Frameworks and Tools We Use


We find it helpful to conceptualise the impact we aspire to achieve with each grant or investment as a cube. The larger the cube, the greater our contribution to a better world.

Funding research

 

The size of the cube is influenced by choices we make on 3 axes: Cause Area, Solution, Organisation/Team.

 

Axis I: Cause Area

All else being equal, the odds of having a big impact (i.e., large cube) varies greatly depending on our choice of cause area.

Framework: We find the Importance, Neglectedness, Tractability, (INT) Framework very useful for comparing the impact we may achieve across cause areas.

Importance: A cause area where there is still a lot of (future) wellbeing to be gained

Neglected: A cause area not many others are focusing on

Tractable: A cause area where the odds of finding an effective solution seem feasible

Open Philanthropy is a large philanthropic foundation that has an excellent track record of using the INT Framework

 

Axis II: Solution

All else equal, the type of solution that is used to make a change within a cause area matters greatly for the magnitude of our impact (i.e., size of the cube). In cause areas where we have a lot of high quality, independent data about the effectiveness of different solutions, we have seen that one solution can be ten or even hundred times more impactful than another one.

Framework: We don’t have the same type of high quality independent data about the cost-effectiveness for all possible solutions. We have found it helpful to first use a Problem-Solution framework that categorizes solutions into different buckets depending on our conceptualization of a) the complexity of the problem at hand, and b) the complexity of the chosen type of solution to this problem. The quality and abundance of scientific evidence on whether solutions actually work differs across these buckets.

Tool: Within a cause area, we can compare a grant or investment opportunity against a Benchmark. That way, we can estimate which opportunity will help us achieve the highest possible impact per Euro spent. We continuously work on identifying meaningful benchmarks.

 

Axis III: Organisation/Team

Within a specific cause area and solution, the difference between the competence of various organisations and teams influences the impact that can be achieved. What do we need to consider when comparing organisations/teams?

Framework: When gauging the competence of an organization/team, we find the Evidence, Beneficiaries, Implementation (EBI) Framework helpful. It looks at the competence of organizations across three dimensions: a) incorporation of the best available evidence for their solution(s), b) deep knowledge of the needs and preferences of their beneficiaries, and c) expertise on effective implementation of their solution(s).

 

Identifying Hits

Thinking consistently and taking into account high quality, independent data can help us to make choices on each of the axes in order increase our chance for maximum impact.

There is, however, a complementary strategy to this ‘top-down’ approach where we start with the cause area, then select the approach and consequently select the organisation/team.

The strategy does not start with selecting the best opportunity within a specific cause, but simply wants to identify those opportunities that have the most impact (i.e., the largest cube) and compare them against one another across different causes. Open Philanthropy Project uses this strategy and refers to it as ‘Hits-Based Giving’.

Tool: We find it useful to make Expected Value Analyses when considering possible giving/investment opportunities across causes. These analyses allow us to (try!) to explicitly quantify and compare between the value that a given amount of capital is likely to create.

Most Expected Value Analyses require making a range of assumptions. As probabilities and values are difficult to estimate, we have to be cautious about taking expected value estimates literally. They’re estimations, and we’re very likely to suffer from a strong Bayesian prior.

 

Read about how to use Benchmarks >>