.

Question: Are the tools we use from ActivityInfo free?

Answer: ActivityInfo is a commercial off-the-shelf platform that you can purchase at an annual subscription. We do have a free trial in place. If you're interested in just seeing how it works, feel free to sign up for a free trial at activityinfo.org.

Question: How to deal with "death by a million spreadsheets" and converge data seamlessly into a proper IMS?

Answer: The first thing that you'll need to think about is having a data model. Take a step back from those spreadsheets and just think about conceptually what are the different data entities that you work with and reflect on how they interact with each other. Then you can start to think about how the tabs in your spreadsheet can be joined together into a relational database.

Question: What is the difference between M&E and IMS?

Answer: There's a lot of overlap. You can think of an M&E system as perhaps being a sub-segment of a fuller information management system. If you've only got a system that deals with monitoring, evaluation data, indicators, and reporting on your progress to targets, maybe you've only got an M&E system. But when you start to collect more data across all of the different domains of your project that help you to answer broader questions, that's when you start to have a fuller information management system.

Question: Knowledge management systems in relation to information management systems.

Answer: Knowledge management takes it another level up. Once you've consolidated your information, drawn out your conclusions, and synthesized them into learnings, that can then be codified and documented in a broader knowledge management system. An information management system helps project stakeholders to take action, often immediate action, to help move their project forward. A knowledge management system perhaps contains broader information about the project that is useful context and useful for applying learnings over the long term.