How do I add funding flows together without double-counting?
This tip applies to the data search. The new FTS database is structured completely differently from the old one. The most important change is that FTS can now connect flows together into a chain. For example, a donor gives a UN agency $10m, but that agency then transfers some of those funds - $8m, let's say - to implementing partners. Although those two flows are both registered in the new FTS database, we can no longer simply add them together to get total funding.
Instead, the "flow model" which drives the new FTS allows us to aggregate flows correctly by drawing 'boundaries'. For example, if we visualise all of our flows as interconnected lines on a big diagram, then we can draw a boundary around any entity - such as a country, plan or organization. Take the aforementioned UN agency as an example. If we draw a boundary around that entity, we can now easily work out which flows are incoming - i.e. funds being transferred to the entity (in this case, the $10m donor funding), and which flows are outgoing - i.e. funds being transferred away from the entity (in this case, the $8m funding being given to implementing partners). There is also a third category, namely internal flows - those which represent funds being transferred entirely within the chosen entity, e.g. from the UN agency's HQ to one of its country offices.
Now, adding flows together becomes easy again. To see how much that entity has received, we can add all the incoming flows together. To see how much it has donated or transferred to implementing partners, we add up the outgoing flows. This way, we avoid any double-counting.
What's the difference between a boundary and a filter?
This tip applies to the data search. When you start a customised data search, you initially specify your search terms and then click on 'Get Data'. This sets the boundary for your search, and determines which flows are going to be considered as incoming, outgoing, and internal - relative to that boundary. You can think of this as the point of view, or perspective, which you are taking on your data.
Once you have the data on the screen in front of you, you can now also set filters which further refine the data by excluding certain flows. However, the filters are not changing your perspective - the boundary stays the same.
To illustrate this, imagine the difference between the following two cases. In the first, you choose the country 'Sudan' as your boundary, and then filter by a UN agency such as 'Unicef'. In the second, you choose both Sudan and Unicef as your boundary. How will this affect the results that you see?
Consider a funding flow made by Unicef's Sudan country office to a local Sudanese NGO who is one of their implementing partners. In the first case, when you select 'Sudan' as your boundary, this flow will be considered internal from the country's point of view. So even when you apply the filter to only show flows which involve Unicef (either as provider or recipient), the flow is still shown as internal. In effect, you have asked to see 'flows from Sudan's point of view which concern Unicef'.
In the second case, however, when you select both 'Sudan' and 'Unicef' as your boundary, you are now drawing a boundary specifically around 'Unicef in Sudan', i.e. the combination of the two entities. From this point of view, the same flow is clearly outgoing, as it represents a funding flow going outwards from Unicef in Sudan to another entity. In effect, you have asked to see 'flows from Unicef in Sudan's point of view'.