Responsible digitalisation in migration governance starts with more than technology

Digital technologies are transforming how governments and organisations deliver services, make decisions and interact with people. Migration governance is no exception. Mobile applications help migrants navigate services; digital identity systems facilitate access to documentation; matching platforms connect people to jobs or places of settlement; and increasingly sophisticated data and AI tools support administrative and policy decisions.

The question is therefore no longer whether migration governance will become more digital. It already is. The more pressing question is what responsible digitalisation should look like.

Our recent GIZ-commissioned study on migrants’ rights and digital solutions in migration examined this question across the migration cycle. It drew on an inventory of 100 digital tools, as well as academic and policy literature and interviews with experts and practitioners. Rather than simply identifying promising technologies, the findings point towards several choices that governments, development cooperation actors, international organisations, civil society and technology providers need to make as digital tools become more widespread.

Start with the purpose, not the technology.

Perhaps the most important question is also the most basic: What problem is a digital tool actually intended to solve, and for whom?

A meaningful difference exists between digitising an existing administrative process and using technology to improve migrants’ access to information, services, or rights. Yet this distinction is often blurred. A technically efficient tool may make a process faster without necessarily making it fairer, more accessible or more rights protective.

This matters particularly in migration governance because the same technologies can serve very different purposes. A digital system can increase migrants’ access to services, but digital infrastructure can also facilitate surveillance or control. The report, therefore, argues that digital technologies should not be treated as inherently enabling or harmful. Their effects depend on how they are designed, who controls them, the purposes they serve and the institutional and legislative environment in which they operate.

For funders and policymakers, this means beginning with a problem analysis rather than with a proposed technical solution. What barriers are migrants facing? Is a new digital tool actually needed? Could an existing solution be adapted? And how will success be defined from the perspective of the people who are expected to use it?

Make rights protection a design condition.

Responsible digitalisation also requires moving rights considerations upstream.

Data protection, non-discrimination, transparency and accountability cannot be checks carried out once a digital system has already been designed or procured. They need to shape decisions from the outset, particularly where tools contribute to high-stakes processes such as asylum procedures, registration, risk assessment or return. This includes basic principles such as collecting only the data necessary, clearly limiting how those data can be used, ensuring that automated or digitally assisted decisions can be explained and challenged, and maintaining meaningful human review.

Importantly, responsibility does not disappear when governments or international organisations purchase technology from private providers. Outsourcing a system does not outsource accountability.

Design with migrants, not simply for them

Another recurring gap concerns who participates in the development of digital tools.

In the inventory, migrants were predominantly listed as users or beneficiaries rather than developers or implementers, while community- and refugee-led initiatives accounted for only a small proportion of the tools identified. This matters because assumptions about what users need do not necessarily correspond to how digital tools are experienced in practice. Meaningful co-design can reveal obstacles that may otherwise remain invisible: language, affordability, connectivity, digital literacy, concerns about data use, or differing levels of trust in the institutions behind a tool. Gender deserves particular attention. The study found that most of the tools reviewed incorporated no gender-responsive design.  

Participation should therefore go beyond asking users to test a nearly finished product. Migrants, including groups most likely to encounter barriers, need to be involved in defining the problem, shaping the design, and evaluating whether the resulting tool works in their real-world circumstances.

Digital should not mean digital-only

One of the clearest lessons from practice is that technology rarely replaces human support. More often, it changes where that support is needed.

A platform may make information available online, for example, while community organisations, caseworkers, or diaspora networks continue to help people understand that information, determine whether it applies to their situation, and act on it. Digitalisation can consequently shift work towards precisely those intermediaries that often have the fewest resources. The report highlights this broader implementation challenge: infrastructure, institutional capacity, ownership and user trust are often as important as the technology itself.

Responsible digitalisation, therefore, means preserving assisted and offline routes rather than assuming that everyone can or should access services digitally. It also means funding the human infrastructure around digital tools. Community organisations and other intermediaries should not be treated as temporary workarounds until a system becomes fully digital; in many contexts, they are essential to making the system accessible.

Measure what changes, not just what is used

Finally, better evidence is needed.

Download numbers, registrations, and platform visits indicate whether a tool has been reached. They say much less about whether it has improved migrants’ access to services, reduced exclusion, protected rights or produced unintended harms. The evidence base remains surprisingly limited. Few of the tools examined in the study were backed by rigorous or independent evaluation, and there is particularly little differentiated evidence about how digitalisation affects different groups.  

A more responsible approach would therefore make independent, outcome-focused evaluation part of the lifecycle of digital initiatives rather than an optional final step. The results should also be shared. Without accessible evidence on what worked, what failed and under what conditions, organisations risk repeatedly developing new tools without learning sufficiently from those that came before.

From better tools to better governance

Overall, these points suggest that responsible digitalisation is ultimately less about identifying the “right” technology than about creating the right conditions around it. That means being explicit about purpose; making rights protection and inclusion design requirements; involving migrants meaningfully; investing in human and technological infrastructure; ensuring long-term institutional ownership; and evaluating effects on people rather than simply measuring technological uptake.

Digital innovation will continue to reshape migration governance. The task now is to ensure that the governance around that innovation develops just as quickly. 

The full GIZ-commissioned study, Meta-study on evidence regarding migrants’ rights in digital solutions in the field of migration, by Katrin Marchand and Lalaine Siruno, is available here.