Last month, the Independent Commission for Aid Impact – the body which scrutinises the UK’s international development work – published a new report on ‘How DfID Learns’. Our aim was to examine how the department learns lessons from its past projects, and uses those findings to shape future projects.
As the lead commissioner for the report, I have read with interest the debate which has followed. Some feel that we’ve been too hard on the Department for International Development (DfID). But whilst DfID does a lot well, it has huge resources to draw on – and its failure to use learning consistently and systematically is a significant missed opportunity.
DfID has allocated at least £1.2bn for research, evaluation and personnel development (2011-15); and its staff are highly motivated, dedicated to their work and learning. DfID has done well in terms of generating knowledge and research, but this is different to learning. Real learning means turning knowledge into action, and this requires more than individually motivated staff. It requires a shift in the approach of the organisation as a whole.
We have seen many good examples of learning and innovation. Thanks to its engagement with intended beneficiaries in Odisha, the lessons of a great livelihoods project have spread across India. In the Democratic Republic of the Congo, DfID staff produced a powerful video that explained a programme’s failure and the lessons to be learnt: I was impressed by this determination to learn from failure. The Learning Partnership set up between DfID and NGOs has shown cross-sector sharing of learning, and the number of professional advisers dedicating 10% of their time to learning is growing.
So why did we give DfID an Amber/Red rating, meaning that significant improvements need to be made? Analysing the 30 reports we’ve produced since 2011, interviewing DfID staff, and analysing 1,700 comments in internal surveys – where much frustration was voiced – we found a very mixed picture.
Let’s pick out a few key conclusions. We found that DfID relies too heavily on individual learning and does not, as an organisation, learn systematically and consistently. It also appeared that strategy and oversight is not held at a sufficiently senior level. And the department doesn’t know how much it spends on evaluation, and doesn’t track or report on the overall impact of the hundreds of project evaluations it commissions each year.
We asked why the section on learning has been withdrawn from the system of annual reports. Other data is produced in large volumes, but staff struggle to put it into use. And managers and leaders are not held accountable for ensuring that learning fully informs decision making or increases the impact of DfID’s work.
Often, staff aren’t given sufficient time to learn – in particular from experience in the field, and from others who have had this experience (which is the learning staff most value). There is inadequate focus on continuous learning from partners, contractors and beneficiaries when programmes are actually being implemented. And we are very concerned that staff often say they feel under pressure to report the positive rather than use evidence to arrive at a decision. The emphasis on showing positive results can lead to biased and selective reporting and staff who feel afraid to discuss failure.
Why do we feel so strongly about this? Because of the resulting missed opportunities. DfID must ensure that expenditure on learning is always turned into action which benefits the poor. We’ve seen many missed opportunities of available knowledge not being used. And the more DfID works in fragile states, the more rapid the staff turnover, and the more dependent it is on those who implement, the greater the risk of DfID failing to learn from experience – both good and bad.
We believe that DfID needs to adopt a consistent and systematic approach to simplifying the evidence available through synthesis to make content more useful. The department needs to learn more, across its global span, from what is happening in the field – not least from the intended beneficiaries, as well as the many partners and contractors it’s dependent on. And it needs to make sure that staff feel able to share openly what works and does not work in practice, and act fast to make adjustments.
It is not just DfID that can learn from these lessons; many of them could apply across government. Our recommendations don’t require additional bureaucracy or systems. The trick is to focus on simplifying, stripping away, synthesising and making sure that every piece of information which is paid for is usable, practical, and focused on always ensuring that it is turned into real impact.
Click here to read the report
Diana Good is commissioner at the Independent Commission for Aid Impact.