Getting it Right Theme

Getting it Right is about helping publishers to publish for the first time. Or about helping publishers to improve the quality of their data. Or about getting better at the GPEDC or Aid Transparency Indexes! What would you like to see under this theme?

I think one thing that would be beneficial for us as a community to discuss is the new publishers process and how can we help to make this as effective and efficient as possible.

The existing process is heavily dependent on the presence and involvement of a few key people within IATI to guide and bring new publishers in. For the maority of organisations, this could probably be handled with clear documentation, access to existing IATI experts (experienced publishers or known specialists within the community) for questions or queries. This would free up time from those within IATI to focus on structural changes and maintenance of the standard and for hand-holding special case publishers who need this additional attention.

We have already started to make this move with the introduction of the IATI Helpzone. But a clear, sensible and well documented publishing process could be a great exercise for a workgroup or team to formulate at the next meetup.

Another one from me today :smile:

Facilitating following the money in IATI

There have been many conversations I’ve taken part in and overheard within the community about the ability (or lack of) to follow the money chain within the set of published data.

I believe that part of this issue is that the process for publishing your funding chain works very effectively if you have direct connected funding within your organsiation, e.g. all incoming funds are earmarked for certain spending, allowing you to identify clear incoming and outgoing transactions as well as have direct relationships with funding and funded activities.

However many organisations seem to operate with a different kind of model. Models including core funding, where incoming funds from a donor is not necessarily targeted at a particular project or programme, but is sent to a central pot, then it is not always so obvious how to track this.

Now if you look at the IATI Standard, there are many options here for working through this within the data, that I think would fit almost any model of organisation.

What I think we need to do as a community is to work out this practically for a number of known working models to provide some publishing advice for those who are trying to do this.

Adrian, I agree that we need to look at different traceability models.

We use one that works for our directly funded organisations (NGOs, private sector partners) and that is baked into the DFID Minimum Requirements. This uses an Incoming Funds transaction that has a provider-org-ref pointing back to the funding project. You can see this in operation on Development Tracker - UK Aid Direct.

We are still challenged by core funding, which in this model would require some form of follow-the-money inside a receiving organisation. That may not always be possible. Our statisticians use “imputed share” for things like multilateral contributions, but that is generally a year or two stale.

Whiteboard/hack session at the #TAG?

Here’s a topic I’d like to recommend for the upcoming TAG meeting:

Auto-geocoding and Geonames

At the 2014 TAG meeting in Montreal I promised to put together something that would extract place names from text and then assign Geonames IDs to those place names, accordingly. We (at the Foundation Center) now have an API and a light web app that does this fairly well, but I want to share it with the IATI TAG to see if it meets other orgs’ purposes. In technical terms, it uses an Alchemy API (we bought a multi-year license), Geonames’ API, and our own internal NetworkX API to help make the best possible choice if given several options (e.g., in cases where the text includes “Springfield”, but there are at least 12 Springfields in the world). We are currently using it for all of our data, but our use cases surely differ from those of other IATI orgs. I’d like to show it for a few minutes, then give folks access to try it out with their data, solicit feedback (online and/or in person) and then see if there are ways we can tweak the algorithm or API to better accommodate everybody’s needs.

I think I’d only need like 15 minutes, and maybe it could be bundled with other geo standards stuff? Any thoughts?


Adrian and John, I agree that the topic of traceability and ‘closing the links’ in the aid chain is definitely one of the important challenges if we want to gain insight in relations between organisations in the development network. At the Netherlands MFA we are asking from all our partner organisations, that they publish in IATI format next year. We are struggling with the same challenges you both mentioned: how to represent core and pooled funding? How to use the existing IATI standard to represent corresponding money flows? How to represent joint funding of alliances of aid-organisations? How to deal with the results data? Etc.

Together with the Cordaid we explored some of these questions. One of the insights gained was that the current IATI standard does provide some possibilities to solve the issue of traceability. There are many different ways to do this though and challenges concerning the use of the IATI standard remain.

One of the key issues is finding the right balance between a very precise and complex solution versus an easy to implement but imprecise solution. We are currently working on a guideline for our partners on how to use the IATI standard implement traceability.

I would be happy to present our thoughts on this subject in the TAG meeting.

I totally agree with you, Adrian, I keep asking for questions for one month and I did not get any clear answer

@ibrahim there is an official new publishers process on the IATI Standard website here:

What would be really useful is to try and find out / track where people most often get stuck.

Thank you very much Adrian for you prompt answer, the link looks promising, I will check it soon.

At Bond, we’ve now worked with over 200 organisations since 2011 to support them to publish, including Southern NGOs, academic organisations and others. The common thing that we find is that reading about IATI scares prospective users, but in practice sharing information via IATI is relatively easy. So there is a perception that because it’s a data standard and uses data language, it’s difficult to do.

What we’ve found is that the standard itself is flexible enough to accommodate most scenarios for NGOs. Tools like AidStream also help greatly. However, it’s really important to have an intermediary explain and support organisations to do this. In the UK, we do this with human interaction (the Bond team), but it doesn’t have to be that way - some people are now finding their way through using the really useful AidStream YouTube videos for example. The right tools for the job definitely make it easier.

Also like Adrian, we find that it is more difficult for larger, federated organisations where money travels internally around the federation, for example from the UK office to a central coordinating office, where it is pooled with other funds and distributed to country offices, where in turn it is disbursed to partners. Or where these organisations have framework agreements which mean that the money is not a straight transfer into a programme or project.

I don’t think there is an easy solution for federated organisations, but I also think IATI may provide an internal reporting framework within large organisations and facilitate the flow of information about funding internally. So these organisations almost have to get their own house in order first before using IATI.

We’ve also found that there almost needs to be a change management strategy in place for larger organisations - IATI is a new way of working that requires resourcing (people/time/IT) up front and then on-going. These resources are not (in the UK and Netherlands experience) funded by donors. Therefore the organisation has to make the decision to invest in IATI over and above other competing demands. To do this, they need to be convinced of the value to the organisation. We’ve also found that creating a sustainable process is also crucial to success. This is not necessarily about starting small - for larger organisations it’s just as easy for them to publish information for 10 or 100 projects as it is for them to publish 1 project. It’s more about embedding IATI within existing systems, sharing the responsibility across different functions and valuing the output, though reuse of their own data.

I’d be really interested in co-presenting a topic with Adrian on this if that’s a possibility.