Our Programmes



Sign up to our newsletter.

Follow ODI

Defining the problem

The pressure to deliver results often limits the amount of time available to define the root cause of a problem. This is a false economy, as it can lead to projects and programmes that address symptoms rather than causes. ROMA helps you understand the root cause of the initial problem you have identified. Two different options are presented below. The ‘five whys’ technique is useful for a first approximation at finding the root cause. A fishbone diagram helps you delve into the issue in more detail.

A first approximation: the 'five whys' technique

The ‘five whys’ technique asks you to identify the initial problem and then answer why it is a problem five times. After the fifth ‘why’ you will have reached a real depth of understanding about the issue. This helps go beyond the initial issues or those that are immediately apparent, to work out what is causing the problem and where the most effective entry points are. Below is an example from a project on public expenditure and financial accountability (PEFA).

Five whys in practice

Public Expenditure and Financial Accountability (PEFA) indicators are widely used to assess whether a country has the tools to deliver good fiscal discipline, to allocate its resources strategically and to use its resources for service delivery efficiently. A suite of indicators is used to make these assessments (see www.pefa.org), one of which relates to public sector procurement, which we explore here.

The initial problem statement might be: ‘the problem is that we get a D on the PEFA procurement indicator, because we do not have a law requiring competitive bidding across government’.

1. Why does this matter?

Without this law there is an incentive not to use competitive bidding in procurement deals.

2. Why does 1. matter?

Without this incentive, most procurement deals are currently done through sole source methods.

3. Why does 2. matter?

Sole source methods can increase corruption and lead to higher procurement costs and lower quality.

4. Why does 3. matter?

We have evidence that many procurement deals have been overly costly and goods are poorly provided.

5. Why does 4. matter?

High-cost, low-quality procurement is undermining the provision of key services across government.

This process of the initial specification of the problem – the need to introduce an externally defined ‘best practice’ to mandate competitive bidding – is in fact prompted by the need to improve the cost and quality of procurement. The latter problem is much more complex but is the root cause that needs solving. However, it is unlikely to be addressed by simply mandating the use of competitive bidding.

It is very unlikely that you will need to ask ‘why?’ more than five times. In some cases, you may find you have reached the root cause of the problem after only three or four rounds of questioning. You will know when to stop because the answers will begin to broaden out rather than narrow down.

Delving into the detail: fishbone diagrams

The ‘five whys’ technique may give you sufficient information to begin to construct a robust objective for your policy-influencing work. However, if it is clear the problem contains several different components that need to be broken down, a fishbone diagram can help you get into the detail (see 'Five whys'). Sticky notes are helpful to brainstorm and group the sub-issues that together contribute to the main issue you are trying to address.

Five whys

Every year, 2 million Nepalis travel abroad for permanent and semipermanent work, leaving through a network of formal labour agencies and informal recruitment agents. A large number of them are extorted – more than 75% pay above the legal maximum fees to secure a job. A large number are also exploited – forced to work inhuman hours in difficult conditions, paid wages considerably lower than promised, being sacked when they fall ill.

This exploitation and extortion occurs along the chain of agents and agencies through which migrants travel in Nepal, among middlemen in India and receiving countries and among employers.

Figure 3 shows the component parts of this problem. Policy does play a role but it is not the most important one. Although laws are in place to penalise Nepali manpower agencies proven to be mistreating migrants, very few such agencies have actually been punished. Middlemen are unregistered, which makes it difficult to enforce laws, and the formal labour agencies that oppose reform have political ties.

Competition for places reduces the demand from Nepali migrants for better treatment. Since the Gulf is a considerable distance from Nepal and Gulf governments are reluctant to lose a source of cheap labour, there is no real political appetite to reduce the exploitation.

Figure 3: Fishbone diagram of the problems facing Nepali migrant workers

Figure 3: Fishbone diagram of the problems facing Nepali migrant workers

  • Begin with the problem as it presents itself.
  • Brainstorm all the individual causes you can think of for that problem: this is best done if it has been supported by detailed research and analysis.
  • Group these causes and develop the diagram as above (you do not need to limit yourself to three sub-groups, but more than that may prove unworkable).

Figure 3
shows the issues Nepali migrants face as they look for overseas employment. It was developed to help design a programme to reduce the high costs of migration through packages of technical assistance and other types of support. The process began with detailed reviews of the literature on migration, interviews with policy-makers and representatives of migrant workers and wider consultations.

Analysing the stakeholders of a problem

Whether you have used the ‘five whys’ technique or a fishbone diagram, an important step is to map which people/organisations/bodies have an interest in the problem – that is, the stakeholders. Both techniques will probably have shown you that a wider range of stakeholders are involved in the problem than was initially apparent.

A stakeholder mapping exercise helps you understand what drives their interest, influence and actions or explains their positions in a programme. It can generate a large amount of useful information about the relationships between different groups of people and how those groups are likely to behave when confronted with the possibility of change.

ROMA uses a simple 2x2 matrix, the ‘influence and interest matrix’ (Figure 4), to map the stakeholders, in four steps:

  1. Clarify the project's overall objective using the 'five whys' or fishbone diagram process.
  2. List all the stakeholders you can think of: it is helpful to put each one on a sticky note.
  3. Draw the axes of the map, as below, on a large sheet of paper, and place the sticky notes on the map. Begin by working out which stakeholder represents the extreme of each quadrant, and work from there. Your choice of where to place them on the map should be informed by some form of evidence. You could write the evidence on the back of the note. If you are working in distinct regions or countries, it will be helpful to construct different matrices for each one.
  4. Go through the sticky notes and work out if you need to break any of them down to identify specific teams or individuals who have different degrees of interest or influence.

Figure 4: The influence and interest matrix

Figure 4: The influence and interest matrix

It is particularly helpful to do this as a group exercise, either with your immediate team or with your primary stakeholders – those you are working closely with to bring about change. Ensure a good variety of people in the group; include those working on communications and M&E, not simply the people implementing the project or programme.

The greater the range of people the more detailed your analysis will be, though you need to balance this with considerations of how you will manage the process. If the group is large (say, more than eight people), it may be useful to allocate someone the role of facilitator, to keep the process active.

Discussions about who is influential, why and what forms of interest they show in an issue can uncover important relationships between the stakeholders that you can subsequently use to develop your influencing objective. It will also make it more likely that you will consider the full range of people and organisations that need to be included.

There are three things to remember when constructing an influence and interest matrix:

  1. Not all people in an organisation will have the same degree of interest and influence. It may be worth separating them out so you can target the main opinion-formers or blockers of change.
  2. It is important to consider all the stakeholders, not just those immediately involved. Separating out your primary and secondary stakeholders using different coloured sticky notes is a helpful way of ensuring you develop the matrix in sufficient detail: the more detail there is, the more likely you will be able to spot common characteristics between groups. This may help you develop your communication strategy (see Developing your engagement strategy).
  3. It can help to work out what incentives may drive each group to either support or oppose change. It is useful to consider the connections, networks, loyalties, patron–client relations, alliances and points of conflict among and between the different groups. Mapping connections between the stakeholders may also be useful.

Using the influence and interest matrix for large or complex projects

For a large multi-country project based at ODI, this matrix was developed in two stages. One stage was done in London by the small project coordinating team, to map the international actors who would be interested in the results of the project, such as donors. The aim was to understand how best to communicate and discuss emerging findings.

The other stage was done by the project teams in each country, working with their local collaborators to draw up a map that would help them understand how best to engage with different actors in the policy process in each country.

Also in this chapter:

  1. Diagnosing complexity and uncertainty
  2. Systemic factors: the political and institutional environment

Visit our ROMA homepage