Methodology

How BARZ Research & Intelligence publications are prepared and checked, where the data comes from and how errors are corrected.

Sources

Official statistics, regulations and standards, primary reports and technical documentation come first. Open data, commercial offers, interviews, client data and calculated estimates are kept apart and labelled. Every value is held in the observation database with its metric, sector and country, value, unit and period, publication and collection dates, the source and its supporting extract, the assumptions and calculation method, its verification status and its update deadline. Costs keep their currency, date, capacity, specification and scope; non-equivalent offers are not compared as interchangeable. Missing values are never written as zero; versions are kept. Client data is segregated from public research.

Source register

Automated quality controls before release

Before release the code checks:

• support for material factual claims
• whether sources actually substantiate those claims
• dates, definitions, geography and reporting boundaries
• formulas, arithmetic and units — formulas are recalculated in code
• comparability of costs and operating conditions
• consistency between narrative, tables and charts
• conflicting evidence
• separation of fact, estimate and forecast
• whether conclusions follow from the evidence
• confidentiality, permitted reuse and duplication

An internal evidence map links material claims to sources and calculations. Publication is blocked when a critical claim is unsupported, a calculation fails, a material contradiction remains unresolved or the inputs cannot support the conclusion. Model-generated confidence scores are never used in place of these controls; automated checks reduce errors but do not establish independent expert approval.

Publication standard

Each analytical brief or study states the decision question and intended audience, an executive summary, the evidence and methodology, original calculations where relevant, alternative options or explanations, scenarios and sensitivity where applicable, limitations and missing data, practical next steps, and the sources, data period, version and review status. Each study identifies the conditions under which its conclusions would change. Monitoring signals give a source, the event date, an explanation of relevance and limitations.

Expert review of commissioned research

Commissioned research is reviewed by a named BARZ employee or external specialist with relevant competence, who checks the methodology, critical evidence, assumptions, calculations, applicability and recommendations. Comments and their resolution are recorded against a specific report version. The BARZ project lead remains responsible for scope and delivery.

Status, updates and corrections

Each publication shows its original publication date, substantive update date, underlying data period and status: current, update due or archived. Dates are never changed merely to create apparent freshness. For a material error, promotion is paused, the affected pages and files are corrected and a dated correction note is published.

Corrections log

• 3 November 2026 — Compressed-Air System Leakage: Measuring It and Pricing the Exposure in the UAE, Oman and Saudi Arabia: retitled from a single-leak framing to aggregate compressed-air system leakage, and reframed the kWh and cost figures as an explicit upper bound that assumes power scales directly with air output, not a recoverable-savings estimate.
• 3 November 2026 — Compressed-Air System Leakage: Measuring It and Pricing the Exposure in the UAE, Oman and Saudi Arabia: added the Sourcebook's load/unload unloaded-power range and explained why a leak-rate percentage does not convert one-to-one into an electricity-cost percentage, since recoverable savings depend on the compressor's control type and operating profile.
• 3 November 2026 — Compressed-Air System Leakage: Measuring It and Pricing the Exposure in the UAE, Oman and Saudi Arabia: replaced three tariff inputs that had been retyped as assumptions (a sourced rate re-entered as a converted value with the observation named in the note) with direct references to the underlying tariff observations, so their source, scope and applicability are checked.
• 3 November 2026 — Compressed-Air System Leakage: Measuring It and Pricing the Exposure in the UAE, Oman and Saudi Arabia: corrected the Oman scenario, which had used an annual consumption above the flat tariff's own stated Cost-Reflective Tariff threshold; the tariff observations now carry their consumption boundaries as checked fields, and the illustrative scenario was reset to satisfy all three boundaries at once.
• 3 November 2026 — Compressed-Air System Leakage: Measuring It and Pricing the Exposure in the UAE, Oman and Saudi Arabia: labelled the UAE scenario Dubai (DEWA) and added that scope to the underlying tariff observation; removed an internal reference to the Directorate's assignment and to this desk's database from Evidence and methodology.
• 3 November 2026 — Compressed-Air System Leakage: Measuring It and Pricing the Exposure in the UAE, Oman and Saudi Arabia: added a limitation that DEWA and SEC separately publish a lower Industrial-category tariff than the Commercial-category rate this brief prices, with its own sourced observations, and explained why this brief uses the Commercial rate instead.
• 3 November 2026 — Compressed-Air System Leakage: Measuring It and Pricing the Exposure in the UAE, Oman and Saudi Arabia: corrected the Executive summary's claim that the three priced results differ only by tariff; the customer category differs too, since Oman's Non-Residential rate explicitly covers industrial use by regulation while Dubai's and Saudi's rates used here are specifically Commercial, not Industrial.

Limitations

Publications are general analysis, not advice for a specific organisation. Conclusions depend on the evidence available, its period and the assumptions stated in each publication. Savings, contracts, approvals and financing are not guaranteed.