1

General structure of the matrix

No activities are specified and the outputs are, in reality, project components. Inputs are given for the entire project and not for specific activities.

Structure the matrix as in the revised example (see Figure B-2), with a purpose for each component and each component having outputs and each output having activities.

2

Generality and the fragmented nature of the matrix content

The matrix provides only a very general overview of the project. In this form, the matrix provides insufficient detail to be a useful guide for project implementation. It is also not clear what is to be achieved under each of the components (outputs).

More detail needs to be included by using a purpose for each component, and outputs and activities for each purpose (see Figure B-2). This will make the matrix longer. However, a summary of the project in terms of the goal and purpose levels can be used on its own.

3

Contents of the second column of the matrix – targets

The targets do not adequately cover the different aspects of the project. They focus too much on quantitative outputs and inputs and not enough on outcomes and qualitative information. The targets do not fully cover the scope of the project for each component, so it is difficult to understand the project fully.

Use key performance questions and key target indicators as shown in the reworked example. Comparing the reworked matrix to the original version, you can see how having outputs for each component purpose makes it much clearer what interventions the project intends to make.

4

Contents of the third column – monitoring/means of verification

The monitoring mechanisms are very general and so provide little guidance for setting up the M&E system.

Include more about the information-collection methods. Additional information about M&E needs to be developed in a separate M&E matrix (see Annex D).

5

Contents of the fourth column – assumptions/risks

At the goal and purpose levels, assumptions are used. At the output level, risks are used. There is no rationale for this, as "assumptions" can be used at all levels.

A risk is an assumption that may not hold true. For example, there is a risk that the assumption about having a market for increased horticultural production may turn out to be incorrect. The revised matrix example includes only risky assumptions, and not killer assumptions or highly likely assumptions.

6

Gender and other equity differences

There is no indication from the matrix that gender and other equity differences have been specifically considered.

Ensure that, where necessary, there are specific outputs or activities that address equity issues. Include targets/indicators related to equity, for example, female-headed households benefiting equally from the project. Ensure that information is disaggregated according to gender and equity differences.

7

Generality of the goal

The goal "Improving livelihoods" is an extremely broad goal. Yet the project does not intend to directly tackle, for example, the health and education aspects of improving livelihoods. So the project implicitly has a narrower focus than the full livelihood goal.

Try to make the goal more specific and more representative of the different purposes. It may be necessary to give a more detailed explanation of the scope of the project’s contribution in accompanying documentation.

8

Targets for the goal level

Increased income is a poor indicator of the overall project goal. Income itself does not necessarily contribute to improved livelihoods. It depends how the increased income is used and how household expenses and work patterns have changed.

See the performance questions and target indicators for the goal in the reworked example (see Annex C).

9

Assumptions for the goal level

The assumptions are so general that they would apply to virtually any project anywhere in the world. So they are not very useful for guiding thinking about the long-term sustainability of the project.

The assumptions should indicate what to look for to see if the project is likely to be sustainable in the longer term. The reworked matrix gives a set of more specific assumptions for the goal.

10

Purpose level

In the original example, there is little difference between the two purposes or between them and the agricultural development output.

There are different ways to structure a logframe matrix. However, for IFAD-supported projects, it is suggested that a separate purpose for each component be used. It is also important to think carefully about whether a project is primarily to achieve a physical change, such as increased agricultural production, or whether it is to focus on institutional and community capacity and the process of development. A good project will achieve both. However, it is important to ensure that capacity-building and institutional development processes are made explicit in the logframe matrix.

11

Purpose-level targets

The purpose-level targets are essentially targets for specific aspects of agricultural development. They do not address the issue of increased capacity for self-reliant agricultural, economic and social development.

At the goal and purpose levels, it is important to ask broader questions about institutional change and how achieving specific production targets are actually contributing to improved livelihoods. See the performance questions for the goal and purpose levels in the reworked example.

12

Outputs

As mentioned above, what are called outputs in the original example are really the project components. If you look at the outputs as given, it is very difficult to get an overview of what the project aims to achieve. The outputs are written only as a title/heading and not as a result or objective.

Outputs should refer to a relatively specific achievement of the project. They should also be used to give a clear picture of the scope of each of the project components/purposes.

13

Agricultural development output

It is not clear what will be achieved under this component. The targets are unclear.

The reworked example shows clear outputs for the project under this purpose.

14

Agricultural development targets

"New crops adopted by 30,000 farmers" is a poor indicator. Taken literally it gives no information about what crops have been adopted, to what extent or how successfully.

It is necessary to make clear that information must be collected about what particular crops have been adopted and to what extent.

15

Agricultural development assumptions

The risks relate to achieving the component (output) and not to the contribution of the component to the purpose and goal. "Department of agriculture is unable to deliver technology" is an assumption relating to the achievement of the component. "Technology is not available" is potentially a "killer assumption".

In general, assumptions should relate to how an activity contributes to an output and how an output contributes to a purpose and so forth. For example, in this project, it is being assumed that extra production will, at least in part, be sold to increase household financial resources. The contribution of the increased agricultural production is based on an assumption about sufficient market demand and prices for the production. Clearly identifying assumptions is often a difficult part of the project planning process. Either the project should be changed to ensure that technology is available as a result of project efforts or the purpose and goal need adjusting to be less ambitious.

16

Community development output

It is not just the community where capacity development is required. For the project to be successful, the department of agriculture and private sector also need to build their capacity.

This output becomes an institutional development component at the purpose level of the matrix.

17

Community development targets

Most of the targets provided are activity or low-level output targets and do not answer the "so what" question.

Make sure there are performance questions that will provide information about, for example, how successful farmers’ groups are in supporting their members to adopt new farming practices.

18

Community development means of verification

"NGO and PMU monitoring" says nothing about what methods or even the general approach that will be used. The MOVs given are so simplified that they provide virtually no information to guide M&E.

Try to be as specific as possible about what monitoring mechanisms and sources of information will be used.

19

Rural credit targets

These targets are all input targets and will not provide information about the outcomes or impact of the rural credit scheme.

Establish performance questions and indicators that will provide information about repayment and for what the credit is being used.

20

Inputs

Inputs should relate to activities and not the whole project. The original example does not have an activity level.

The use of the second and third columns in the matrix change at the activity level. The second column is used for inputs and the third column for budget information. Monitoring activities is necessary, but it is easy to track what activities have been completed through basic project records. Consequently it is not necessary to provide details about indicators and monitoring mechanisms at the activity level.