The credibility of any award lives or dies in its judging room. This page publishes exactly how the World Public Service Excellence Awards evaluates every entry: the five criteria and their weightings, the structure of the jury, the anonymisation process, and the safeguards that keep commercial interests out of results.
All categories — from customer service awards to public administration awards — are scored against five pillars. Weightings vary slightly by category family and are published in each criteria pack before nominations open.
| Criterion | Typical Weight | What Jurors Look For |
|---|---|---|
| Citizen Impact | 30% | Verified outcomes for the people served: satisfaction, access, speed, fairness, wellbeing. Before-and-after data with credible measurement. |
| Innovation | 20% | Novelty appropriate to context — new models, technology or partnerships — judged by what it changed, not how futuristic it sounds. |
| Efficiency | 20% | Better use of public money and time: unit-cost movements, productivity, waste removed, demand prevented. |
| Inclusiveness | 15% | Whether improvements reached the hardest-to-serve: accessibility, language, digital exclusion, geography, equity of outcome. |
| Sustainability | 15% | Institutional durability: governance, funding, skills and culture that keep results alive beyond one leader or budget cycle. |
After the deadline, the secretariat removes identifying branding where feasible and allocates each entry to a regional panel of at least five jurors with relevant domain expertise. No juror receives an entry from their own country or from any organisation they have worked with — conflicts are screened against the published register.
Jurors score independently against the five criteria before seeing any colleague’s marks, recording written rationale for each score. Statistical screening flags outlier scoring patterns for review, protecting entrants from a single harsh or generous juror distorting results.
Category panels convene to debate shortlists, and cross-regional moderation aligns standards worldwide. Top-ranked entries undergo claim verification — source data checks and referee calls — before winners are confirmed and locked by the Awards Integrity Board.
Our jury is built for competence rather than headlines. It comprises former ministers and permanent secretaries who have run institutions, city managers who have balanced budgets in public, academics in public administration and service design, senior advisers from international organisations, and prior winners who know exactly what excellent looks like from the inside.
Jurors serve a maximum of three consecutive cycles to keep perspectives fresh, complete calibration training each year using benchmark entries, and sign the integrity code covering confidentiality, conflicts and gifts. Their names are published; their individual scores are not — protecting candour while preserving accountability through the moderation record.
For entrants, the practical meaning is simple: your entry will be read by people who have done comparable work, and your feedback letter will reflect a professional peer review. Finalists frequently describe the written assessment as the most useful external evaluation their service has received — one reason organisations re-enter these excellence awards year after year regardless of results.
A selection of the 2026 jury. The full jury list and conflict-of-interest register are published before judging begins.
Award programmes fail in predictable ways: pay-to-win economics, sponsor influence, opaque criteria, judges marking friends. The World Public Service Excellence Awards is engineered against each failure mode, and we publish the engineering. Nominations are free, so no revenue depends on entry volume in any category. Sponsorship is sold only against the gala and content — sponsors never see entries, never meet panels during judging, and appear nowhere in the scoring chain.
Criteria and weightings are frozen and published before nominations open, so no entrant competes against a moving target. Every juror conflict is declared in a public register. And the Awards Integrity Board — independent of the secretariat — audits score distributions annually, hears appeals within 21 days of results, and holds the power to withdraw an award if material claims are later disproven. That power has existed since founding precisely so it rarely needs using.
Criteria documents are written for precision; entries are scored by humans applying them at pace. Understanding how jurors actually read closes the gap between a technically compliant entry and a high-scoring one. Begin with weightings as a time budget: citizen impact carries thirty percent of the score, so it deserves roughly a third of your words, your strongest evidence and your opening paragraph. Entries that spend a thousand words on project history before reaching outcomes force jurors to hunt for the scoring material — and a juror reading their fourteenth entry of the evening should never have to hunt.
Each criterion has an implicit question and an implicit trap. Citizen impact asks who is better off and how do you know; its trap is activity data — calls answered and systems launched are inputs, not outcomes. Innovation asks what changed in the model; its trap is technology worship, describing the tool rather than the redesign of the service around it. Efficiency asks what was released — money, hours, capacity; its trap is savings claimed without baseline costs. Inclusiveness asks who might this have missed; its trap is the unexamined average. Sustainability asks what survives leadership change; its trap is enthusiasm offered as governance.
Jurors also score confidence in your evidence, implicitly. Independent sources — audit offices, statistical agencies, commissioned evaluations, regulator data — anchor a score; internal dashboards support it; unattributed claims erode it. One verified number outweighs five asserted ones.
The written rationale jurors record alongside every score becomes your feedback letter, and it follows the criteria in order — which means you can pre-empt it. Before submitting, score your own draft: give each criterion a mark out of ten and force yourself to cite the sentence and the evidence file that justify it. Sections you cannot self-score above six are sections the panel will not score above six either; strengthen or honestly frame them. Teams that run this exercise with someone outside the programme — a finance colleague, an internal auditor, anyone professionally suspicious — routinely gain two or three final points, which across a 2,400-entry field is frequently the distance between the long list and the shortlist.
Two structural notes complete the picture. Moderation means your entry is ultimately measured against the global field, not only your panel’s batch: standards are aligned across regions in the cross-regional stage, which is why a small administration with rigorous evidence can and does outrank famous institutions with general claims. And verification means your top-ranked entry will be checked: source documents requested, referees called. Nothing in your entry should be a sentence you would hesitate to show your auditor general — not because the process is adversarial, but because the trust governments place in these results is built one verified claim at a time. Write to that standard and the criteria stop being an obstacle; they become the clearest brief your programme’s story has ever had.
From the entrant’s side, judging is a quiet season — and knowing the schedule turns silence from anxiety into information. In the first week of October, every entry receives confirmation of anonymisation and panel allocation; no content questions are asked of entrants from this point, so an inbox without messages means the process is simply running. Through October, jurors score independently: each entry is read in full by at least five panellists, scores and written rationale are recorded, and the statistical screening reviews distributions for outliers. Entrants whose referees are contacted for verification should brief them simply to answer plainly — referees confirm facts, they do not advocate.
In late October, category panels convene to debate provisional shortlists, and the cross-regional moderation aligns standards across the global field. Finalists are notified by email on 26 October, hours before public announcement, with embargoed materials for their communications teams.
Through November, verification completes on top-ranked entries — source documents requested through the portal, referee calls scheduled across time zones — and the Awards Integrity Board reviews the full results record before locking outcomes. Winners are not informed in advance: results are revealed live at the gala on 11 December, which finalists consistently describe as the point of attending. Within thirty days of the gala, every shortlisted entrant receives the written jury assessment, and the appeals window runs for twenty-one days from results for any entrant who believes process was not followed.
The calendar’s discipline serves a principle: every entrant, from the largest ministry to the smallest municipal team, experiences exactly the same process on exactly the same dates. No early results for the well-connected, no extensions for the prominent, no private previews for sponsors — the schedule is the same because the standard is the same, and the standard is the product.
Download category criteria packs inside the entry portal and write to exactly what the jury will score.
