Hello Gonzalo,

From my experience with research management for the US Navy, I can tell you facts about research agendas in general. To have a viable and long-term research program, you must be able to show progress in your research projects by using metrics. This will be key in obtaining continuing funding from your sponsors and in attracting additional sponsors. 

Take defense projects in the US, for example. During the first year, we funded many ideas that looked good. However, in future years, we funded only the projects that produced good results as determined by metrics. Some of the metrics we used can be applied to research in general. Defense is similar to agriculture because both are focused on practical applications. If the research does not lead to viable transitions into the production sector, perhaps it should not be funded. To be fair, I have notices that some projects take longer to produce results than others. Five years is a good time period to test a project. Good research projects should be significantly well developed to the point where they produce promising transitions into practical applications. This is the ultimate test.

However, along the way, you may wish to count the number of publications and patents, as well as to apply agricultural-specific metrics, such as reports of how many farmers a particular program has helped.

Your request for proposals should very carefully outline the evaluation criteria that you will use in the initial selection, as well as the metrics you will use to evaluate whether a continuation of funding is appropriate. Although some people will complain about this, you should enforce a strict adherence to format. Make sure your proposal authors submit a technical proposal as well as a cost proposal. 

Uniform format will make the evaluation process more efficient and you will be able to consider more proposals with fewer evaluators. Limit the proposal size for the sake of efficiency to, say 50 pages. Emphasize that even though the proposed researcher has stated a fact in response to one question, if it applies to another question the fact should be restated. This will prevent your evaluators from having to search the entire proposal to find the answer to each question. I have been on both sides of research - as a researcher and as an evaluator. If you have specific questions, I will be happy to help you.

You may want to browse my ResearchGate web page for more reports regarding research, knowledge management, DARPA programs e.g. HPKB, and metric.

Good luck with this.

Thank you Marion for your kind response and considerations, fully in accordance with the related metric.

A few principles you may want to think about

Agricultural research programs should include:

1. Farmers, their resources, requirements and welfare.

2. Provincial and national requirements or goals for agricultural products.

3. Current production levels, demand, imports and exports.

4. Cost:benefit ratios under different agroclimatic zones.

5. Agricultural inputs, their availability, their use rationalization and their effects on the environment.

6. Organic agriculture practices.

7. Research teams, funding, result-oriented goals etc. as mentioned by other experts

Roland and BR have brought up some very good points. First, two points from Roland's list of basic steps. My comments are after the ++

1.Identifying a research niche and planning an independent program

++The research niche should include a difficult problem that has widespread influence so the research will be worth the effort, but it also should address a tractable problem. These are not easy conditions to satisfy simultaneously. However the best research programs do satisfy the conditions.

2.Obtaining funding

++This seems to be the hardest step for many. Research managers need to consider what it will cost not only for the researchers themselves but also what it will cost to administer the program. Administration costs include writing the request for proposals, evaluating the proposals, and later, evaluating the results of the research.

++Now, from BR's list:

2. Provincial and national requirements or goals for agricultural products.

++This includes applicable laws. Addressing a high priority on the provincial or national level will increase the probability of success in terms of continued funding for multi-year projects.

3. Current production levels, demand, imports and exports.

++Ideally, you will want to consider how to increase the demand AND the production simultaneously, so a market will be available for the increase in produce. For example, if demand is high but production low, this could include an analysis of why production cannot keep pace with demand. If production is high and local demand is saturated, the analysis could focus on how best to export the products and how to remove the obstacles for doing so.

4. Cost:benefit ratios under different agroclimatic zones.

++Also in regions that have different laws.