Once your organisation has committed to IATI publishing and the project team or IATI Champion is in place decisions need to be made about what, how and when data will be published. This information should be initially recorded on the ‘Publishing Information’ tab of the Implementation Schedule and then copied into the relevant fields of the publisher’s Registry account when it is created.
One of the most important decisions to be taken before an organisation starts preparing to publish to IATI is to decide how to present their particular business model in a way in which their work is best translated into an IATI activity. It is up to the Publisher to define their units of aid and they can specify any number of hierarchical levels in which to best organise activities. Their work might, for example, be broken into themes, programmes, projects, sub-projects and/or contracts. It is up to the Publisher to decide whether a hierarchical structure best tells the story of their organisation, or whether such a framework is irrelevant to the data that IATI requires to be reported.
If Units of Aid are defined via a hierarchy, an explanation (to be ultimately documented in IATI Registry account) should be provided on how the units of aid are split between the hierarchy levels . For example, activities at hierarchy level 1 (or the parent level) could cover things like programmes, umbrella projects or grant streams. Units of aid at hierarchy level 2 (or ‘child’ level) could include projects or sub-projects. It is also possible for activities to be related on a horizontal basis (‘siblings’). For example, a group of projects sitting under a programme could be classed as sibling activities. Related activities can be identified through providing the related activity identifiers when reporting on each individual activity.
It is also important to differentiate what data will appear at the different levels. For example, activity budgets could be reported at level 1, while transaction level data is reported at level 2. This also helps to avoid double counting, with monetary amounts being reported at the lowest possible hierarchical level (and not repeated or aggregated at higher levels). Lower level activities will inherit data from their parent activity.
Organisations typically publish updates annually, every 6 months, quarterly, monthly and even daily. Ideally it would be a requirement for every publisher to publish as frequently as their information changes. However, it is recognised that this is not always possible so it encourages publishers to select the most frequent publishing frequently that their organisation is able to support.
Timeliness of data relates to how current the data is when it is published. IATI differs from other reporting standards and systems, by emphasising the importance of publishing timely data, that is reflective of the swift-changing nature of projects and activities. With this comes a tension between provision of timely data and quality data.
If you are publishing quarterly do you include transactions that have occurred right up to the day that you publish your information or are your most recent transaction those that occurred a month or so in the past? Timeliness is normally constrained to the accessibility and functioning of an organization’s own accounting processes and systems.
it is important to consider how accurate your data is when it is published. The data published does not need to be audited in the same way as would be done with close of year accounts. However, ultimately, the decision on data quality will be taken by each individual organisation, depending on the internal procedures and processes that data may have to go through before being put in the public domain. It will also depend on whether an organisation approaches IATI as an open data initiative or a process for publishing verified statistics. Many organisations publish information taken from unaudited accounts throughout the year that is then revised and republished once the accounts have been through the audit process. If data is initially published in an unvalidated or estimated format it is important to consider how and when your data can be recreated and published in its formalised format.
Publishers should also outline the quality assurance processes or any details of their considerations around what they consider to be verified or unverified data, and where their IATI data sits on that scale.
It is very important that an organisation considers which of its activities it will publish information to IATI about. Ideally it would be a requirement that a publisher provides information of all of its activities. However, this sometimes not always possible and initially a Publisher might decide to only publish information in relation to those activities for which they are contractually bound to publish to IATI. However, it is hoped that over time a Publisher will move to publishing to IATI for all of its work.
If applicable, some organisations may need to establish thresholds in terms of the value of activities or transactions that they are publishing. For example, an organisation who works on large projects costing thousands of dollars may wish only to publish projects over a particular value (for example $500).
Similarly, if an organisation generally only makes or receives transactions over a particular value, they may wish to aggregate all transactions, in-coming or out-going that are under a particular value, per organisation, per publication period.
As with exclusions, information on thresholds should be clearly published in order to help users of the data understand the context it is being published in.
IATI enables publishers of information to the Standard to exclude certain information from their data. However IATI requires that the policy of how those exclusions are made is made public, as part of the Registry Record in order to provide context to those who are using the data. Reasons for needing to exclude data depend on each organisation but can vary from security concerns to the need to protect commercial information. If your organisation carries out works which could be considered by states or societies as illegal or unacceptable, for example in the fields of human rights, gender or governance, you may not want to publish all the details of your work in order to both protect those who are doing the work as well as the intervention itself. Similarly if you are working in sensitive geographic areas you may not wish to publish details of your work and the areas in which it is taken place. In such circumstances you can elect to only publish minimal details or even to not publish any details at all.
Considerations when defining an exclusion policy
- Does your organisation/government have any existing Freedom of Information laws or policies that may impact the IATI exclusion policy?
- Are there particular data fields where you would consistently have to remove data?
- Are there entire programmes/activities that would need to be excluded from your IATI data?
Each organisation needs to consider and understand where it sits within the aid delivery supply chain and report accordingly. Basically an organisation should report on activities for which they either receive payments or make payments for or are directly involved in in some other way. This means that if an organisation has an HQ which makes payment to a country office then both the HQ and country office should report separately to IATI
As its name suggests IATI is about Aid Transparency. However, it is not just about publishing data for ODA activities. Information can be published about any activity which falls under any of the aid categories as defined at http://iatistandard.org/codelists/flow_type/ .
Consideration should also be given to which specific aid activities are reported on. Ideally IATI would want details of all of an organisation’s aid activities to be published and not just those that have funding which comes with a contractual requirement to publish.
When you publish your information you do not give your information to IATI to look after. You will instead create an account on the IATI Registry that will hold (along with other information) the URL of where your IATI files are located. IATI therefore asks its publishers to host the datafiles that contain their IATI information on their own web servers. This is because data that is located within a publisher’s own web space has greater integrity due to it being located at source. Therefore you will probably need to work with your IT or web team in order to identify a location where your IATI information can be hosted. If you have decided to create an IATI specific web page it is a good idea to include links to your published IATI files?
The success of IATI is dependent on intermediaries making information accessible for different stakeholders by taking the information published for the IATI information providers and using it to produce databases, tools, applications, and analysis for newspaper articles or radio programmes, etc. It is envisaged that these intermediaries would range from partner country governments, CSOs in the north and south, community groups, parliamentarians, journalists, individuals, researchers.
To enable this ecosystem of different ways of accessing information, intermediaries must be given legal permission to use it. This can be done on a case-by-case basis, where each time someone wants to use the information they must ask for a license from the copyright owner, or a license can be issued up front that states the terms for using the information. For practical reasons, as well as to encourage the use of the information, it has been agreed that IATI will seek to proactively license the information. In addition, in the case of IATI, where information is provided by multiple sources it is particularly important that the licenses used are compatible with each other.
In parallel with the IATI process, there is a movement towards open government and open data movements within international organisations. There are strong links between the goals of these initiatives and IATI, and the principles of open data (as defined by the open knowledge definition) are a key element for all of them. For data and databases, to be “open” means that users have the ability to:
- combine (mash up) datasets from different providers;
- add additional data and select which data records to include or exclude in derived works;
- change the organisation of the data (its schema) and change the database to a different format;
- copy and distribute the information.
The TAG Secretariat, in collaboration with a small group chaired by the World Bank and including an open data/intellectual property lawyer, produced a set of recommendations for licensing. The Open Aid Information Licensing Standard, agreed in February 2011 at a meeting of signatories and the IATI Steering Committee, is that information published through the IATI standard should be licensed under an open license. It is a set of principles that must be adhered to, rather than a prescriptive set of terms and conditions.