Skip to content
SME AI Impact

Open Research Asset

Structured dataset & data dictionary

The full dataset powering this showcase is published openly in CSV and JSON formats for researchers, journalists, and enterprise practitioners. This dataset is ungated and free to download.

Dataset summary

Records

30

15 Innovation + 15 Adoption

Sources cited

38

Across 4 evidentiary tiers (T0–T3)

Coverage benchmark

94%

Average across 10 evaluation dimensions

Licence

CC BY 4.0

Open attribution licence

Data dictionary

Definition and typing for each field in the CSV export.

Field nameTypeDefinition & usage
idintegerNumeric index of the record ordered canonically by category and listing.
cohortstringCollection cohort: "award-2026" (sealed 30 award winners), "practitioner" (Maple Commerce deployments), or "community" (verified non-award SME submissions).
provenancestringOrigin tier: "official-award" (IMDA & SBF citation), "maple-commerce" (implementer-built), or "community-submission".
disclosed_implementerstringNamed implementer entity where conflict of interest is disclosed (e.g. "Maple Commerce"), or empty.
namestringOfficial registered entity name as cited in the IMDA/SBF award announcement or client submission.
slugstringUnique URL slug used for permalink routing on this showcase.
categorystringAward category: "innovation" (developed proprietary AI) or "adoption" (adopted ready-to-use COTS solution).
sectorstringIndustry classification (e.g. Business Services, Manufacturing, Logistics, F&B).
useCasestringFunctional application domain (e.g. Workflow automation, Document processing, Computer vision).
aiTypestringPrimary technical architecture archetype (e.g. LLM application, Agentic workflow, Computer vision).
headlinestringOne-line summary of implementation and primary business outcome.
confidence_gradestringAssigned evidence grade: A (independently supported), B (canonical source supported), or C (limited public evidence).
coverage_pctintegerResearch coverage score (0–100%) across 10 structured evidence dimensions.
official_solutionstringNamed AI solution or platform from the official award citation.
primary_metric_labelstringDescription of the headline published performance metric.
primary_metric_valuestringPublished quantified outcome value (e.g. "80% reduction", "2.5 hours/day saved").
all_metrics_countintegerTotal number of documented metrics for the winner in the dataset.
stack_layers_countintegerNumber of discrete technical stack layers documented in the case study.
named_vendorsstringSemicolon-separated list of disclosed third-party software/cloud vendors, or "None named".
sources_countintegerTotal number of cited references and public documentation sources.
sources_beyond_t0integerNumber of independent third-party sources (news, client profiles, partner case studies) outside the canonical award citation.
company_urlstringVerified official website URL of the winning company.
researched_atstring (ISO)Timestamp of initial research extraction and baseline verification.
updated_atstring (ISO)Timestamp of latest evidence verification or citation upgrade.

Licence and attribution note

This dataset is licensed under the Creative Commons Attribution 4.0 International (CC BY 4.0) licence. You are free to share, copy, redistribute, and adapt the material in any medium or format for any purpose, including commercial applications, provided you give appropriate credit.

Recommended citation: Wong, I. (2026). Singapore SME AI Impact Awards 2026 Dataset. BestCourses.sg. Available at: https://bestcourses.sg/sme-ai/data/

Note: Company names, award citations, and trademarked terms remain the intellectual property of their respective owners and the award organisers (IMDA & SBF).