Working towards a more open, equitable data ecosystem for charities
Introduction
**NPC** is funding and partnering with a project team to understand the data ecosystem around charities between now and February 2022 as part of their **Open Philanthropy** strand. Within that strand, this piece of work is aiming to highlight where funding should be focused to support better, more open and equitable use of data from, about or within the charity sector.
This work is just getting underway with a small project team led by **Tom Watson** from The Good Ship. The team also currently comprises David Kane, Mor Rubinstein and Tom French. Our intention is to centre this on existing / potential use cases and data interactions to establish typologies within them, their prevalence and what support should be resourced (and for/by whom).
In this joint blog from the project team, we set out our understanding of the work, why we think it’s important and some of the things we will focusing in on. We’re also after your input too, so please see the form at the bottom of this page…
Setting the scene for this work
Charities of all shapes and sizes provide, produce and use data. Sometimes they are required to by others — i.e. they have to do it. Sometimes it is because it helps them achieve their mission better — i.e they want to do it.
When a charity uses data (their own or others’) or provides their own data to another organisation — for example, for regulatory, funding or project purposes — a relationship is formed with that organisation for a particular use of the data. But that doesn’t mean that it’s the only and final use of those data…
The organisation receiving the data may be receiving similar data from other charities too. By combining these data from multiple sources, the organisation essentially creates a new data set. This new data set can be the starting point of a new relationship with another organisation for a different purpose/use. And so on. Here’s one of many potential examples…
All of these different ways the data are used are valuable to the sector as a whole, in terms of how organisations operate within it, how it’s described (by itself and others) and how others know the impact it’s having. They also benefit society in a broader sense as it can help to allocate resources — funding or otherwise — based on an understanding of who’s doing what and where.
The relationships, purposes and their benefits aren’t restricted to data which originate from charities alone either. Data from other sources, such as local and national government, can be the basis for relationships to form between organisations in the sector around a particular purpose(s) too.
What are we trying to understand?
There are a number of challenges for the sector and others in realising the maximum benefit of these relationships. These include, but are not limited to:
Not all relationships and purposes are visible to everyone.
- This, where appropriate, prevents others from re-purposing the data at the aggregate level for new uses.
- It also increases the likelihood of disparate, duplicate and incompatible approaches to data emerging in isolation.
Not all data relationships are reciprocal.
- This may introduce or exacerbate power dynamics that, in turn, foster inequality or inequity.
- This may lead to a weaker than possible data ecosystem.
- This may be symptomatic of concerns around data sharing and protection.
Not all relationships are as effective or efficient as others at making the data use as valuable as possible.
- The purposes and benefits for an organisation’s data that happen a few relationships away from the provider of the data are not always communicated back to them.
- This misses an opportunity to incentivise better data quality and sharing at source.
Here’s an example:
These challenges are understood well in principle. And there is a growing community and set of initiatives underway to bring people together around them. But the challenges have yet to be analysed to date in a way that focuses in on the specific relationships, uses and types of available support which could be resourced in order to overcome them in the most efficient way.
Building on and developing work that has gone before, this is why we’re looking at this now. And we’re going to do this in the open to encourage transparency; principles which will be be required for this work to be successful in the longer term.
Questions to answer
This piece of work seeks to answer 4 main questions:
1️⃣ Which current uses of data could be shared or brokered better, and what are their characteristics?
2️⃣ What uses of data are missing or being ‘blocked’ in some way, and what are their characteristics?
3️⃣ Where, to whom and for what purpose should funding resources be focused to make improvements based on the answers to the first two questions?
4️⃣ What power dynamics are at play in these relationships, data uses and the support utilised to bring them about, and what are the implications of these dynamics in terms of what future funding or support looks like?
What are the some of the more specific questions at the relationship and purpose level?
Underneath the headline questions above, there are a number of more granular lines of enquiry to take to unpick the data, the relationships formed around them and how the product of those relationships are used (or not).
Some terminology and why we’re using it
Throughout this work we will use a number of different terms. Some of these terms are commonly used, while others are emerging through other pieces of work. Either way, we feel it is important for us to highlight and describe/test some of these terms now to avoid ambiguity as this work unfolds.
🔁 Interactions
Definition: The relationships that form between organisations around a particular data set in order that a use can be achieved.
Why it’s helpful: These relationships are the mechanism through which meaningful data use happens, or could happen. They provide the opportunity for developing further uses for the data too. The culture and dynamics of relationships will also determine how openly data uses can be shared, learned from and promote equity.
🔓 Roles
Definition: The nature of how an organisation is involved in an interaction, such as whether they provide, consume, analyse or regulate the data, knowingly or otherwise.
Why it’s helpful: Understanding how people interact, the power they have and what types of organisations tend to fulfil particular roles will help to identify where support may be required, for whom and in what circumstances. This will also help to understand which organisations in which roles do not receive feedback on how their data are used.
✅ Use cases
Definition: A contained end-to-end process where data are identified, a relationship — an interaction — is formed around them and the use(s) are known, including the tools employed to get there.
Why it’s helpful: Identifying and analysing use cases in isolation will help us to understand the constituent parts leading to meaningful data use in this context. It will also help us to understand the prevalence of particular use cases, where there’s duplication and what things are most likely to lead to onward use.
🔗 Chains (linear or cyclical)
Definition: The sequence of linked use cases that build on raw data to create different uses at different levels for different purposes.
Why it’s helpful: Chains are symptomatic of more openness within use cases. They may also be representative of habitual ways of working that could be reshaped to promote equity better and meaningful use for more people. Learning about chains will help us to understand the ingredients for linking together use cases effectively and providing the necessary feedback or incentives for different roles within interactions that bring about desired outcomes.
🚧 Blockage
Definition: Something within a chain or a use case that stops the original or further use being achieved.
Why it’s helpful: By identifying these things and understanding the use cases or chains within which they are most prevalent, we can think about some of the ways by which we can improve data use or build longer, more open chains.
⤴️ Inefficiency
Definition: A point within a use case or chain that could be tweaked to be better at bringing about effective data use for the most people.
Why it’s helpful: By identifying these things and understanding the use cases or chains within which they are most prevalent, we can think about some of the ways by which we can improve data use or build longer, more open chains.
⚙️ Catalysts and conditions
About this term: The support, resources, culture and/or organisational traits that facilitate better, more open interactions, use cases and chains.
Why it’s helpful: Understanding and defining these things — external and internal — will guide where funding should be deployed or expanded upon, for whom and in what circumstance.
Note: We recognise that other terminology and methodologies exist in this space to describe data ecosystems, such as ODI’s Data Ecosystem Mapping Tool, for example. For this specific work, however, we wanted to settle on a few key elements and define them for the context within which the work sits.
What we’re doing next
To answer all of these questions, we’ll be doing three main things between now and the end of February 2022:
- reviewing existing initiatives or research to uncover and categorise the use cases that have happened to date;
- having as many conversations as possible to reflect, test and refine our assumptions and ideas, as well as learn about things we may have missed;
- creating open outputs of everything that are designed to be built upon.
By doing this, we hope we will find a clear way to describe the current landscape, and the challenges or opportunities it represents. In turn, this will help us to make recommendations for where funding can best support development in this area to maximise the openness and usefulness of data in the sector, whilst minimising the risk of compounding any existing power dynamics.
How you can help and get involved
❓ Can you point us in the direction of existing resources that may have documented or categorised some of these use cases, opportunities and challenges before?
❓ Do you have a use case you can share with us, successful or otherwise?
❓ Do you have any thoughts and ideas on how we have framed things in this blog
Tell us:
For reference…
NPC are funding this work to begin understanding the following questions:
- What does the data ecosystem look like in the social sector?
- How can mapping and understanding the landscape help us identify opportunities to accelerate practice?
- In particular, where are there opportunities for open data, data sharing and reuse, data standards, and other collective approaches?
