Sometimes, “killer facts” rival shock photos in humanitarian communications. These hard-hitting statistics are certainly intended to mobilise public opinion, but their increasing politicisation calls into question their credibility. How can we produce reliable data and preserve their power of conviction in a world of post-truth and widespread mistrust?
For several decades, understanding and measuring the needs of crisis-hit populations have been a key pillar of humanitarian work. The dynamic picked up with the commitments in the Grand Bargain in 2016,[1]Inter-Agency Standing Committee, Qu’est-ce-que le Grand Bargain ?, 1er juillet 2017, https://interagencystandingcommittee.org/grand-bargain-official-website/quest-ce-que-le-grand-bargainwhich stepped up the requirement for accountability and efficiency in resource allocation for organisations ranging from international funding bodies to non-governmental organisations (NGOs) working on the field.[2]Center for Humanitarian Data, Série de notes d’orientation. La responsabilité des données dans l’action humanitaire. Partage responsable des données avec les bailleurs de fonds, décembre … Continue reading Almost automatically, these pieces of data became key components of advocacy practices aiming to convince and steer donors, decision-makers and the general public.
The strategic use of statistics in humanitarian communication campaigns goes back further in time, starting in the 1980s with the rise of professional advocacy, particularly in British NGOs. They popularised the use of killer facts – punchy, simple, memorable figures – to grab the public’s attention, simplify complex issues and quickly galvanise support. Nowadays, these data are displayed on subway advertising hoardings, social media and viral infographics, giving emotional messages an aura of scientific legitimacy and irrefutability. This strategy, analysed by Pascal Dauvin in particular (Pascal Dauvin (dir.), La communication des ONG humanitaires, Éditions L’Harmattan, 2010.))is in line with a rationale of professionalising advocacy, with impact often taking precedence over nuance. The link between information and communication is therefore forged around “killer” facts, aiming to both inform but also galvanise opinion, and encourage donors and decision-makers to take action.
“Use of scientific evidence to demonstrate a state of affairs or promote a cause has perhaps never been such a necessary and complex undertaking.”
However, in the fake news and post-truth era, use of scientific evidence to demonstrate a state of affairs or promote a cause has perhaps never been such a necessary and complex undertaking. The increase in the volume and sources of data in the humanitarian sector has gone hand-in-hand with growing politicisation. This politicisation can partly be explained by financial challenges, particularly the need to prioritise how funds are used. Geopolitical considerations, such as managing the image and reputation of stakeholders involved in crises, also play a part.
States are not the only entities to grasp the strategic importance of humanitarian data. All the actors involved in a humanitarian conflict or crisis – be they armed groups, opposition forces, guerrilla movements or humanitarian aid stakeholder organisations – may seek to influence, monitor and even control the production and dissemination of data. Although their statuses, aims and means differ, they all regard data as a way of leveraging power to legitimise their action, steer the dominant narratives or challenge others’ narratives.
In this context, how does the production of reliable humanitarian data represent a major advocacy and communication challenge? How do you produce credible humanitarian data and maintain its power to convince and be channelled into tangible initiatives to help the affected populations?
The growing significance of data in the humanitarian sector and advocacy
Since the start of the 2010s, data collection and use have become increasingly important in the humanitarian sector, with a growing emphasis on the efficiency and transparency of humanitarian initiatives. Nowadays, data plays a pivotal role in understanding the needs of affected populations and in guiding the programme and operational choices made by NGOs and funding bodies. The whole Humanitarian Programme Cycle under the leadership of the Inter-Agency Standing Committee (IASC), which directs the large majority of humanitarian funding at a global level, is based on aggregating and analysing thousands of pieces of data from crisis-hit areas. Under this coordination system, specialised stakeholders are specifically tasked with data management, analysis and sharing. One such stakeholder, the Centre for Humanitarian Data, created by the United Nations Office for the Coordination of Humanitarian Affairs (OCHA), plays a key role. Located in The Hague, the centre aims to improve humanitarian decision-making through improved data use. The centre supports humanitarian stakeholders in four main areas: data management, data visualisation, ethical responsibility and predictive analytics. The centre also runs the Humanitarian Data Exchange (HDX), an open platform enabling millions of datasets from United Nations agencies, NGOs, governments and other partners to be centralised, organised and shared. HDX facilitates access to reliable and up-to-date data on humanitarian crises, while promoting common data use standards.[3]OCHA, The Center for Humanitarian Data: Connecting people and data to improve lives, https://centre.humdata.org ; OCHA, Présentation du Centre pour les données humanitaires et HDX, 13 mai 2024, … Continue readingThis shared data platform has become a cornerstone of modern-day humanitarian coordination.
Moving beyond global prioritisation, data has a direct operational impact in the field. A striking example dates back to the Russian invasion of Ukraine in February 2022. While media and humanitarian attention was focused on the refugees fleeing to neighbouring countries, a coalition of humanitarian stakeholders – IOM (International Organization for Migration), UNHCR (United Nations High Commissioner for Refugees) and OCHA – published an internal displacement map on 15 March revealing that close to 6.5 million people were internally displaced in Ukraine,[4]IOM, REACH, UNHCR and OCHA, Update on IDP Figures in Ukraine, 18 March 2022, https://www.unocha.org/publications/report/ukraine/update-idp-figures-ukraine-18-march-2022-enuk; … Continue readinga much higher figure than the initial estimates of 700,000 internally displaced persons (IDPs).[5]International Organization for Migration, Ukraine Internal Displacement Report – General Population Survey – Round 1, 16 March 2022, … Continue readingThis revelation radically changed perceptions of the crisis: it was no longer just a refugee crisis, but also an internal displacement crisis. On the day after publication, funding bodies began to redirect their funding to the interior of the country and humanitarian organisations (UN, NGOs, etc.) started to redeploy their teams to eastern Ukraine.
“Regardless of how robustly data is produced, there is no guarantee that it will be taken into account or that it will influence political decisions or galvanise public opinion.”
This example illustrates the importance of data when seeking to understand a crisis, clarifying the story being told and steering decisions. However, it also implicitly brings to light a more ambivalent reality: regardless of how robustly data is produced, there is no guarantee that it will be taken into account or that it will influence political decisions or galvanise public opinion. Data may be ignored, played down or misused in highly politicised contexts.
Consequently, unlike the case of Ukraine, other crises, such as the repeated capsizing of migrant boats in the Mediterranean and the airstrikes in Gaza, are documented by precise, plentiful and widely accessible data. However, these figures struggle to bring about major political change or galvanise international action on a scale commensurate with the situation. This contrast highlights a fundamental tension: data cannot speak for itself. Its impact depends on how it is interpreted, conveyed and incorporated into strategic narratives, or, on the other hand, how it is sidelined or neutralised in environments in which political interests hold sway.
Given the increasing importance of data in the humanitarian sector and in Western culture in general, pieces of data – as killer facts – have become key components of humanitarian communication and advocacy. These punchy, memorable statistics, designed to grab people’s attention and galvanise public opinion, are tending to replace (or supplement) the “shocking” photographs used since the 1970s to visually encapsulate the seriousness of a situation. Just like those photographs, these figures condense complex realities into simple messages and are highly emotionally charged, albeit not without raising ethical questions.[6]Françoise Duroch and Maelle L’Homme, “Ethical considerations around the use of humanitarian imagery”, Humanitarian Alternatives, issue 21, November 2022, p. 92-105, … Continue reading
Killer facts give humanitarian messages the aura of scientific legitimacy and irrefutability to which we referred, increasing their persuasive power. These figures are used in advocacy campaigns and appear to be effective mobilisation tools. However, their impact does not solely hinge on their content: their impact is strongly influenced by the political and media context in which they are publicised and by the level of trust placed in the organisation publicising them. In an information-saturated environment marked by mistrust of institutions, the perceived legitimacy of humanitarian stakeholders becomes a determining factor in the data being heard, believed and taken on board.
Consequently, just as images may be rejected for being sensationalist or misused for political purposes, killer facts may be ignored, challenged or twisted if the publishing source is not viewed as credible. This underlines the importance of humanitarian organisations fostering a relationship of trust with their target groups – funding bodies, media outlets, decision-makers and public opinion – so that their data can really have an impact on debates and decisions.
The datafication of humanitarian practices has been the subject of criticism, notably focusing on the tendency to reduce human needs to quantitative indicators, which are often disconnected from the socio-cultural reality of the affected populations. Though data plays a key part in decision-making, this critical role raises ethical and political questions.[7]Centre for Humanitarian Data, Guidance Note Series – Data Responsibility in Humanitarian Action: Humanitarian Data Ethics, January 2020, … Continue readingAs Joël Glasman states, humanitarian data is not neutral.[8]Joël Glasman, Humanitarianism and the Quantification of Human Needs: Minimal Humanity, Routledge, 2019.Data is developed in specific contexts, influenced by power relationships, and can mask some unquantifiable needs. His assessment is shared by the Humanitarian Data Science and Ethics Group (DSEG), which reiterates that data production must go hand-in-hand with ethical, transparent and participatory governance to prevent data from reinforcing inequality and marginalising some voices.[9]Kate Dodgson, Prithvi Hirani, Rob Trigwell et al., A Framework for the Ethical Use of Advanced Data Science Methods in the Humanitarian Sector, Humanitarian Data Science and Ethics Group, 2020.
In this regard, Hugo Slim delves further into the issue by calling for an ethical redefinition of humanitarian needs.[10]Hugo Slim, How should we define and prioritise humanitarian need?, Norwegian Center for Humanitarian Studies, 13 November 2023.He stresses the need to recognise the intrinsic moral tensions involved in any attempt to prioritise needs and advocates for a more humane approach, driven by compassion, simplicity and awareness of the limits of humanitarian action. This approach urges us to move beyond purely technical frameworks and take on board qualitative and contextual dimensions when assessing needs. The methods used include discussion focus groups and semi-structured interviews. They enable locally perceived needs to be highlighted. These needs are often missing from standard indicators. For instance, displaced women may express concerns about safety in distribution centres or about access to menstrual hygiene products, yet these issues are rarely included in standard quantitative indicators. Similarly, conflict and environmental assessments help identify risk factors that are invisible in numerical data, such as intercommunity tensions about water access or the differential impact of a crisis according to gender or age.
The challenges in producing credible, reliable data and their use in the era of fake news
As well as the extensive usage of killer facts, or, more generally, scientifically produced data, in political decision-making, mistrust of science, institutional sources and the promotion of “alternative facts” have increased considerably over the past decade. In the United States (US), trust in scientists dropped from 87% in 2020 to 73% in 2023, before slightly rebounding to 76% in 2024,[11]Agence Science-Presse, Confiance en la science : une remontée ?, 16 novembre 2024.but the downward trend certainly ought to resume in 2025, given the huge attack on scientific research by the Trump administration, marked by budget cuts, mass redundancies and restrictions on academic freedom. In the fake news era, relying on scientific evidence has become both an absolute necessity and a dizzying challenge.
“Even when based on robust data, killer facts are only effective if the publishing source is seen as legitimate and trustworthy.”
Killer facts have an ambivalent role in this context. They are often extracted from robustly produced data and are a condensed, simplified, publicly accessible version of the information. The effectiveness of killer facts is based on their ability to turn complex analysis into punchy messaging. However, even though this simplification facilitates communication, it can also lead to nuance being lost and data even being misused. Moreover, widespread mistrust of institutions, including scientific and humanitarian institutions, directly impacts how these messages are received. Even when based on robust data, killer facts are only effective if the publishing source is seen as legitimate and trustworthy. Scientific legitimacy is no longer enough to guarantee credibility in an environment saturated with competing information. This emphasises the importance of humanitarian stakeholders increasing transparency, traceability and education about data in order to restore a bond of trust with their target groups.
This crisis of trust is part of a broader phenomenon: the growing politicisation of humanitarian data. Although humanitarian data’s central role in decision-making processes has been cemented, data production and use are increasingly influenced by political, economic and geostrategic interests. States and other parties involved in crises – local authorities, armed groups, non-state actors – seek to control information flows to direct aid, shape storylines or legitimise their positions.
In Syria, for example, UN and Human Rights Watch reports documented cases of the regime manipulating humanitarian data, with the aim of channelling aid to areas that were supportive of the regime, to the detriment of other vulnerable populations.[12]Human Rights Watch, Everything is by the Power of the Weapon: Abuses and Impunity in Turkish-Occupied Northern Syria, 29 February 2024, … Continue readingMore recently, in December 2024, the US publicly challenged the findings of the Famine Early Warning Systems Network (FEWS NET), an early warning system based on scientific data, regarding the risk of famine in Gaza,[13]Joe Federman, Israel is “nowhere close” to meeting U.S.-set aid goals, say Gaza charities, Video, Associated Press, 12 November 2024, … Continue readingthereby illustrating how robust data can be rejected on political grounds.
So how can credible data for use in decision-making be produced today? Robust, independent, transparent methods are needed to guarantee the credibility of humanitarian data. Data checking mechanisms, as well as an open debate on data collection and analysis processes, are key to ensuring data quality. Collaboration between humanitarian stakeholders, including affected communities, enables more representative and relevant data to be produced. Above and beyond methodological rigour, placing affected populations at the heart of needs assessments involves a more fundamental transformation of data production practices.
Firstly, it means actively including the relevant communities in all phases of the research cycle: from designing tools to interpreting findings. Participatory approaches, such as focus groups, semi-structured interviews or perception surveys, help highlight needs that are often invisible when using standard indicators. Secondly, it involves building local capacity, developing community feedback mechanisms and embedding data collection in accountability and safeguarding principles. It is not just a question of adjusting technical tools, but instead of reappraising the way in which needs are defined, measured and used to guide humanitarian responses. In our view, the production of credible humanitarian data to maintain data’s power of persuasion and its capacity to be channelled into tangible initiatives should involve the different actions.
Building a robust evidence-based system
“Data should be supplemented by qualitative information provided by communities to gain an understanding of their local dynamics and priorities.”
In response to the prospect of a funding cut, there is a need to build a person-centred, evidence-based ecosystem, guaranteeing precise, contextualised needs assessment. Data should be supplemented by qualitative information provided by communities to gain an understanding of their local dynamics and priorities. The rollout of such an ecosystem requires significant resources and multi-actor coordination, which can prove difficult to achieve in crisis contexts. The aim is not just to produce data, but also to robustly interpret the information to learn lessons that will be useful for decision-making.
Such an ecosystem cannot be developed without solid analytical frameworks to organise, compare and interpret the collected data. Indeed, the production of relevant, contextualised data is just a first step. The data also needs to be incorporated into analytical tools that are coherent, shared and recognised by all humanitarian stakeholders. The Joint Intersectoral Analysis Framework (JIAF) aims to do just that. The JIAF seeks to harmonise the evaluation of crisis severity and pilot a more equitable distribution of resources.
Producing robust analytical frameworks
The JIAF needs to be strengthened, particularly in its coherent, systematic application across different contexts, to enable a comparable analysis of the severity of crises and an equitable distribution of resources. This involves addressing existing methodological challenges while guaranteeing a transparent rollout of the JIAF. This approach is in line with Hugo Slim’s above-mentioned thinking,[14]Hugo Slim, How should we define and prioritise…, op. cit.urging us to move beyond purely technical frameworks and incorporate human and contextual dimensions into needs assessment.
Making decision-making processes more transparent
Prioritisation and planning must be based on scientific data, while explaining the process and sources used. This is one of the few ways of alleviating the effects of the growing politicisation of humanitarian aid by making choices clearer and decision-makers more accountable. By the politicisation of humanitarian aid, we mean the political misuse of aid by states or geopolitical actors, constituting a threat to the impartiality and neutrality of humanitarian work. Aid is used as a diplomatic lever, a tool for territorial control, or a means of political legitimisation in some contexts.
Given these risks, it is vital that critical data systems are maintained, i.e. mechanisms able to provide reliable, independent, real-time data on the most urgent humanitarian needs. This notably includes real-time monitoring systems, which provide early warnings about famines, unmet basic needs and mass population displacement. These systems also play a key role in providing early warnings when natural disasters occur (drought, flooding, earthquakes) by facilitating a swift, targeted response. It is crucial that they are maintained to guarantee responsive, evidence-based humanitarian action that is less vulnerable to political pressure.
Clear, contextualised strategies for communicating humanitarian data
Producing reliable data is not enough in itself. The data needs to be rigorously analysed to grasp its significance and then needs to be communicated in a way that is clear, transparent and tailored to the context. An effective communication strategy does not just boil down to presenting quantitative findings. Instead, it must also explain the methodology, recognise the limitations of the data and emphasise the legitimacy of the information.
This transparency does not guarantee acceptance, but it is a key lever for increasing public and decision-maker trust and for limiting the impact of disinformation. This involves putting together key messages on the source of data, data robustness and data’s practical use for directing action. Working together with local stakeholders is also vital to tailor messages to the cultural and linguistic reality, and, in this way, facilitate local buy-in. This approach increases data legitimacy, reduces the risk of manipulation by political decision-makers and limits rejection by populations. Finally, including data in human storylines while maintaining scientific robustness may help galvanise public opinion without falling into the trap of excessive simplification.
“Reappraising humanitarian data also means recognising that information is never neutral.”
The challenge is no longer to simply produce reliable data, but instead to reappraise data’s role in humanitarian work. In the future, the humanitarian sector must move beyond technical requirements in order to build inclusive, ethical data systems that are truly useful for affected populations. However, this also involves questioning what we choose to measure and why. Each and every piece of data conceals methodological choices, power relationships and sometimes blind spots.
Reappraising humanitarian data also means recognising that information is never neutral. It involves taking a gamble that by combining scientific robustness with transparency, while listening to communities, data can once again become a lever for justice and change, rather than a simple management tool.
Translated from the French by Gillian Eaton
Picture credit: Mati Flo
