This consultation closed on March 1st, 2019 and the discussions are no longer accepting comments. If you would like to contact IATI, please email

Many thanks to all those who took the time to comment on the discussions. Summaries of the discussions below will be available below on Friday 8th of March, 2019. We thank all our Moderators for facilitating the discussion and synthesising the outcomes.

From the 4th to 18th of March, IATI will hold a survey on data use as the next phase of the consultation to develop the IATI Strategic Plan. To learn more about IATI, visit

With IATI recently celebrating its ten-year anniversary, this online consultation is an opportunity to spark dialogue around essential priorities for the initiative’s next three years. Anticipating the next generation of partnership and data needs, your responses will help IATI to ensure it responds to the rapidly-evolving development finance, open data and transparency agendas, to increase the use of development cooperation data. 

You may wish to read through two background documents prepared for this consultation (an external and internal scanning paper). These papers examine the current international cooperation and open data landscape, as well as IATI’s progress and achievements since its inception in 2008, and may be useful tools to inform your participation in the consultation.

Please feel free to comment in as many threads, and respond to as many or as few questions, as you like. Though the consultation will largely be hosted in English, comments in French and Spanish are also welcomed. You may also submit contributions to to be posted on your behalf, should you encounter any connectivity issues.


1. How could information reported through IATI respond better to partner country needs for relevant and easily-useable data that can be leveraged for national development planning processes, and foster development partner accountability at the same time? 

  • Beyond what was accomplished through the 2016-2018 Strategic Direction, what are the priority areas in which IATI can further improve the quality, timeliness and usability of data (for example, placing greater emphasis on encouraging publishers to use the added value fields such as geolocation, results, linked transparency, humanitarian and activity documents)?
  • How are users of data in partner countries (different government partners, civil society, journalists and activists) engaging with existing and emerging development cooperation providers? How could IATI support the harnessing and use of data on this cooperation at the national level?

2. Given the evolving development cooperation landscape, what new directions do you foresee for the IATI Standard over the next 3-5 years?

  • Which new or existing data fields, additional datasets, and business processes, would you prioritise to ensure interoperability of IATI data with fiscal and statistical country systems and / or alignment with development planning and programming processes?
  • Should IATI, as a standalone standard, respond to technological developments that will enable connectivity between data sources? If yes, how?
4 Feb 2019 - 28 Feb 2019


Tim Davies Moderator

Considering Strategic Directions for Development Cooperation Data and Standards

The forum on “Considering Strategic Directions for Development Cooperation Data and Standards” generated wide-ranging discussions. From a grounded analysis, five general themes have been identified, and are described below, relating to (1) maintaining focus; (2) supporting use of data; (3) standard development and simplification; (4) standard documentation; (5) positioning in relation to other standards.

(1) Staying the course: delivering on original commitments

Over the last decade IATI has substantially expanded the quantity and detail of available machine-readable data on the volume, allocation and results of development expenditure. Yet there is a widespread perception that IATI is still not delivering all the information that aid recipient governments need, nor is the data being adequately used to support development co-operation and co-ordination.

There is a strong view that IATI should not lose sight of the original Accra commitments, and the primacy of partner country governments as data users. This needs action on three fronts:

  1. Ensuring the quality and relevance of structured data from the largest donor organisations (where there are often still gaps). Some suggested de-prioritising smaller donors; others suggested distinguishing data from large and small donors in terms of what is important, and moving away from the metric of overall number of publishers.

  2. Increasing the reliability and usability of services supporting IATI data use. This requires both financial backing and staff resource for those tools, and more focussed user-centred design. D-Portal and the Data Store are seen as core infrastructure.

  3. Revisiting partner country needs, to make sure prioritisation is guided by current evidenced user needs, rather than external ideas of what is needed, or continued efforts to pilots of early hypothesis and ideas about what is needed. Suggestions to focus on Ministries of Finance, and on donor country office awareness of IATI, to improve country engagement with IATI data were both put forward.

Sticking to the original commitments does not mean ignoring the changing landscape of aid. There was support for a broad view of aid and development cooperation, taking into account humanitarian, climate finance and south-south co-operation. Extending to cover flows which look very different (Foreign Direct Investment, Remittances) did not receive support, although important questions were raised about the need for IATI to critically examine its support for describing loans, and supporting debt management. Understanding how IATI fits into, and aligns with, ‘Total Official Support for Sustainable Development’ (TOSSD) measurement was raised by one contributor as a crucial area for the future.

(2) Taking IATI to the people

The design of IATI was based on a ‘publish once, use everywhere’ idea. This remains relevant, with contributors describing a range of important uses of IATI, from re-packaging data for humanitarian actors and placing it alongside their existing operational data sources, through to importing data into AIMS systems, or using it in innovative third-party tools like AidHedge. However, the idea of ‘publish and they will come’ has proven false. Getting IATI data into use at country level, or taken up by third-party developers, requires active outreach and engagement, and sustained funding beyond one-off money for pilots. Whilst activity might be carried out by partners, IATI should be more active in fostering and supporting the partnerships that can get data to where it is needed, in the format and level of detail it is needed in.

Contributors described the need to raise awareness of IATI data, and to scale up capacity building: not just through expensive in-person trainings (which, when used, should be delivered by suitably skilled trainers), but also through online resources such as videos and high quality documentation. Capacity building should respond to user needs: including by teaching about IATI alongside other data sources, and working with Universities, civil society and other institutions. Supporting publishers to create awareness of IATI in their own country offices, and engaging country offices more with improving IATI data was also described as a key way to build trust in IATI data.

For country use, IATI data also needs to be available in local languages. It is not currently clear whether or not machine-translation can meet this need. One respondent also noted the importance of creating feedback loops, so that grass-roots civil society groups can engage with IATI and feedback information on projects, rather than just having ‘read-only’ access to data.

Across a number of comments, the importance of better understanding user need was underscored. IATI could strengthen its capacity to follow through on user-centred design approaches.

(3) Keeping it simple: fitting IATI to different communities of practice

There was a vibrant conversation on the question of important data fields, who they are important for, and who is best placed to provide the information.

By recognising that there are multiple use-cases for IATI, and multiple kinds of publishers (both by constituency, size and position in the aid distribution chain), it may be possible to describe a minimal core of IATI data, and then a range of different sets of fields required by different communities of practice. Testing whether fields are widely used could lead to more fields being removed from the standard, and to publishers being able to identify the fields they should most focus on.

However, views were mixed on whether IATI can realistically move towards a simpler core standard: some saw the 10-year anniversary as the ideal opportunity for a major refactor/rewrite. Others suggested the current governance process tends to prioritise growth, but not rationalisation of the standard. One institutional donor noted that frequent changes to the standard are problematic for both big and small organisations.

Discussions of geo-data, and an on/off budget field, illustrated the importance of working out who should provide certain data, and of investing in documentation of data fields. For example, one comment suggested local NGOs in the aid delivery chain are much better placed than large bilateral donors to produce geolocation data, whilst they may have no information on whether financial flows should be counted as ODA or not. Supporting local co-ordination may not require financial information, but may need an emphasise fields on that describe the who, what and why of development assistance in a particular country.

Improving the simplicity of the standard for publishers and users may require the core IATI community to spend more time on standard development and governance, and to explore new technical approaches and architectures. One comment suggested that IATI could support ‘confirmation’ of data between delivery chain actors, noting that “when data is mutually confirmed, by the immediate partners, we have no reason to question the quality.”

(4) Better documentation can speak volumes

A number of responses pointed to the importance of IATI Standard Documentation, and the need to invest more time in making sure documentation is complete, comprehensive, bug-free, and accessible.

In a number cases the meaning of an IATI data field can only be understood when their is an accompanying documented business logic. However, at times, guidance or documentation promised in the upgrade process has not been delivered. Documenting best practices, based on learning from existing publishers, could add substantial value.

Documentation is not just about technical definitions: it is also a way of helping different practitioners speak using ‘common language’, recognising that when we bring together country partners, technical experts, government donors and NGOs, we need to make sure we have shared understanding of key concepts.

It may be useful to explore new methods, including face-to-face workshop sessions, and task-and-finish working groups, in order to accelerate the completion of the guidance that is needed. Good documentation is more than text, and involves examples, diagrams and learning materials.

(5) Joined up standards

There were relatively few contributions relating to where IATI sits alongside other standards, although one response noted the importance of positioning IATI as an open data initiative, and avoiding an 'IATI silo’. The need for continued alignment with statistical standards (DAC/TOSSD) was raised, and one comment noted the value of having comprehensive mapping to partner country data on budgets, expenditure and reporting. In considering the Humanitarian sector, an understanding of the existing standards that exist can help IATI add value rather than duplicate existing work.



Tim Davies Moderator

Welcome to this IATI Strategic Plan consultation. I'll be your moderator for this section of the discussion on "Development Cooperation Data and Standards".

I'm here to help the discussions flow, and I'm happy to answer any questions of clarification too. I really look forward to hearing your experiences, ideas and reflection.

The questions above are quite detailed, but in essence we are being asked to consider:

  • Strategies for improving data quality

  • Strategies for improving data use


  • New (strategic) directions for data and standards

You might like to respond to these themes by sharing a personal experience or insight from your work with IATI. You may have comments on the ideas in the scanning the Horizon document (the last few pages are the most relevant to this thread of discussion). Or you might like to reflect on how you would re-design IATI strategy and technology if you were starting from scratch in 2019 to help us think more broadly.

Whatever your input, I hope that with everyone's contributions we'll have a really good constructive and wide-ranging discussion over the next few weeks. 

David Megginson

I'll just mention that UNOCHA's Centre for Humanitarian Data is piloting a collaboration with the IATI technical team to make IATI data available to humanitarian actors side-by-side with other humanitarian data (eg needs assessments, damage reports, facilities lists, etc) on the UN's Humanitarian Data Exchange (HDX). This kind of collaboration is important, because aid actors don't generally know to go looking for something like IATI data in a different place, so we want someone working in (e.g.) South Sudan to see that country's IATI data beside all the other operational and monitoring data.

There are no doubt many other similar situations where we could put IATI data in front of people more readily (data department stores, if you will) rather than forcing them to make a separate side trip just for IATI data. No matter how well-designed future versions of the IATI Datastore and D-Portal may be, a lot of aid-data stakeholders will never go there. Being able to spread that data around to places the users actually are that is a big benefit of our using open licenses.

Tim Davies Moderator

Thanks David. This is a really good example of the 'publish once, use everywhere' idea that was at the heart of early technical discussions on IATI. 

I'm wondering though (and this is a question for anyone reading) if there is anything we need to particularly consider for the future strategy to move beyond pilots towards long-term flows of data from IATI into these other spaces? 

My understanding is that the strategy to date has (implicitly/explicitly) been to assume that third-party 'intermediaries' will play this role of taking IATI data and putting in front of more people: but in ten years it's not clear how often that has happened, or has happened at scale. 

David Megginson

Tim Davies You're right that it doesn't seem to happen on its own -- we probably need to reach out and manage explicit partnerships for the high-value places, at least (they'll want some kind of agreement on terms of service/data quality, etc).

This sounds like the right kind of "strategic partnerships" described in the Outreach stream. We need to dedicate resources to identify opportunities for such partnership and develop the required tools/guidance/whatever is needed.

Poncelet Ileleji

Looking at strategies to improve data quality personally talking from the perspective of someone working and living in a developing country like the Gambia, wer have to look at grass root engagement especially aspects of citizen generated data, its a good to start to me.

Tim Davies Moderator

Thanks Poncelet. 

When you suggested 'aspects of citizen generated data', do you mean that IATI might need to do more to have a feedback loop - so that donors engaged with IATI think of this not just as 'publishing', but more as a two-way channel of communication about their projects.

Or this more a case of delivering capacity building that combines IATI data with citizen-generated data at the local level - such as when a local project might combine crowdsourced data on the quality of school buildings, with IATI data on the funding for them.

Poncelet Ileleji

Tim Davies:  Your first opinion is what I was thinking of so we have a feed back loop.

Saripalli Suryanarayana

I am sure this is a real way to understand the developmental perspectives in terms of SDGs.One aspect no one will be willing to discuss and come out is the need to adopt a designer web and platform for the future vehicles for public use.This and as well the issues concerning the Data on the vehicular pollution,and thermal energy use might be beyond these discussions.

How ever we all agree to discuss the immediate needs of water,sanitation,and perspective growth of cities.They can be smart/or even safe but pollution is the problem.Innovation in to renewables have solved some problems.How ever carbon emissions,from socially imposed finacial burdens[financed by various agencies] out of the thermal energy still are with the people of each country,be it developed or devloping.

The problem of politics over policy persistently lie to show a developed society in to developing and showing many factors needing international intervention.The idea in presenting such data and documents lie solely for quid pro quo[money laundering].

Such projects unverified leaves behind,a wide gap in conception and achivement.Data is key to find real communities,with in each society,and finding a national way is most important.Where national productivity is some where near the requirements,then consutancy or manpower [by IATI] shall only be addressed,so than the distribution channels are streamlined.

Heritage and culture  in most asian counties or societies as a sytem needed to be addressed in meeting the disasters arising out of over exposure to sun,and contracting new infections.Hnad wash,use of safe cloth for protection,menstrual hygiene and so on so fourth.

Tim Davies Moderator

Thank you Saripalli.

Am I right in summarising what you are saying like this?:

  • The focus in IATI appears to be mainly on the finance/money side of development;

  • But there are many other things that need to be understood, including environmental impacts of development projects, and cultural factors that impact on health and wellbeing;

  • IATI could be used to more fully understand development in line with the Sustainable Development Goals (SDGs) - but this would require more careful design of the platform to include these wider issues;

This seems to me to link well to the idea that (a) the IATI Standard might need to develop to focus more on the 'what?', the 'why?', and the 'what happened?' of development projects; and (b) that stronger links will need to be made with other data standards and infrastructures that can describe other aspects of development. 

Would that be a fair interpretation?

Saripalli Suryanarayana

Yes you are right,experts to get data on changes in environment,finacial dealing,food production,habits in living are more a need of many countries.

The governing policies,market approach,storage of food production,migration trends.Jobs developed and fullfilled at local leves are key to forming an assistance policy.

Thank you for understanding the process of developmental mechanism needs of modern,changing technological systems.These systems needs to be adopted at low cost with viable human intelligence.

Spend properly,wisely.

David Megginson

Hi, Saripalli. Your points are excellent, and I think it comes down to deciding where IATI's boundaries are. In the humanitarian-aid universe, we have lots of other coordinated international data initiatives, eg

- Humanitarian Needs Overviews (HNO) for multi-sector needs assessments in select crisis countries

- Displacement Tracking Matrix (DTM) for displaced populations and their needs/services

- Integrated Food Security Phase Classification (IPC) for food security

- Standardized Monitoring and Assessment of Relief and Transitions (SMART) for malnutrition and mortality

etc etc. We appreciate IATI for financial transparency and activity reporting, but I don't think it would make sense to try to expand it to cover all aspects of international aid, when there are already well-established sister initiatives in place for those. UNOCHA's Humanitarian Data Exchange (HDX) collects many of these data types, and is a good starting point if you're searching for a broader perspective than you can get through IATI alone (disclaimer: I work with HDX).

Cheers, David

Joshua Powell

Most of these ideas are rehashed from prior work/studies on IATI (by Development Gateway, PWYF, Oxfam, and Mark Brough), but the key priorities from my perspective (thinking primarily of partner country government use) are:
1. Publication in partner country language. Many of our AMP partners are in Francophone Africa, and the burden of use for English-only publications is particularly high when trying to match project titles, understand descriptions, etc.

2. Ensuring that the DataStore (and related APIs) are up and working (and that APIs include all key fields). Much of the work of IATI-AIMS integration, after initial configuration, can be automated (or at least partially so) if this is in good shape.

3. Helping publishers create more awareness of IATI data within their own country offices. In order to develop better data quality (through country office validation) and facilitate better trust/cooperation between AIMS units and country office staff.

4. Value add fields (particularly on geography) and indication of on/off budget status. There have been several conversations on whether an on/off budget field is needed (I think its absence is a significant barrier to partner country use), but at the least the channel of delivery and related fields should be published.

5. Attracting more funding to support country-level use. At the moment, most of the funding that exists is for pilots or proof of concept work. The bulk of investment has been on publication and tools, with only more recent forays into funding for use. The Data Use Fund is a good step, but the scale of funding is quite small to achieve lasting change/uptake. I think the Bangladesh, Madagascar, and other experiences show that having more sustained funding/programming is critical to achieve everything needed (technical, skills, process design, change management, etc.). 

Tim Davies Moderator

Thanks Josh: these are great points. 

On partner country language: I was struck when this consultation was set-up, that the team from GlobalDev Hub said they've found over the years that the machine translation on this platform does a pretty good job and converting discussions, and allowing cross-language dialogue.

Do you know if there have been any experiments with machine-translation of IATI data, and how far this is useable or not?

I'm curious whether the potential costs of adding some translation later into IATI have been quantified. 

Taryn Davis

Just emphasizing Josh's point #3. Getting organization country offices to understand and accept their IATI data is one of the largest barriers to its use in country government systems. There needs to be a reconciliation of data at this level in order for it to be usable and trusted at the country level.

On #3, bringing the perspective of someone at HQ: I agree that we need to develop better data quality and trust. But it is very difficult (read: impossible) to create awareness of IATI data in country offices if government officials themselves do not at least signal an interest to use IATI data. We really need partner country officials to engage more on this, to create the "pull" for donor country offices to take an interest in IATI data.

Another thing that would really help would be to have a killer d-portal that country offices could use for day-to-day work. 

On #4, fully agree on geography, but do want to reflect here the gist of the Twitter discussion since not everyone will go. An on/off budget field as has been described in some reports is pointless as only partner countries have the required information. What we need is better data on implementing organization type/channel of delivery.

#1, 2 and 5: yes, absolutely!

David Megginson

Excellent points, all. Donor data is generally too high level to be useful in-country, and local implementing partners (who know the most detail) usually lack the time, capacity, and/or incentives to do IATI reporting, so on the development-aid side, the sweet spot would be the programming organisations—the iNGOs and multilaterals who take the big grants from donors and break them down into smaller projects and programmes. Their activity data (unlike the donors') could be genuinely useful at the country level, especially if they got better at localisation. DFID  the Netherlands MFA have both focussed on getting their immediate grant recipients to report, but we have to make sure the reporting guidelines consider country-level (operational) requirements as we as donor-level (results/transparency) ones.

Tim Davies Moderator

Thanks to everyone who has been taking part so far in the discussion. If you’ve been meaning to post something, please do try and do that this week: then we’ll have time to explore the links between different contributions in the final week of consultation. I’ll be writing up a synthesis of all the comments here to feed into the strategy. 

Below you can find a quick summary of some of the responses so far. As a reminder, we’re looking for any of your thoughts that can inform the IATI strategy relating to data quality, data use, and the future of data and standards. 

How could information reported through IATI better respond to country needs?

  • David Megginson pointed to to collaborations that are placing IATI data side-by-side with other humanitarian data: taking IATI data (or extracts of it) into other portals and platforms. 

  • Saripalli Suryanarayana suggested a need to focus more on the ‘what?’, ‘why?’ and ‘what happened?’ questions for projects, rather than the current focus on ‘where?’, ‘how much?’ and ‘who was involved?’ and ‘how is the money classified?’

  • Josh Powell put forward the need for more localisation into partner country languages, and the need to raise awareness of IATI at partner country level. The importance of stability of APIs and DataStore was also noted.

  • Josh Powell also brought attention to the importance of ‘on/off budget status’ fields, noting the absence of this information is a significant barrier to country use of IATI data 

What new directions do you forsee for the IATI Standard over the next 3 - 5 years?

  • Over on twitter, Harsh Desai argued for the importance of geographic fields, explaining “Location is one of most under-populated #IATI fields”, but that there needs to be better articulation to donors of the value proposition for providing geodata. Josh Powell also noted the importance of geodata. 

  • Poncelet Ileleji called for more grass roots engagement with data, creating feedback loops so that IATI can be used for two-way communication, not only data publishing.

  • The need for more funding to support county level use, going being small scale pilot funding was noted by Josh Powell.

Steven Flower

Thanks so much for these thoughts and discussions

I've two points to share.

1. Regarding the artefact that is the IATI data standard,and ideas around what and how it may evolve, I'd like to propose a trend to consider: reduction!

Over the first years of the IATI standard we've seen several changes, which have generally seen the standard grow in terms of scope and focus (alongside some important bug fixes we've made, based on user feedback).  We've now got more possible vocabularies to consider, and a multi-dimension of results data, for example.

To counter this, we quite often hear feedback that IATI is "difficult" to understand or use, coupled with a need for simplicity.

Now, I'm not suggesting that we reduce IATI down to simplify things, but I do think people would welcome an evaluation of the various facets and functions of the standard, with a view to rationalising.  We've ten years of data being produced with the data standard, so a wide range of use cases to draw from.  During research, I think we'd find parts of the standard that are inefficient, or "not as intended".  Our temptation may be to close loops through more complication of data, rules and conditions of use - but I know a number of us in the community believe that a reduction of the standard might be our best way forward.

I won't now detail the contradictions of particiapting-org vs transaction/provider-org (for example!) but just wanted to flag that we might want to welcome the standard being a lighter and more specific resource, rather than something that could grow to be many things.  Indeed, if the standard were to concentrate on some core exchange concepts, these might be adaptable to others sectors very easily.

2. The second point I'd also like to make here is more of a request in terms of how we measure the success of the standard.  I'd like to ask: when do we stop counting (the numbers of publishers, or activities or volume of transactions)?  

There's a common thread around IATI that 900+ publishers / 1 million activities / BILLIONS of $ of transactions is something we should measure, and continue to use as a metric.  Whilst I agree this is something that has been helpful during our growth, this kind of data starts to creak in terms of meaning and legitimacy.  We all know that there aren't really 900+ active and engaged publishers, so why the need to keep a log?

To my mind, we should positively welcome the point where we cannot confidently count (or double count!) IATI data because we realise this is not a measurement to drive us forward.  To this end, I would welcome a vision for IATI where the point of entrance for a new publishers does not signal an addition (to turn the clock forward another turn) to the initiative, but a new point on the network.  It's down to us to engage this node, not just count it.   

Saripalli Suryanarayana

Yes exactly inter wooven sytems of heriditary and culture,and local systems may ask other systems to be addressed first.Like a house at ground level may not be safe enough unless a cot is given.Masquito bites can not be avoided unless treated nets are made vailable.People are ineffective unless sanitation is effectively addressed,So the point where and how to start to have safe and developed life is important.

Herman van Loon

I like and support Steven’s line of thought: less is more in terms of quality and usability. A less complex standard will it make it more straightforward to publish and use the data and develop tools. There are a number of aspects to consider here:

1 – More focus on those parts of the standards which are relevant for many users

2 – Taking into account the role in the network of a publisher

3 – Focusing more on IATI as an opportunity for improving cooperation instead of  transparency and compliance

Good quality data  is i.m.o. not about publishing as many as IATI elements as possible. Looking at the actual publication of IATI data elements in the dashboards shows that there is a very large ‘tail’ of data-elements published by only a very limited number of publishers. This begs the question if we are really improving the usability of the standard by accommodating every specific publishers wish or that we should focus more on the standard as a common language (meaning focus on those parts of the standard which are relevant for many users to exchange meaningful information)?

It is imo also important to consider a publishers role in the delivery chain. An example: a local NGO is usually very well equipped to produce high quality geolocation data since it knows exactly where the actual work is being done. This in contrast with large bilateral donors who often work through intermediates (multilaterals, iNGO's, etc.) and therefore have no information where an activity exactly is taking place. On the other hand, a large bilateral donors exactly know if an financial flow is ODA or not. This kind of data is totally irrelevant to publish for an NGO.

Currently there is a large number of IATI publisher who only publish the minimal number of activities necessary to get funding from their donors. Their data is incomplete and therefor potential opportunities are lost (e.g. for coordination of activities between organizations). As an IATI community, we could encourage those publishers to publish and use their whole project portfolio (there are good examples of organizations who do this and who benefit from their effort).

Tim Davies Moderator

Thanks Herman. I'm reminded in this discussion of some of the advice we had from Jeni Tennison when scoping the design of the Open Contracting Data Standard. The paraphrase that for IATI, it would mean:

  1. IATI should identify and define communities of practice which have different analytical goals or purposes to which the data will be put. There should be a small number of these (around five) to begin with.
  2. Each community of practice should identify the metadata that is required to enable their particular purpose.
  3. IATI should identify a minimal core set of metadata that is required by all communities of practice; it's likely that this is a very limited set, comprising just an few fields.
  4. Publishers of IATI data should declare (in a machine-readable way, but also through iconography on their websites) which purposes they aim to satisfy through the data that they publish.
  5. Developers of tools that work with IATI data should declare (in a machine-readable way but also through iconography on their websites) their purpose(s), ie which sets of metadata are required for the tool to work. This makes it clear to users of the tools which data can be analysed with it, and to publishers which metadata needs to be published so that they can use the tool.

I don't think OCDS has actually achieved this in practice: although personally I think it remains a really interesting approach. It would allow 'sub-communities' in IATI to emerge, focussed on the different overlapping use-cases for the data.  

Could we take IATI in this direction? 

Herman van Loon

Thanks Tim for pointing this out. Yes, I think it might be worthwhile to explore this line of thought in greater detail. The initial focus might be what data could be expected to be published by each constituency (local NGO's, private sector, multilateral, government, bilateral donor).

Maybe there should also be a separate  focus on what is the minimal core of IATI elements to be published by every publisher.

Númi Östlund

Great input so far!

We have a very limited perspective on IATI, but I'll share some of our thoughts and comments. Hopefully, it can provide some food for thought especially when it comes to usability. 

As background, we connect to IATI and use data to make a financial analysis for activities. We don't follow the working groups, we are not really part of the active community. We just use the data. 

When we developed our IATI-module, we looked around to find good examples of how others had done, and what the lessons learned was. We were especially interested in the different kinds of use-cases IATI had. What had people done with the data? Result: We just couldn't find all that much (please note that I not saying it doesn't exist, we just couldn't find it.).

We turned to the IATI community, and found great discussion and momentum. But we perceived it as focused on reporting, data quality and related issues. How we read things is really well summed up by Steven above: The focus seems to us to be about counting (publishers, activities, amounts, etc.). 

OK, enough nagging. What we found was: a lot of great data, but really no usage momentum.

Our suggestion: If you like counting, change the perspective to usage. How many use cases should IATI have? That's a real indicator for impact.

And how about adding a perspective on different use cases. There are so many potential uses of this data! Make sure its not "just" dashboards presenting the data, like d-portal or They are great, but what else do we have?

One this is for sure, the IATI community (I'm including us here!) are not going to come up with new use-cases, that not how it works. Usage of data is pushed by innovation. You need to get the data to potential users, and see what they make of it (and we might be interested to find who those potential users are!). I'd love to see innovation competitions for example, with cash prizes for the best use cases. With all the tech development and innovation, this unused data is such a goldmine! Get all the ML/AI people in the room. 

And hopefully, once we get usage, we have a strong demand-side that can help us promote publishing and also help us prioritise work/updates on that side of things.

Tim Davies Moderator

Thanks Númi,

I really appreciate this input: and must say that I was so happy to see AidHedge when it emerged, as for me it was exactly a marker of IATI maturing that this seemed to come from outside the usual suspects in the community. 

You are spot on about lack of discoverable use-cases. There have been various attempts over the years to curate tools (AidInfoLabs; IATI Wiki; OpenDevToolkit; IATI Tools Review), but for some reason none appear to have been sustained. Definitely a need for reflection on this and working out why. And possibly some quick wins here to make more of the work that has been done - to better showcase what is working with IATI data. 

I also really like your point on where innovation comes from. Out of curiosity, were there any competitions / bootcamps / mentoring packages etc. that helped AidHedge develop (even if not IATI linked?)? 

Númi Östlund

Tim Davies 

Thanks Tim! 

Well yes, helping potential users find inspiration and lessons-learned is, of course, good (we wrote a series of blog posts on our experiences using IATI as well). That will help those with an interest in using the data. 

What would be really interesting would be more open activities to really push for new data usage. I'm thinking of things like open hackathons or competitions. There are all these hubs popping up around the world where "impact start-ups" are developing. Look at Norrsken House, as one example. Their catchphrase is "Using technology to optimise the world for people and planet", and they are 350 entrepreneurs sitting there. There are similar places all over the place, also in partner countries!. And I wonder who many of these entrepreneurs have ever heard of IATI. Just imagine a series of hackathons or competitions, trying to harness that!

We got innovation funding under a program for "social entrepreneurs" from the Swedish innovation agency. They are also working a lot to promote open data use, in various sectors. And there are many similar programmes internationally, that could also be used to highlight the potenial of IATI data. 

Aria Grabowski

1. How could information reported through IATI respond better to partner country needs for relevant and easily-useable data that can be leveraged for national development planning processes, and foster development partner accountability at the same time? 

  • Beyond what was accomplished through the 2016-2018 Strategic Direction, what are the priority areas in which IATI can further improve the quality, timeliness and usability of data (for example, placing greater emphasis on encouraging publishers to use the added value fields such as geolocation, results, linked transparency, humanitarian and activity documents)?

First of all usable data is massively connected to tools and d-portal just doesn't have the financial backing and staffing to really be a robust easy to use tool.  Addressing that can't just be about having data experts figure out what they want this needs to have significant user testing and feedback in a coordinated and strategic way to really drive the functionality of a new and improved d-portal.  

In terms of usefulness of data placing an emphasis on activities/services provided/results is huge, project details are what everyone needs, and pushing all publishers to place documents when the coding doesn't work and isn't feasible yet is a great first step. Ideally the activities should be linked to sub-national locations. These documents HAVE to be clearly labeled, if there are four documents with the same name can we add dates or which language its in so its easy to click on the information I want as the user.  Some research we are in early phases on may suggest that having gender analysis included may be helpful.  Random links to home pages or generic donor websites that have nothing to do with the project should cease to exist on the related documents list. How can we limit crappy data input by publishers, I think this would really help with the usefulness of information, I don't have a suggestion for how exactly to do that besides some of the tools that others have worked on that flag issues or additionally information in documents that isn't included in coding, but that has its own set of challenges because they people entering data are frequently not the people that know the project details to know if the algorithm has gotten right or wrong.  It would be really helpful to have filters to search projects by whether of not they have certain fields so for example I want a health sector project from 2018 in Lesotho, with a gender equality policy marker, but I need to know the results so I want either project documents of results data and if the file doesn't have it doesn't show up on my search.  I also want everyone to start reporting on policy markers, being able to identify cross cutting issues is huge, and tying that to detailed project documents is a big value add of IATI that nobody else has.  

  • How are users of data in partner countries (different government partners, civil society, journalists and activists) engaging with existing and emerging development cooperation providers? How could IATI support the harnessing and use of data on this cooperation at the national level?

I think there are three huge issues here.  One a lot of people don't know what IATI can do for them, we need to highlight this, not just that it exists, but what they should care and how they can use it for their goals.  Two IATI data is quirky, not easily self explanatory, and takes an while to figure out how to use it to your advantage, training for this is a time intensive and hard to do activity, but something we should invest in.  Training should be designed so that people walk away saying okay this just be came a tool I can use now, web based videos can supplement this or help people that are more familiar learn some tricks, but cannot be the only training.  Also running training is a skill, it is not something that anyone that understands IATI can do, making sure you have the right trainers is critical.  Three there needs to be an easy to use tool (d-portal version 2) that really is easy to use and designed for this set of users with their input, feedback and user testing driving design.  I imagine if we get it right for this set of users, it will work for all the donor country users as well, and if donors need some specific function for them specifically, they have the resources to create their own tools.  

2. Given the evolving development cooperation landscape, what new directions do you foresee for the IATI Standard over the next 3-5 years?

  • Which new or existing data fields, additional datasets, and business processes, would you prioritise to ensure interoperability of IATI data with fiscal and statistical country systems and / or alignment with development planning and programming processes?
  • Should IATI, as a standalone standard, respond to technological developments that will enable connectivity between data sources? If yes, how?
Reid Porter

I came here to share a few thoughts, and (happily) found many of them stated or alluded to by others, but I might adjust the framing a bit.

I feel like we shifted our priority from data quality to data use a bit abruptly, as if the approach that we've used/are using for quality (advocacy, tooling, etc.) would work for use. But I tend to think of data use as the outcome; so what are the inputs and outputs needed to incentivize the behavior change we're seeking?

  • To David's, Josh's, and Aria's point, we need to serve up data reliably and make it available in as many places where potential users congregate as possible. The work done on FTS and HDX is a great step for the humanitarian community. What are the other platforms out there, what's the priority order, and whose doors do we need to start knocking on? Luckily, this isn't something that needs to go through the Secretariat and add to their workload, though air cover is always appreciated.
  • To Steven's and Herman's points, culling the standard (or shifting to an OCDS-esque extension-rich model) would yield numerous benefits, but how realistic is it that we'll get there? As Andy Lulham wisely pointed out in Kathmandu, the current upgrade process has an unintended consequence of ensuring an ever-increasing size of the standard; in other words, we can add new elements as decimal/minor upgrades, but removing something, causing a breaking change, and generally moving someone's cheese requires a integer/major upgrade, and more generally a higher level of effort. Our velocity to date does not suggest that radical change is likely, and the level of knowledge that's needed to understand the various proposals floating out there in the ether prohibits a large, diverse institution like the TAG from ever producing a groundswell of support. On the other hand, and more positively, Steven floated about some new ideas in Kathmandu, like a smaller, more technical group of 'experts' (though he didn't like the term 'experts') that could recommend such a process to the larger TAG for their input/approval. Also, as Herb Caudill recently pointed out, 10 year anniversaries make for great opportunities to do something drastic with tech platforms, though of course there are risks involved.
  • We've talked a lot about publishing carrots (IATI dashboard used to 'name', opinionated review suggestions in publishing/validation tools) and publishing sticks (donor requirements, IATI dashboard used to 'shame'), but what about data use carrots and sticks? If that phrase seems off-putting to you, consider the curious case of GIS/geographic data. That tech has been available and in use for decades as a city planning and urban development tool, and now of course everyone makes SOME effort to use geo-enabled tools, visualizations, and analytic products. But what happened in the middle? There was never a hard and fast requirement that implementing parnters had to deliver a dynamic web map or buy an ESRI product, but there were several soft and subtle incentives. For instance, donors (at the very least, USAID) started including a requirement that proposals discuss how the project would use existing geographic data, and what analysis they would do to inform their activities. Eventually orgs started putting GIS savvy in job descriptions, and eventually GIS people multiplied and started bulking up their organization's approach to GIS data. How could we similarly incentivize development practicioners to use IATI data as a routine practice? How could we introduce positive organizational peer pressure to get orgs to do more without feeling told-to?
  • To Steven's and Numi's points, and to wrap this up, if we think about data use as an outcome, it forces us to think about a different set of metrics focused on users that we don't know personally discovering IATI data and using it as part of their work. How many use cases, how many downloads, etc. all get us closer to the app-equivalent metric of "daily active users," but how would we know that IATI data is making IATI users successful? FWIW, I really like Numi's idea of competitions and innovation more broadly, especially when coupled with Jeni's idea of incubating communities of practice that could tell us what they needed, rather than us trying to guess.

Not sure if that gets us anywhere, but there's my two cents.

theo van de sande

Don't know whether it helps to raise your two cents Reid, but it might be helpful to make a distinction between the potential (breadth) of the standard, vs. the actual use of the standard by publishers. Use cases might/should serve as the prime indication for publishers to make use of the standard. Instead of sanitising the standard, I would propose to develop a stronger relation between publication and use of data. We should, from that perspective, also consider raising the number of mandatory elements for publication with a harmonised, standardised way of reporting on them given their strategic value. These mandatory elements should be determined from a demand/use perspective, NOT from a supply/publishing perspective. Of course, publishers should always feel free and be encouraged to publish what they want, from their perspective, encouraged and supported therefore by a broad standard.

To me, the problem is not the breadth of the standard, but the lack of publishing practice on relevant/strategic elements. Provided of course that all published data are made (easily) accesible through our data store. The latter will enable all data users to meet their own use case.

Herman van Loon

theo van de sande I agree that the breath of the standard should in principle be defined by the information requirements of IATI data users. The current complexity of the standard and the large number of data-elements published by a limited number of publishers (in 2017 about 30% of the standard's data-element were published by less than 10 publishers) is i.m.o. a reflection of the fact that those data-elements are of limited value to the IATI community (otherwise there would have been more pressure on publishers to publish those elements). It is perhaps the result of the infancy period of IATI in which accommodating publisher requirements was necessary for the growth and viability of IATI.

Breadth of the standard comes with a price: these rarely published data-elements do put a burden on the standard. Tools (data store, data validator, dashboards, registry, documentation, d-portal, etc.) become more complex than necessary and the standard becomes more difficult to use ('IATI is complex').

I think we should therefore not hesitate to depreciate rarely published data-elements if no significant user-group and use-case can be found. But maybe even more important than the depreciation of those data-elements, is that we have a a far higher threshold for adding new data-element to the standard than we have right now. Only changes to the standard for which there is a clearly defined use case which are relevant for a significant user group, should i.m.o. be eligible.

Tim Davies Moderator

I was just reminded in the discussions about a tighter-defined standard of the famous quote "If I Had More Time, I Would Have Written a Shorter Letter".

I think there is an important direction of travel here towards simplification of the standard - but it may be important for us to consider the resources that would demand. Paradoxically, keeping something simple can be much more complicated than just adding elements. It might need thought about the kind of governance models, and community engagement needed. 

Reid Porter

Herman van Loon I like the criteria you laid out regarding new elements. I'm minded though that during the Open Ag days, our donor (a significant player) had a keen interest in something that would aid 'findability' of relevant data, both from the perspective of the user (querying projects related to cashew nuts) and the perspective of the publisher (including the search terms 'cashew nuts,' 'soy beans,' etc. without the need to assign percentages). It was arguably a well-documented use case ("I want to search for projects related to cashew nuts"), and more arguably a fairly intuitive use case ("says what it does right on the tin!") It's been included in AidStream, advertised as part of the 2.03 upgrade, and discussed in the context of several different applications (ag, humanitarian, SDG tracking, etc.) And so, the 'tag' element was born. Interestingly, it just celebrated it's first anniversary on Feb 20. A year later, only 1 publisher actually uses it!

Granted, there was an abortive effort on our part - the 'pushers' for the element - to get people to use it, and the original User Zero lost interest, and it's been harder to explain the difference between tag and sector than we anticipated with certain publishers. Maybe it was a bad add, maybe it just hasn't had time to percolate through the IATI-verse, who knows. I submit, we can't know for sure how reality will play out.

So a stricter criteria for adding elements/attributes is all well and good, but what's our timeline/process for pruning and pairing back? I'll reiterate my point, which Andy raised originally in KTM: our system for upgrades nearly guarantees that the standard will continue inflating because of our (understandable) hesitancy to introduce breaking changes. Short of frequent breaking changes and the ripple effect of downstream changes/investments that result, the only other alternative I've heard is better use of extensions that can be considered 'trials' or group-specific after-market add-ons that exist in perpetuity. But that's at least one additional major change effort it and of itself. 

So what's the way forward? How do we keep the standard from growing complex, and what's the criteria for 'unnecessary complexity'?

leo stolk

Very interesting views on how IATI can further improve the quality, timeliness and usability of data. 

Steven, I agree we should stop counting the number of publishers and data sets as markers of success. A significant portion of published data sets are very poor and difficult or impossible to be used at all, as key elements are missing (location information, implementing organisation information etc)

See large value in Tim Davies suggestion to use the open contracting approach of for a number of communities of practice with clear data use (data demand) defined rules on element use. Possible best complemented by a more strict common mandatory set of elements as suggested by Theo. Together this could to some degree simplify the standard, a stricter core and a more modular approach for the rest. 

From the carrot perspective IATI could attribute one or more markers, a quality stamp or star, to a data set, when it has hit the core demands, and one or more community of practice rule sets. It would stimulate higher data quality in data sets and would allow community of practice to detect and and fetch compatible data sets for their use. 

Tim Davies Moderator

Thanks everyone for the great inputs: we're really digging into some of the ways in which IATI technical approach could evolve.

I wanted to bring back in one key part of the opening questions, which is to ask how the direction of IATI might be shaped by the "evolving development cooperation landscape" in the next 3 - 5 years?

Some of the aspects of the evolving landscape mentioned in the External Scoping paper (see links at the top of this thread) are:

  • domestic public resources as the central pilar in financing for development;
  • domestic private investments remaining significant, though often volatile;
  • growing Foreign Direct Investment (FDI) and remittances;
  • Official Development Assistance (ODA) falling short of international commitments;
  • growth of 'single issue' vertical funds, and private 'impact investors';
  • growth of regional funding in areas such as Association of South East Asian Nations (ASEAN);
  • linkages between aid and tax;
  • partnerships beyond finance - including knowledge transfer;
  • work on improving practice through 'Global Delivery Initiative (GDI)' and 'Global Learning for Adaptive Management (GLAM)' amonst others;

To these I might add the growing importance of climate finance, to address growing climate risks and protect development gains. 

Do any of these have implications for the technical future of IATI? 

Lieke Ruijmschoot

Thanks for the opportunity to engage on IATIs direction and the discussion so far. I support those that advocate for a renewed focus on data use, although of course a way to achieve that is to ensure data quality. It is a bit of a chicken and egg-discussion as we say in Dutch, as quality, being defined as 'usefulness' depends on the inputs by data providers but is defined by data users....

I would like to bring a different issue into the discussion which is one of caution. I miss in the External Scoping Paper the trend of shrinking civic space and increased repression, which has profound impacts on international cooperation. In particular in the area of lobby and advocacy work which is the focus of many Dutch development NGOs and their international partners. It changes the aim of transparency into a dilemma. Striving for complete transparency can bring dangers to civil society organisations, in particular in settings where even receiving foreign funding can be seen as evidence of subversion and a reason to deligitimise an organisation. In an era where once on the internet is forever on the internet, we tend more and more to anonimise and generalise our results. This is of course in stark contradiction to some of the objectives discussed here.

In this light I would like to caution the calls for geolocating every result, as well as connecting with too many other data sources. And as a point of departure for these issues and new directions, revisit the vision and mission of IATI with the safety of its users (in this case I mean data providers) in mind.

leo stolk

Fully support Lieke's expression of concerns of decreasing civic space and increased repression, and how it will impact on the level of transparency. Unfortunately more often data needs to be excluded, in cases where publication evokes risks and threats to those that engage in the activity. We could add the 'whenever responsibly possible' a published record should be a useful as possible. 

Tim Davies Moderator

Thanks Lieke Ruijmschoot and leo stolk - these are really great points.

I'm wondering if there is any empirical work on managing disclosure or non-disclosure of aid project information that we might draw on in the consultation write-up? There has been some work around open contracting and beneficial ownership data I'm aware of for example, but not sure if any privacy studies took place early on in IATI's development. 

And as a provocation - can I ask whether any of these concerns should get us to re-examine the open-closed binary of IATI, or are there ways we might also develop infrastructures that allow more detailed data-sharing between trusted partners, alongside fully open data? (i.e. creating space for the 'shared data' as well as 'open data' part of the ODI's data spectrum)

leo stolk

Tim Davies

Not aware of any write up on management practices on disclosure or non-disclosure of aid project information. May be tricky to get an comprehensive overview of the practice from publishers, as disclosing the scale and the measures of non-disclosure may undermine the non disclosure. Even within the IATI community views on supporting civic space and tolerance may differ..... 

It is up to every publisher and its stakeholders responsibility to recognize and analyse the risks associated with transparency, and take its own appropriate measures. 

In practice publishers regularly share activity data incl financial and result information regarding non disclosed activities to trusted parties in hard copy or other ways outside IATI. In my view it is not within IATI's scope to design an infrastructure for this. 

Björn Haßler

Hello all,

My proposal is for IATI to track digital outputs of projects. 

If we think about IATI's role in a Theory of Change, then finances are clearly an important input. These financial inputs lead to outputs, some of which might be hard to track (from IATI's perspective). E.g. a new school building can be easily monitored locally, but it's hard for IATI to check whether say 1000s of schools have been built (and to a good standard).

However, each project does produce digital assets, and (unlike physical assets) they are easily deposited / tracked etc. In fact, for some projects (such as education projects) these digital assets are a significant sink of resources. I.e. in education projects, significant funding is used in the creation of new digital assets (such as new training materials). However, often these assets do not become publicly available (intentionally or unintentionally). 

A key motivation for this proposal is that, given a record of project inception (on a funders website or on IATI), it is surprisingly difficult to locate resulting digital outputs. Usually, there's a whole chain of content producers, sub-contractors, etc, that makes it laborious if not impossible to actually locate the materials that should have been produced. (Very happy to give examples in a personal communication.) In my mind, being able to track those 'chains', and making them transparent, is very close to IATI's mission.

The proposal: I'd like to put forward that at the time of registering a project with IATI, the project should also register what digital assets they will produce. (A lot of the time, such a digital assets register already exists internally, put in place when the contract with the funder was negotiated.) Then, as the project progresses, the digital assets that are produced are deposited with IATI (or deposited into a recognised archive and linked from IATI).

Of course, more than likely digital outputs change as the project progresses: That's fine, and the digital assets register can be updated at any time. The point is not an absolute benchmark, but to communicate to others (at a point in time X) what has been published so far (up to X), and what is intended over a reasonable period of time (X + 6 months).

Types of digital assets could include reports, educational content, data, etc. Ideally, this would be 'open' content (c.f. Digital Principles, open access, etc), but the main purpose of the idea above is that digital content becomes available to others in a more straight forward way.

Thanks for hosting this consultation!

Mark Brough

I agree with a lot of what Josh has written above (with a slight difference of approach on the on/off budget field).

A few thoughts from my experience of trying to use data with governments at country level, particularly in Bangladesh and Liberia, as well as Myanmar. Essentially, I would argue that we need to work harder to meet the original commitments of IATI (David M says something similar here). Sometimes the really radical, powerful and exciting thing is to keep focusing on what you are already doing (or should be doing) rather than the shiny new stuff :)

1. I actually don't think we focus anywhere near enough on Ministries of Finance in partner countries ("you would say that!" - I know, but hear me out). In most/many cases, they are the ones responsible for both their own budget and the implementation of the national development plan (in some countries, Planning Ministries are also important). So if "ownership" is to mean anything, they need to know what is going on in their country. They are also the ones most likely to have some kind of aid management or coordination unit responsible for managing this kind of data. They are our best hope for shifting the conversation from aid data collection to improving the way aid is delivered.

In general I think we should be working with those existing institutions rather than trying to go around them and potentially undermining them. I think IATI's biggest potential has always been to improve the way aid is delivered and thereby make a significant contribution to improving development outcomes. We need to do more to realise that potential. I think that means having a better understanding of constraints around use of data by partner countries (especially ministries of finance) and working to overcome them.

2. We still need to get the basics right. For me, this means focusing on the largest organisations (bilaterals and multilaterals especially), and ensuring that their data is frequently published and of a consistently high quality. While some donors are doing great, if we do not have most of the largest organisations publishing up to date data, with disaggregated financials, and recognisable project titles and descriptions, then from my perspective focusing on other (albeit interesting and potentially very useful) areas can be a distraction. I think the IATI Secretariat has a key role to play here; it would be good to think through more how that role dovetails with (or can be coordinated with) outside actors.

3. We should be working with those already using data that looks like IATI, and working with them to understand how IATI could help make their lives easier (by improving efficiencies) and allow them to deliver their own outputs better (by improving quality of data collected). I think some of the work around data use could be recast to think more about "how can IATI data help you in your existing work" rather than "how can you use IATI data". That's an easier sell than pushing people to start doing a new thing and adding to their existing workloads.

4. We should take a broad view of what constitutes "aid" or "development cooperation" (e.g. humanitarian, climate finance, south-south cooperation) but we shouldn't be distracted by stuff that looks very different (e.g. FDI, remittances). We should also check whether the standard is really able to deliver for core business (e.g., can IATI data be used for debt management -- one of the key places where partner country governments need really good data -- the answer is currently "not really"). E.g. talk here with the Commonwealth Secretariat and UNCTAD, the main two providers of debt management systems.

5. We should do more to make sure centrally-provided infrastructure is stable and reliable and responsive to user feedback. There are some positive signs here, such as the new Datastore (assuming future maintenance is prioritised -- this has been a key challenge in previous years), but there are other areas where challenges remain (for example, reliably keeping in sync with third-party codelists; providing easy access to codelists and reference documentation in other languages (the translation of the main IATI site to French is great); improving the speed/process by which proposed changes get resolved).

Finally: the internal reflections paper makes reference to the 2018 Institutional Review, but doesn't really discuss which of those recommendations will be enacted over the next few years. It would also be good to have some reflections on how to work with and strengthen the community around IATI, while providing more clarity on the Secretariat's workplan and reporting progress against it (the new quarterly updates are a great start). To be even more radical, the Secretariat could even publish their own quarterly results data to IATI ;)

Tim Davies Moderator

Just logging (so I don't forget it when writing up) that there is a reply from Bill Anderson in Tweet over here to one part of Mark's post: 

Publish What You Fund Team

How could information reported through IATI respond better to partner country needs for relevant and easily-useable data that can be leveraged for national development planning processes, and foster development partner accountability at the same time? 

Donor country offices need to be involved. We've found that a lot of useful data is housed within the Mission, but is not 'sent up' nor often shared to wider civil society. This is a big loss. We know that a number of donor agencies are doing this already, but if all donors could get their country offices to engage more, not only would quality likely improve but so might use. IATI needs to accept that running / funding a handful of in-person training sessions is not sustainable. Instead, for capacity building, it should be looking more at investing in more accessible and intuitive platform(s) for accessing the data, be that D-Portal or an alternative, and embedding it in already established institutions, such as universities. Ultimately, the IATI community is not able to provide enough training in enough places in enough countries to make a dent in the progress it needs. It should consider more sustainable methods. 

Ole Jacob Hjøllund

It is important to stay loyal to the vision/mission and strategic priorities of IATI, referring to the needs of all stakeholders. It is also important to recognise the meaning of the different roles of organisations, before we can specify when we should place greater emphasis on which added value fields.

As an example: If the publisher has the role ‘4 – Implementing’, then it would be correct to place emphasis on e.g. geolocation and descriptions/activity-documents in the country’s language. But if the publisher has the role ‘1 – Funding’ we should hope for un-earmarked or loosely earmarked funding, and thus place far more emphasis on e.g. type_of_aid and the org_id of the participating organisation identified as immediate partner, receiving the transactions – expecting to find more detailed information downstream.

Just like Steven, I would like to expect 'reduction' as one of the outcomes of a future major upgrade. Some reductions could and should be guided by developments that has taken place since Busan, so please apologise – I’m looking backwards for a minute, before I look forward.

Committing ourselves to the development of the ‘common standard’, the Busan outcome document specified that it should build upon a.o. IATI (version 1) and DAC-standards. Thus there is nothing wrong in the fact that our IATI activity-standard is burdened by a lot of ‘added value’ tags from the CRS++ format. But since then, DAC-reporting on ODA has changed from flow-based measurements (that makes sense in IATI-format as well) to a calculated Grant Equivalent measurement, that would not provide any value whatsoever to the users of the IATI-standard. So the join between the pre-Busan ‘old standards’ is broken, and we need to think carefully about how to revitalise the concept of the ‘common standard’.

Simultaneously, the DAC has developed a new measurement framework, based on flow-data, embracing south-south and triangular cooperation, emerging donors and mobilised private finance: TOSSD. This must take center-stage when we decide to move for a major upgrade – i.e. within the next strategic period, but not right now. We should allow the new TOSSD-format to mature, and we should still recognise CRS++ as a standard with which we should ensure interoperability, but it seems likely that a shift from CRS++ to TOSSD as the major, statistical standard on SDG-relevant flows and activities could reduce the current CRS++ tags in our activity-standard by 50%, and the remaining tags would be far more widely relevant for publishers.

Allowing IATI to become a vehicle for the collection of TOSSD data would present the added benefit of shared presentations: TOSSD will deliver standardised statistical presentations, designed to meet the needs from the recipients’ perspective.

A IATI-TOSSD alliance would be a win-win scenario, as I have stated in discussion 2. But it is not all there is to say about strategic directions in this discussion. In IATI, we should go nuclear; we are the only community that can, so that is why I think we must move deeper into the details. After all, we are the nerds. Not necessarily it-nerds, but the peer-community of the people that actually knows how our respective organisations use ‘data’, and what kind of information we would benefit from having exchanged digitally in the future. This must become one of the main efforts in the next strategic period: To demonstrate how (IATI-standardised) digital data-exchange between immediate partners will serve several purposes in efficient and effective development cooperation. And when data is mutually confirmed, by the immediate partners, we have no reason to question the quality. But being truly nerdy: The globally transparent access to data becomes a fringe benefit of data-driven development cooperation.

There is one more issue I would like to add, and to some extend I think of it as the elephant in the room; the fact that there is not one way to improve data quality. If we address the different tags in our standards, and ask how to ensure the quality of each and everyone of them, the answer will not be the same across the various tags, nor across publishers. So we should avoid the assumption that ‘someone should put in more efforts’, and focus our attention on how different data-use scenarios ensures quality. Then we should find ways to prove it, and demonstrate that our standard fits the purpose.

To take just one example: As stated in the beginning, I consider the modality (type_of_aid) to be among the most important added value fields for funding organisations, but how can I ensure the quality? ‘Validation rules’ you may think, but that would just ensure that I use an allowed code, and that is far from ensuring that I use the right one. Having given it some thoughts, I can only come up with one valid answer: Getting it confirmed from the recipient. If the recipient confirms that a specific agreement is of that specific type, then I can insist that all other data-users should trust that data-element in my publication.

This is just to illustrate that the very use of data, applying the IATI-standard to obtain a very distinct confirmation from partners, can be the best and arguably only way to ensure data quality. But across all the tags in our standards, I find that the right strategic direction for us to pursue is to bring the usecases we can offer to the table, and join the efforts of several other organisations and initiatives to define and describe good practice.

There are several international recommendations available, when it comes to good statistical practice. Some of these are indeed relevant for us as well, but they tend to suffer from the weakness (those I have read) that they assume availability of resources to ensure quality in a rather old-fashioned gatekeeper sense. This would constitute a killer-assumption for an initiative like IATI, and is indeed the reason why IATI was not launched as a statistical initiative. So, we do not have guidance on what data quality is nor how to enhance it, until we write it ourselves, or (preferably) in collaboration with strategic partners.

Herman van Loon

Thank you Reid Porter for your example. It perfectly illustrates Andy's point that the current governance structure is geared to adding more and more detail and therefore complexity to the standard and the necessity to control this complexity in order to avoid valuable resources seeping away to maintain elements in the standard without much added value .

Some initial rough idea's to make this control more operational:

1 - have a dedicated integer upgrade for reducing the standards complexity and correcting 'errors';

2 - new elements to the standard start as an extension. These extension can be promoted to the core IATI standard when they are used by multiple users and being published by multiple publishers (meaning there is a proven added value);

3 - data-elements which after x years in the standard are not published by more than y% of the publishers become automatically eligible for depreciation in the next integer upgrade. The burden of proof that such elements cannot be depreciated, lies with the user community;

4 - ask which users will actually start using the new data-element when available before adding anything to the standard (is anybody now using this TAG element?);

Mark Brough

I like the idea of discouraging additional complexity and being a bit more rigorous on whether we add new elements to the Standard. But I would caution a bit on the idea of "simplifying" the Standard, which has come up a few times. The risk here is that users will then have to deal with both the old Standard (if they want to use any existing data) plus the new "simpler" Standard (if they want to use any new data). So rather than making data users' lives easier, it could make them quite a lot harder...

Instead, I think there is a lot that could be done to improve the documentation, especially so that newcomers to the IATI Standard have a less steep learning curve. That process could perhaps also help address errors and inconsistencies without necessarily needing a more far-reaching and potentially disruptive overhaul.

Ok, let's put the tag element issue to rest. :-)

I believe the main use of this element will be the SDG goals and targets - to identify which SDG(s) the funding intends to support. It was really good for the IATI community to anticipate this need and make sure the was a place in the standard to put the data. Unfortunately the DAC decision to use SDG goals and targets as flags/tags took a long time. Now that the DAC decision is made, all that remains to do is for the IATI Tech Team to approve the addition of the codelist to the tag element - then publishers can start publishing them!  There is a good use case for this data and it would be great to see outreach efforts in SDG-related fora once some data is available. 

That being said, I do have sympathy for the general idea to contain the tendency to bloat the standard with new elements or attributes. 

Saripalli Suryanarayana

Physical house hold data and IRIS/FINGER PRINT to have a valid Number for health care and poverty eradication is important.Linking such with the bank account does what exactly what India has done to get proper house hold/individual data in as large as 1.28 billion.

In fact china is using the Photo graphy for PIGS/and Cows.So also parental DNA analysis,with its 1.30 billion population.

World gets benifit if such Neglected Tropical/Rare disease along with communable vectors are found at Public health places the WASH programme becomes sucessful.

Matt Geddes
  1. Across various forums we have a long list (with some key priorities) of things that would help aid recipient countries to make better use of the data. Having workstreams on data use is real progress, but we have known most of these things for quite a while. Josh (language) and Mark (debt management use) raise two above – there are some real elephants in the room (double counting, activities vs projects, officialness of data, not comprehensive etc). I suspect the problem is rarely technical or lack of knowledge. I think that the system we (IATI funders, secretariat, community) have setup for prioritising what to spend our collective energies on doesn’t prioritise these issues. Until we investigate why so few of these issues have been solved in the past 8 years (and change something) then I don’t expect to make much progress. It may be that we prioritise new features, or non-breaking changes, or don’t have the resources, - all very justifiable decisions, but I think being a bit clearer about what has happened in the past will be a key part of a more honest way forward than pretending we just need more workstreams, case studies etc. In a similar vein, I remember seeing the the Rwanda IATI profile said that GoR think IATI is not worth it/doesn’t cut the mustard at the moment – this seemed nuclear to me, but it barely caused a stir on the forums – maybe it did at the TAG, but I doubt it.

  2. I suspect there are a lot of IATI features that don’t do what they were designed to do – that is normal, and not a bad things for an evolving effort, so it makes me really tempted by the simplification agenda and basics suggestions above. This could be the start of a 'tick-tock' where 2.01 expands the standard, 2.02 drops some bits, and 2.03 is the long term support version. One I wanted to know was whether the inclusion of contact email addresses makes a difference. Has anyone ever emailed one of these and got an answer? The answer might be yes, and my suspicions are unfounded, but I think the idea that we have an assessment of each field is a good one if we are to keep the standard evolving. I also put secondary publishers in this list – I think this is a great feature but many of the tools don’t even show them and so we should reassess how we handle this (hopefully keeping them!).

  3. I don’t see a framework through which changes to the IATI standard or core tools are assessed. The example I am thinking of is the decision that the new datastore will have a different API to the old one. This means that all tools that use the old datastore API will be broken. This might be the right choice, but I think we are caught in a difficult position of having been an engaged and well wishing community who are small enough to just make good calls based on a comments thread and a google doc link, and a large organisation that needs proper procedures for these things. For example, I was really surprised that when we ask people to publish using IATI data, we do not give a guarantee that the version people use will not be depreciated for X years. Similarly, the decision to start dropping non open license data. I think IATI needs to take the next step up as a provider of a service and start offering these sorts of things. Both of these needed to go through a transparent procedure that examined what the impact would be, and there should be standards for what kind of impacts are acceptable, especially if we want people to start building tools etc. There is a reason that the OECD DAC is slow at changing things, and despite it winding me right up, it is not all bad – we need to evaluate whether the current standard management model is the one that is most supportive of the standard and its use, or whether a few more years of evolution, even if we burn some fingers is the right call.

  4. I don’t think that IATI as a system is sustainable yet and would like to see a workstream addressing that before it is too late. Donor (publisher) support will fade over time (see the Paris Declaration - or imagine DFID being rolled into the UK Foreign Office). IATI needs to think hard where will the support the standard needs will come from in that case. Try playing with the scenario of no core funding to the secretariat from the current funders, what would need to change for IATI to survive? Would IATI survive without the secretariat? Does it offer enough that people would be prepared to pay?

  5. For the FDI/remittances brigade, before spending time adjusting IATI to handle it, we need to show an existing use case for that data to be in IATI – I haven’t seen one yet. It is not like the activities/transactions model, or the self publishing model is likely to be a good fit for such metadata sparse and often non-public data. Most of the other ‘beyond ODA’ stuff also isn’t actually here yet – and those that are e.g. DFIs, blending, vertical funds, climate funds etc – are generally still aid driven and therefore in IATI already. So aside from the joy of buzzwords, I would de-prioritise that workstream for now.

  6. I agree that we need to stop focussing on number of publishers. The number of published reports that use IATI data might be a good replacement – even better if it also included the number of recipient country budgets/aid flow reports – and this returns to my first point – the metrics IATI uses are sometimes a dead giveaway to it’s mindshare (maybe we don’t realise that we have left the startup stage yet and still need hugely positive self-promo). Some new metrics might be a better stick to beat ourselves with as we move to a longer-run sustainability and data use outlook.

I didn’t enter a fancy title or employer/job description when registering so in case someone wants it: My experience of IATI is a) someone who is often asked to write reports showing the international aid flows to a country or to a sector and b) I worked with Mark to help Bangladesh add IATI import to their AIMS, and now support Somalia to make a new AIMS which will have IATI import (and publishing) as a core feature. Before that I spent 6 years as embedded TA to aid management units in Sierra Leone, South Sudan, Mozambique, Uganda, Afghanistan and Somalia – hence primarily focussed in Ministries of Finance etc.

Harsh Anuj


As a user of IATI data to inform the development community regarding the challenges faced during project implementation in specific contexts (see, I have found it to have great potential. It can allow our analytics platform to seamlessly pull data (specifically project documents and attributes) from different organizations in a standard format.

Having said that, I feel that there is a lot that different organizations can do to make their own data more standardized (i.e. prior to publication in IATI). As a global aggregator of development data, I feel that IATI is well positioned to drive this mandate within different organizations. This is of course related to the issue of transparency as well. I hope that going forward, IATI can leverage its unique position to drive organizations to publish more data in a more standardized fashion.

Finally, I feel that both IATI and development organizations need to give more thought to publishing data in a form which can be useful. A major value of the data published on IATI resides in the insights which the development community can derive from it to do things better.

Best regards,


Hi Harsh

Thanks for these comments. As someone working on the donor/publisher side, I would find this kind of feedback extremely useful to convince my own organisation/colleagues to improve our data. Yet we never hear from data users. I don't know if you've ever send an email or something to a publisher to tell them what is wrong with their data, what should be more standardized etc. If so, how did it go? If not, why not, how could we convince you to?

In other words, do you have ideas about how we - or the IATI community - could do to facilitate specific, detailed feedback on data? What would work and what wouldn't, in your opinion? 

Alberto Lizzi

Thank you, Harsh, for your comments which I find very useful and practical. 

Indeed, at UNDP, we are looking at how to make data more useful and to move from simply reporting results to improving results. It is a very important -and difficult- shift that revolves around learning from results and challenges (lessons from evaluations and project reports) to make sure that UNDP is able to improve from its own (and partners by sharing on ) know how. 

The biggest obstacle we experienced in making the shift happen is the lack of data standardization, making analysis via machine learning very hard. To give you a more concrete idea, because of the lack of standardization, we were only able to analyze roughly 5% of the lessons. It would take considerable resources in terms of time and workforce to clean up and make the rest of the lessons machine-readable. 
By closely engaging with other UN agencies, I can report that UNDP is not alone in this. 

IATI looks well positioned to promote common standards that would allow its members to lower these barriers to analysis. Standards should move beyond results and include the reporting of challenges and successes to allow organizations to draw more actionable information from what works and what doesn't. The issue of standardization could be then tackled by promoting machine-readable formats that analytical outfits like -and others- can leverage. 

Thank you for the opportunity to discuss this important topic.



Andie Vaughn

Responding for USAID, U.S. Department of State, and Millennium Challenge Corporation (MCC):

We are late to jump into the conversation, but skimming through the above, it looks like we are aligned with many of the opinions above - we've enjoyed reading through all of these ideas.

The US believes we need partner countries to respond to the first question.  We all likely have an idea of what could be improved or where the gaps are, but ensuring the partner country constituency is guiding this conversation and these needs will be vital to ensuring any efforts put forth by publishers is successful.

Notably, however, most of us recognize that there are challenges with the accessibility of IATI data for the vast majority of users. XML is highly technical and difficult for most users. In addition, more casual users typically rely on visualizations to explore datasets. D-portal is a useful tool, but does not show regional data reported to IATI and does not provide data for every field that is collected. Currently, it is not clear that the average user with only a cursory knowledge of foreign aid would be able to navigate the data published to IATI to answer research questions or inform decisions. The challenges with accessibility mean that a large number of organizations with a role in development partner accountability and national development planning are excluded. 

Continuing to push publishers to fulfill existing fields with quality information should be the focus of IATI for the immediate future.  Additionally, highlighting examples of how this data is being used and why it is needed (particularly if these requests come from partner countries) will help encourage publishers to act – to modify systems, policies, and data collection. 

IATI should also focus on informing grassroots and local organizations in partner countries about IATI and how to use the data. This should involve outreach to a variety of organizations that could benefit from IATI’s data—not just organizations with sophisticated data operations. For cost effectiveness, identifying relevant intermediaries, such as professional associations, that could partner in reaching potential IATI data users could be an approach.

Regarding the next 3-5 years:  IATI is a valuable resource for connecting the dots and making links so that users of data and data publishers can more efficiently and effectively utilize their resources.  IATI needs to have the tools to make those connections – the datastore and d-portal and ensuring the data available in both of these is valid, quality information, are vital moving forward to ensure that data needs and use cases can be met.

To the extent there are fields in partner country budgets, expenditures, results reporting, etc., having comprehensive mappings to these data against existing standards and fostering intermediaries (ie AIMS development) to develop these country specific mappings where needed, would be very useful.

IATI may need to put in place decision criteria for accepting new data fields or data sets related to specific sectors or initiatives (for example adding fields for each SDG target versus mapping existing fields to targets).

Petya Kangalova

It has been extremely useful reading the responses to this consultation and a lot of really important points have already been picked up in the discussion thread.

1. How could information reported through IATI respond better to partner country needs for relevant and easily-useable data that can be leveraged for national development planning processes, and foster development partner accountability at the same time?

I believe the key here is not to talk about data quality and data use as two separate components. In order to improve data quality you need to ensure that users’ needs are met and visa versa creating this feedback loop we have been talking about, the binding part of which is the IATI standard.  Addressing question one I think we need to ensure that we:

  1. Standardise the standard - ensure that at a technical level all bugs are being fixed, documentation about the standard is clear and consistent and standard processes are clearly evidenced.

  2. Bridge the language gap (and I don’t necessarily mean it in terms of English, French,etc.). How can we talk about IATI so that we speak the ‘common IATI language’. How can a technical expert talk to partner country officials and make sure they understand each other? Can we all speak in plain IATI language? I think that requires training not just for partner countries to improve their knowledge of the standard but training for technical IATI experts to be able to understand and address partner country needs.

  3. I don’t believe there is a problem with the breadth of the Standard but rather the lack of recommended guidelines and publishing best practices. Over the next few years, we should prioritise improving this, with the aim to understand what are the relevant and strategic elements that partner countries want to find and use in IATI data. Allowing such a feedback mechanism and really listening to partner countries on what they see as data quality improvements would allow us to develop those guidelines. Tim’s point about allowing  'sub-communities' in IATI to emerge, focussed on the different overlapping use-cases for the data, would support the simplification of the standard.

2. Given the evolving development cooperation landscape, what new directions do you foresee for the IATI Standard over the next 3-5 years?

On the second question the points I’d make are:

  • Prioritisation and relevance of IATI elements really need to be driven by demand by the IATI users (partner country government needs)

  • Ensure that everytime we talk about the IATI Standard (at a technical level), we position it in the wider context of the open data movement and other data standards. Make sure we don’t sit within an IATI silo.

Manongi Ntimbwa

For data to be helpful in Tanzania and perhaps the whole of East Africa. AWARNESS needs to be created to the media people, students at lest from Masters to above level, researchers and policy makers. Once the said are well informed of the IATA data sources the level of use will increase and so be impactful. 

Thus, the comming strategy should include awerness creation.


Posting this on behalf of Mr Mokute Mopolo, Pour le Coordonnateur National en mission, République Démocratique du Congo:

C'est avec un réel plaisir que nous venons participer à la consultation en ligne pour aider à façonner le plan stratégique de l'IITA 2019-2022! commencé depuis le 4 février.

Notre attention se focalise sur Discussion 3: Considérer les orientations et les données pour la coopération au développement.​​​​​​​

La République Démocratique du Congo est un pays fragile à cause de nombreux conflits armés qu'elle a connu, n'ayant pas assez de ressources internes pour financer son plan de développement, sollicite régulièrement l'aide internationale pour palier à cette insuffisante. Voilà pourquoi elle avait opté de mettre en place une base de données dénommée " Plateforme de Gestion de l'Aide et des Investissements" en sigle PGAI qui a pour mission principale de capter toutes interventions des partenaires techniques et financiers. Malgré la présence de cet outil nous n'arrivons pas à capter à temps les données des projets financés par les bailleurs.Nous avions mis beaucoup d'espoir à l'IATI et étions le premier pays à importer automatique dans notre base les données postées dans le site de l'IATI, malheureusement, l'expérience n'a plus continué.Malgré tout nous réitérons notre adhésion à l'IATI et continuerons toujours à exploiter les données pour la coopération au développement postées sur le site. 

Que les donneurs publient leurs données au format que nous avions proposé à l'IATI pour un meilleur échange automatique.


Pour le Coordonnateur National en mission


Posting this on behalf of Erik Hesseling, Department International Programmes, IATI Coordinator:

I did not participate at the discussion on the Strategic Plan till now. The main reason is could not trace the suitable discussion.

My mayor issue is that I would in favor of having more effort on how IATI publishers can make better and more frequent use of the IATI data and XML’s for the purpose of there own organization. For transparency, visualization, reporting and dashboards.

We as Netherlands Enterprise Agency we refresh frequently our datasets so others use the XML. But even more important is the use of datasets for IATI visualization website This website becomes more and more important as management tool. For policy makers in my organization but also Dutch MFA.

But there are even better IATI project websites like those of UNDP, African Development Bank, Oxfam Novib, Agence Française de Development. Enabel, Cordaid, e.a.

I would be in favor of having in the years ahead a more close exchange of views, experiences and best practices among IATI publishers who use the IATI xml for transparency and management purposes by having this kind of website.

Best regards,

Erik Hesseling


Posting this on behalf of Kabura Junie, Assistante en Base de Données Suivi-Evaluation des Programmes, Secrétariat Permanent du Comité National de Coordination des Aides "SP/CNCA", Burundi:

La question d’affichage, de téléchargement  et d’impression des données est encore pertinente.

Quant à moi, j’aimerais voir sur le site les informations désagrégées par Ministère tutelle du projet

Natalia Magradze

2. Given the evolving development cooperation landscape, what new directions do you foresee for the IATI Standard over the next 3-5 years?

The “primary” audience for IATI data use should continue to be partner country governments. To this end, user friendliness and data quality remain paramount to the uptake of the IATI data by developing countries. Therefore, a concerted effort is needed to better understand the impeding factors to the lack of use of IATI data by the partner country governments. Conducting an assessment in several countries would be useful looking at (1) what tools and technical support might be needed to integrate IATI data into country government internal systems including AIMS, (2) what kind of data sets they would find the most useful, (3) what publishing cycle and reporting frequency is the most optimal.

Saripalli Suryanarayana
True Natalia the data needs to be seggregated in such a way that unless connected/the details of healt-DNA-property are not used by the IATI.The health details and Public health shall be preserved seperately,as they preserve the banking/finacial-prpoerty are preserved.So also the house hold data shall not be a election data for political parties. Protection of data/division and acessabulity shall be controlled.Be cause many issues are interconnected.

Posting on behalf of Mishiko Seino, Programme Officer, Performance Management & Accountability, Field Results Group, UNICEF:​​​​​​​

UNICEF comments:

  • If used properly, the geo-location fields has great potential to scope who is doing what, and where within a country at the national and subnational level. 
  • Documentation has limited value if only attached to data at an individual activity level. It could show that an activity was implemented and how much was spent and for what, but how does this show what was the development and humanitarian results at the country level? Having an option to provide documents at the outcome or country level can show progress made towards achieving national or global goals and also promote a common understanding of the development and humanitarian results. 
  • As IATI evolves, the objective in the future would be to provide data that is easy to understand and used by anyone who is interested. Can there be a built-in tool within the registry to connect the data from different publisher for aid traceability? For example, when looking through the data of one publisher, could there be a flyout display of suggested linked iati-activities from another publisher from the registry?

Please find below consolidated input from the European Commission, received from Andrea Saviola:​​​​​​​

Geo-location, in particular sub-national location, is important. However, it is correct to place the right emphasis on the delivery chain (as explained by Herman van Loon in his post).

Since the initiative is looking for strategies to increase data quality and data use, it is of utmost importance to ensure stability in the standard, or at least leaving more time between the different versions. After having aligned to 2.02, a new 2.03 version was already under preparation. This affects both big and small organisations which do not have the time to adapt publications that they are already challenged by new requests.