Recently the founder and CEO of a small but highly effective Nigerian CSO working to empower marginalised women and girls gave a plenary speech at a major pan-African Conference. In the run up she had reached out to her network for current data. She wanted to highlight some key health indicators of vulnerable girls living in informal settlements in Africa’s major cities.
It turned out there was no easy way to come by such data. The few published articles referenced datasets at least 10 years old; data available from online portals was too high level to be useful. In the week leading up to the conference she gets a break; there is recent survey data available from a credible African research and thought leadership organisation. After contacting them and explaining the context and urgency she is told to make a formal request through a form on the organisation’s data portal. An automated response tells her the application will be considered and to await the response. Sometime later she receives an email telling her to log back in to learn the outcome. The good news is the request has been approved; the bad news is over a week has passed and the conference is over; also the available data is in a file format she doesn’t understand and can’t read. There are some summary reports available but these are in PDFs with no easy way to access the underlying datasets.
This is a true story — and it’s far from unusual. Right now the vast majority of African organisations working in the development space do not have ready access to data they can use for policy advocacy and programme delivery.
It’s instructive to consider a bit more deeply why this is so — and there are clues in our example above. Much of the conversation about data and development in Africa (including my own earlier piece on Building Takwimu) tends to talk in terms of ‘data shortages’; a ‘lack of high quality data’. But this language doesn’t capture the problem in practice, because there is no outright shortage of data for policy making in Africa. Excellent policy relevant data is being generated through official government channels, market transactions and the activities of multiple foundations and CSOs.
The immediate problem is most of this data is not finding its way outside of siloed programmes and value chains — and the underlying problem is actors are not incentivised or enabled to bring this data into use around a shared policy agenda/process.
So what’s happening? To try and understand this better I met and discussed the issues with a variety of people over recent weeks during a visit to Nigeria, Kenya and Uganda — across government, CSOs, academia and industry — focussing throughout on a simple positive question : “How can we get more high quality African data into use for effective policy advocacy and programme delivery”.
The answers are understandably multi-faceted, but some clear strands emerged, with implications for policy and practice. For simplicity let’s think in terms of two broad organisational scenarios:
A. Organisations that generate primary policy relevant development data, through surveys and other forms of original research or as a bi-product of transactional activity.
B. Organisations that can put such data to use to advocate for better policy, or to design, communicate and deliver better services, e.g. in health, agriculture, education or financial inclusion.
Considering the former first — why doesn’t more of these organisations’ data find the light of day to inform policy and practice? Here are some of the issues:
- Commercial data sources don’t have ready/easy licensing and tracking mechanisms for reuse for non-profit/humanitarian purposes.
- Government bodies are wary of releasing their data due to concerns about quality/consistency, formatting, comparative standards, licensing regimes and (political) reputational risk.
- CSOs inconsistently release data because of concerns about quality/methodology, formatting, sample size/relevance.
On the demand side of the equation there are many media houses, CSOs, policy makers and planners who, if they had effective access to this data, could add value and repackage it to build stronger advocacy positions, or for programme design and delivery. However, much of the time, even where data is openly available it remains effectively inaccessible. To some extent this reflects capacity constraints in their ability to discover, manipulate and re-use available data as actionable content, especially in smaller/grassroots actors. But it also reflects problems in how data tends to be published — spread across multiple different online locations, frequently locked in PDFs, poorly described, lacking context, explanation; not visualised for ease of understanding.
Economists call this ‘market failure’. One one ‘side’ there are organisations who can supply policy relevant data; on the other there is a strong and growing demand for that data. But the market does not efficiently bring these actors together to support greater use of data for policy making and programme design, despite the huge economic value this would bring.
What has all this to do with Takwimu? Quite a lot. Based on our experience to date, I’m convinced:
- there is an important role for an authoritative neutral platform that can reduce costs and risks for multiple players to publish, discover and reuse high quality original African datasets.
- Takwimu’s ability to contextualise, visualise and build insights on top of development data can make it substantially easier for a range of different actors to build evidence based advocacy positions and delivery better targeted interventions.
In the next stage of Takwimu’s development we’re going to be a market maker, mobilising African data out of silos and into the hands of changemakers.
*thanks to Justin Arenstein for insights and inputs.