Public data register

    Accountability

    Corrections log

    When we get a published figure wrong we correct it and record it here. Price-cap corrections are read straight from the change log that backs the figure, so this page cannot drift from what actually changed.

    Price-cap figure changes

    • 23 September 2026
      Q4 2026 (Oct-Dec)

      £1,663 corrected to £1,723

      Ofgem Oct–Dec 2026 cap for a typical direct-debit household (+4%, announced 26 Aug 2026). Previous cap is the Jul–Sep figure on Ofgem's revised typical-use basis (£1,663), not the £1,862 old-basis figure, so the two are like-for-like.

      Source: Ofgem

      Inspect the price-cap record
    • 18 July 2026
      Q3 2026 (Jul-Sep)

      £1,641 corrected to £1,663

      Corrected homepage/forecast to match Ofgem's published Jul–Sep 2026 cap for a typical direct-debit household. Note: £1,862 is on Ofgem's old typical-use basis; on the revised basis used from 1 July 2026 the Jul–Sep cap was £1,663. Previous cap corrected 24 Oct 2026 from 1680 to £1,641 (Ofgem's Apr–Jun 2026 cap, old typical-use basis, per Ofgem press release 27 May 2026); the Jul–Sep change was a 13% like-for-like rise.

      Source: Ofgem

      Inspect the price-cap record

    Editorial corrections

    • Placeholder supplier labels ("Supplier A" to "Supplier F") were removed from the supplier performance report charts and replaced with the named suppliers and their actual scores.

      Anonymised placeholders had been published in charts intended to show real, attributable supplier performance.

      Evidence: src/data/research.ts — supplier performance report chart series

    • A data period (H1 2026) and a downloadable scoring table were added to the supplier performance report.

      The report cited general source pages but not the dated datasets or the calculation sheet behind each score.

      Evidence: src/data/research.ts — dataPeriod and dataTables on the supplier report

    • References to "National Grid ESO" were corrected to NESO (National Energy System Operator) across methodology and source listings.

      The organisation was renamed, so the old attribution was inaccurate.

      Evidence: src/components/Sources.tsx — COMMON_SOURCES.neso

    • The forecast page was reframed as a 12-month outlook with explicit confidence levels rather than a definitive prediction.

      The earlier framing implied more precision than a modelled outlook can support.

      Evidence: src/pages/EnergyPriceForecast.tsx

    • Numerical estimates for future cap periods, and their high/medium/low confidence labels, were withdrawn from the outlook page. The page now shows only the cap periods Ofgem has confirmed, and states plainly that no verified forecast is available for later periods.

      The estimates were written by an automated weekly job with no recorded inputs and no check by a person, so they could not be reproduced or evidenced. The confidence labels reflected how far ahead a period was, not measured accuracy.

      Evidence: src/data/forecastEvidence.ts — FORECAST_EVIDENCE; src/pages/EnergyPriceForecast.tsx

    • The outlook page now shows Ofgem's current headline figure of £1,663 for 1 July – 30 September 2026 and the confirmed £1,723 for 1 October – 31 December 2026, alongside the equivalent figures on the older 2023 consumption values (£1,862 and £1,935).

      Ofgem reduced its typical consumption values in 2026, so the headline annual figure changed while the rates did not. Showing one basis without the other made the site's figure look inconsistent with Ofgem's published headline.

      Evidence: Ofgem press release, 26 August 2026 — https://www.ofgem.gov.uk/press-release/energy-price-cap-will-rise-4-october-2026

    Where a correction predates the change log, no date is shown rather than an estimated one.