Developing a bitesize data strategy

A template for thinking about data for strategy and strategies for data for nonprofit organisations who are time-poor, just getting started in their data journey, or feel overwhelmed and need a ‘reset’.

In working with organisations through the programme of strategic data use workshops, I’ve been challenged to develop resources that hit the ‘just enough ‘ sweet spot of challenging and stimulating thinking, but without overwhelming people or complicating things to the point where people disengage or fail to see how it can be applied to their own situation.

The Data Learning Loop

The first step is being able to step back and see the bigger picture. In the past (p.150) I’ve used the single and double loop learning model to understand knowledge creation and learning loops, and it seemed like a good model to apply in this context too.

The Data Learning Loop model is a simplified view of how the different elements of an organisation feed into data actions, and the subsequent ‘levels’ of learning that take place. It’s not by any means perfect, but workshop participants seem to have found it helpful for taking a step back and thinking about the high-level structure and flow of their work.

1. All nonprofit work starts with an organisation’s values, which may manifest in its mission, vision, and ethos.
2. It is guided by an organisational strategy, which may be codified in an explicit strategy, or be seen in a theory of change, broad strategic questions, impact indicators, or even project aims.
3. The data strategy should help organisations to use data to deliver their strategy (among other things), and I’ve included some suggested sections of the data strategy (discussed below in ‘Strategy for data’).
4. The physical ‘data work’ broadly contains four actions: collecting data, supplementing data, analysing data, and communicating with data.
5. We then critically reflect and evaluate the outcomes of our data action at three ‘levels’ of thinking. The first is single-loop learning, where we ask “are we carrying out our actions in the correct way?”. The next is double-loop learning, where we look back at our guiding strategies and ask “why are we doing what we do?”. The third is triple-loop learning, we go back to our underpinning values and ideology and ask “how do we even decide what is right?”.

As I said, it’s not perfect; like all models it excludes some things in order to prioritise the display of others. It’s quite simplistic and it fails to recognise the role of process evaluations or agile learning loops, but I think it’s a useful starting point for framing our thinking.

Data for strategy

There are a couple of ways of thinking about how data and strategy interact. I like Marta Stelmaszak Rosa’s split of ‘strategy for data’, ‘data for strategy’, and ‘data in strategy’.

To help us think first about ‘data for strategy’ the first page of the Bitesize Strategic Data Use template encourages us to consider what it is that we want to use data for. This could be a desire to influence someone, demonstrate expertise or impact, or simply begin exploring a dataset.

At this point, it’s helpful to consider how we approach the data. We could 1) approach it with a hunch, hypothesis, or strategic aim in mind, or 2) approach the dataset with an open mind and see what conclusions can be drawn from it. In reality, we often use a combination of both approaches, but it’s useful to reflect on how you are approaching the data and how that may inform your conclusions and course of action.

The rest of the first page of the Bitesize Strategic Data Use template asks who the end audiences are, how the data use supports specific strategic aims, what data sources (internal and external) can be used, and what the next steps are to delivering this outcome.

As with the The Data Learning Loop, it prioritises these categories at the expense of others. For example, there is no reference to resources, responsibilities, or a detailed description of the final outputs, but I think it’s a useful starting point for distilling our thinking.

Strategy for data

Whilst the first page of the template refers to discrete uses of data, the second page looks at the essential elements of a broader ‘strategy for data’. It includes a series of prompting questions under the headings ‘Capacity and tools’, ‘Ethics and responsibility’, ‘Data quality and limitations’, ‘Ownership and culture’, and ‘Capacity and storage’.

Capacity and tools	Ethics and responsibility	Data quality and limitations	Ownership and culture	Sharing and storage
– What could be done to increase organisational capacity for data use and data literacy skills? – Are we using the right software and approaches for the questions we are trying to answer?	– Who could be affected by us using data in this way? – Do we have the necessary data policies and agreements in place? – What steps could we take to minimise any risk of harm?	– Is the data formatted in a standardised way? – Do we know the source of the data, and any limitations of it or biases it may contain? – What level of detail is provided, and what other data can it be linked to?	– Who is responsible for the (ethical) use of data in the organisation? – Is data use supported by senior management and trustees? – Is data used to question practice?	– Do we know what data is stored where? – Do the right people within the organisation have access to the right data? – Who outside of the organisation would benefit from having access to this data?

It might be useful to jot down some brief responses, or raise the questions in a team meeting. It’s not exhaustive (e.g. it doesn’t consider who will undertake the work), but it’s a useful starting point. The template contains space for considering the next steps for an organisation; this might be a useful place to write down aims to be accomplished under this heading, and any immediate barriers or work needed to progress them.

Once you’ve worked through both pages, there’ll be further work needed to codify the strategy into a phased action plan, considering who does what and when, how you’ll know you’ve been successful, and how to bring the wider organisation on this journey with you (comms, resources, etc). But before this, I think the template is a useful launchpad for considering how you can and do use data strategically, and where you might begin with developing a strategy for data.

If you would like to chat further about the Data Learning Loop, data for strategy, or strategies for data then please do get in touch. I’ve put some resources that I’ve found helpful on the resources page.