This study provides a novel human-machine hybrid approach for creating learning about how knowledge management
interventions can improve the performance of agriculture and food innovations to achieve sustainable development goals. It
analyzes 2339 evidence resources, 500 of which are full texts reaching approximately 9600 pages of published high-quality
academic text. It provides a big spectrum of results and presents the main findings in a concise way. It informs the main
management question od what works in knowledge management interventions in improving innovation systems in middle and
low-income countries. It provides high granular information generated in North Africa, Near East, East Europe, Central Asia,
and contributes to overall IFAD learning in the NENA and CEA region and beyond.
It shows that only a tiny share of the evidence sources provided enough granularity to program better interventions. In addition,
existing information was mostly about general knowledge management rather than other operational aspects like capture,
retrieval, enhancement, etc. Moreover, how different types of knowledge management interventions differ has not been
sufficiently articulated. Although impacts were discussed intensely, how it happens through effective delivery and scaling
processes was hardly articulated. Most of the existing evidence is at the level of continents and regions rather than at the
country and sub-country levels. In most of cases, heterogeneous areas and countries were bulked into a single group hindering
the performance of knowledge management interventions significantly. In continents, the evidence available on different subregions varied significantly. Therefore,
The study showed that evidence sources informing about the role of knowledge management on SDG 1 (poverty) and 2 (hunger
and food security) were much less than the average of all SDGs. Also, the number of IFAD-funded authored and informed
evidence sources were minimal. Nevertheless, evidence could inform all IFAD result management indicators except the
government budget. Especially information on nutrition, the only IFAD indicator target that could not be achieved in recent
years, had the largest evidence. Furthermore, there was no single complete evidence resource for Moldova among some 2300
analyzed by the synthesis. Morocco and Sudan were in the medium range when it came to the availability of the evidence.
Therefore, the study confirmed that the geographical focus of the IFAD regional SKiM project addressed a gap. Therefore,
The evidence synthesis has also some natural follow-ups to design more effective and efficient knowledge management
interventions that can fast-track impact. The immediate follow-up is to extend the scope of the evidence synthesis to technical
reports and gray literature produced by IFAD and other key partners like CGIAR, FAO, Asian Development Bank, African
Development Bank etc. Knowledge management is a practical discipline in which learning is captured also in non-scientific
forms. The methods and tools developed for this evidence synthesis can reduce the time necessary to cover gray literature
significantly. Also, synergies with existing toolkits of IFAD, like the ones developed by ATHENA, could further increase the
gains for IFAD.
The study concluded that
1. Actionable, context-specific knowledge management intelligence is a bottleneck for designing transformative
delivery and scaling interventions required for impactful agriculture and food innovation systems
2. IFAD regional knowledge management efforts were crucial but insufficient in achieving the major poverty and
hunger ambitions. It is necessary to increase investments in knowledge management with a co-investment model.
3. Extending the evidence synthesis by leveraging existing IFAD and partner tools can be a quick win in designing
better-informed knowledge management interventions.