How to anonymise judging data without losing the insights that matter

Of course, awards program teams do not set out to expose personal data. But without a clear approach to anonymisation, even well-intentioned reporting can surface more than it should.

Program managers sit on a goldmine of data. Submission materials, scores, judging comments—all of it holds real potential for analysis and program improvement.

At the same time, pressure to handle personal data responsibly has never been greater. But this is not an impossible challenge. Respecting data privacy and producing meaningful analysis are not in conflict. It simply comes down to how you handle the data.

The tension between transparency and privacy

Awards programs generate data at every stage. Entrants submit applications, judges assign scores and leave comments, and programs track deadlines and activity throughout the cycle. All of that is valuable for internal quality assurance as much as for reporting to boards, sponsors or the public.

The problem arises when raw data makes it possible to identify individuals. A judge whose scoring pattern is visible in a report, or entrants whose names appear in aggregated outputs are scenarios that undermine the principle of data minimisation and erode trust among everyone involved.

At the same time, blanket data restrictions are not the answer. Without analysis, program managers lose the benchmarks that inform good decisions: Which categories attract the most applicants? Where do judging scores diverge significantly? What has changed since last year?

The solution is targeted anonymisation: the structured removal of personally identifiable information before data enters any analysis or reporting workflow.

What anonymisation actually means in practice

Anonymisation goes well beyond redacting names. Under the GDPR, data is only considered truly anonymous when identifying an individual is no longer reasonably possible, including through the combination of multiple data points.

For awards programs, that means addressing four key areas:

Remove personal identifiers. Names, email addresses, phone numbers and similar details should be stripped from any dataset used for analysis. Internal IDs or anonymised codes can replace them without any loss of analytical value.

Aggregate results rather than individualise them. Instead of reporting on a single judge’s scores, report on the average across all judges within a category. Trends remain visible; individuals do not.

Give small groups extra protection. When a category contains only two or three entries, results can easily point back to specific people. Consider applying a minimum threshold before publishing data at that level, or grouping smaller categories together.

Handle free-text comments carefully. Qualitative feedback from the judging process is especially sensitive. Comments can reveal personal views, writing styles or indirect identifiers. For reporting purposes, they should either be omitted entirely or summarised at a high level.

Practical tips for program managers

1. Map your data flows from the start
Deciding early on which data is collected and for what purpose saves significant effort later. A simple overview covering what is needed for operations, what goes into reports and what is archived helps establish sensible boundaries from the outset. This aligns with the principle of data protection by design and by default, as required under GDPR.

2. Focus analysis on categories and time periods
Rather than drilling into individual results, look for patterns at the program level. Questions like “Which category had the highest average judging quality?” or “How has participation changed over three years?” deliver genuine insight without requiring any personal data. Our guide on metrics that matter for awards explores this further.

3. Apply role-based access consistently
Not everyone working on a program needs access to the same information. Judges should see the submissions assigned to them, not the scores of their fellow panellists.

Program administrators can work from aggregated reports rather than raw data exports. This separation supports both data privacy requirements and the integrity of the judging process.

For a deeper look at governance and access controls, see Governance, risk and control: Integrated safeguards in award programs.

4. Review exports before sharing
Before reports are sent to external stakeholders, such as sponsors or media partners, a manual or systematic check should always take place. A short anonymisation checklist makes it easy to catch anything that might have been missed.

Balancing privacy and insight with Award Force

Award Force is built with the understanding that privacy and functionality need to work together. The platform’s reporting tools allow scores to be analysed at category level without surfacing individual juror data. Averages, distributions and variances are all accessible through aggregated views.

Award Force’s configurability means access rights can be set precisely for each role. Who sees what data, and at what level of detail, is fully within the control of the program team. This is particularly important when external judges or multiple organisations are involved in the same program.

Award Force also upholds security and data protection standards in line with GDPR requirements, giving programs worldwide a reliable foundation, both technically and legally.

Data privacy as a mark of quality

Taking data privacy seriously is a demonstration of professionalism and respect: for entrants, for judges and for the program itself. Organisations that protect the trust placed in them also protect the credibility of their results.

Well-anonymised data still delivers meaningful insight into trends, strengths and areas for development. Thoughtful analysis makes the essentials visible – what the program achieves and where it is heading – without compromising the privacy of the people who made it possible.