Case studiesFinancial research
Did Ørsted do what it said?
Ørsted, the Danish offshore wind company, has made dated commitments and forecasts for over a decade. Reliability API sampled its public documents from 2014 to 2026 and checked what it could.
Five promises and forecasts, and what happened
Reliability API checked 29 claims from Ørsted’s documents. These five are two promises and three forecasts, each with a date it could be checked against, chosen to show a mix of outcomes. Each line runs from when Ørsted said it to when it was due, and the mark at the end is the verdict. Select one to read the quote, the reason and every piece of evidence.
-
Certified sustainable biomassPromise, said 2017 Delivered on time
“In 2017, our target is that 60% of the biomass we buy during the year must continue to be certified as sustainable.”
Why, in plain words
An auditor-verified report gives 72% certified, above the 60% target.
The full reasoning
The 2017 status report on sustainable biomass, verified by the auditor NEPCon, gives a measured 2017 documented share of 72% for the Bioenergy & Thermal Power plants' forest biomass. All of it was certified, with none counted through alternative documentation (E1, E2, E3). That is above the 60% target, which matches the 2017 industry-agreement requirement (E4). The report was published by Ørsted, but the figure was checked by a third-party auditor and is a specific measured result, not a restatement of the target.
Evidence
- E1NEPCon Certificering ApS (verifier) / Ørsted Bioenergy & Thermal Power A/SJuly 2018
- E2NEPCon Certificering ApS (verifier) / Ørsted Bioenergy & Thermal Power A/SJuly 2018
- E3NEPCon Certificering ApS (verifier) / Ørsted Bioenergy & Thermal Power A/SJuly 2018
- E4Danish Industry Agreement presentation (Søren Larsen, EU4Energy Vienna Policy Forum)24 May 2017
-
2018 earnings forecastPrediction, said February 2018 Resolved false
“EBITDA (business performance) excluding new partnership agreements is expected to be DKK 12-13 billion in 2018 compared with DKK 12.7 billion in 2017.”
Why, in plain words
Earnings came in at DKK 15.0 billion, above the forecast range. "Resolved false" means the forecast was wrong, not that the news was bad.
The full reasoning
Ørsted's 2018 annual report states that EBITDA excluding new partnerships rose 18% to DKK 15.0 billion (E1). The Q4 2018 presentation separately identifies DKK 15.1 billion from new offshore partnerships, which confirms the excluded items (E2). DKK 15.0 billion is DKK 2.0 billion above the top of the predicted DKK 12-13 billion range, beyond the DKK 1.5 billion threshold, so the claim resolves false. Both sources are from Ørsted, but they are its specific reported financial results on the same basis as the guidance.
-
Sell half of Hornsea 1Prediction, said February 2018 Resolved true
“We still expect a 50% farm-down of Hornsea 1, either in H2 2018 or in 2019.”
Why, in plain words
Ørsted sold half of Hornsea 1 in 2018, inside the window it gave.
The full reasoning
An independent trade publication reports that Ørsted signed an agreement in September 2018 to sell half of Hornsea Project One to Global Infrastructure Partners (E2). The same report says the 50% divestment was completed in November 2018 (E1). Both dates fall within the H2 2018 to 2019 window, and the stake was exactly 50%, which meets the true criterion.
Evidence
- E1offshorewind.biz27 November 2018
- E2offshorewind.biz27 November 2018
-
Hornsea 1, the largest offshore wind farmPrediction, said 2017 Mostly true
“With capacity of 1.2GW, Hornsea 1 will be the world's largest offshore wind farm when commissioned in 2020, and our biggest investment ever.”
Why, in plain words
Hornsea 1 was fully running in May 2020 and was the largest offshore wind farm at the time. "Mostly" because the "biggest investment ever" part rests only on Ørsted's own presentation.
The full reasoning
An independent report for Ofgem states that Hornsea 1 was fully operational, with full export capacity reached on 4 May 2020, inside the 2020 horizon (E1). Ember describes Hornsea 1 as the world's first offshore wind project over 1 GW and says it was only later surpassed by Hornsea 2, which supports it being the largest at commissioning (E3, E4). The 'biggest investment ever' part rests only on the subject's own 2016 investor presentation (E2), and the evidence does not confirm the exact 1.2GW figure, only that it exceeded 1 GW. Those two gaps make the verdict mostly true rather than true.
The two model checks disagreed on this one, so it is flagged.
Evidence
- E1Grant Thornton UK LLP for Ofgem11 November 2020
- E2DONG Energy (Ørsted)27 April 2016
- E3Emberundated
- E4Emberundated
-
Coal exit by 2023Promise, said 2017 Resolved false
“We have decided to stop using coal altogether at our power plants from 2023.”
Why, in plain words
Ørsted set a coal exit for 2023. Danish authorities then ordered three units kept running during the energy crisis, and the exit moved to 2025. The verdict records the miss; the evidence records the reason.
The full reasoning
Ørsted's own 9M 2022 report says Danish authorities ordered it to continue and resume three coal- and oil-fired units until 30 June 2024, and that its former target of stopping coal by Q2 2023 was replaced by a 2025 target (E1, E2). The 9M 2023 report shows coal was still in use during 2023: coal use fell 33% but did not stop, and Studstrup unit 3 burned coal until Q2 2023 (E3, E4). The criteria say the reason for coal use does not change the test. Because coal stayed in material use through 2023 and the exit was moved to 2025, the claim resolves false.
Every verdict on this page is machine-generated and not yet reviewed by a person. “Resolved false” on a forecast means the forecast was wrong, not that the news was bad.
The scores, explained
A score is the share of checked claims of one kind that held up, from 0 to 100.
The range shows how sure that is. Few claims give a wide range; more claims narrow it.
-
About 84 in 100 factual claims held up, likely between 73 and 93, from 16 checked claims.
Factual claims 84 73 93 -
For promises, 60 in 100 held up, likely between 40 and 79, from 6 checked claims. Low confidence: too few to say much.
Promises 60 40 79 -
For predictions, 46 in 100 came true, likely between 27 and 65, from 7 checked claims. Low confidence: too few to say much.
Predictions 46 27 65
Scores from 8 documents, 2014 to 2026, sampled from 32 found. What a score means
For your coverage
The same report for any listed company you cover, by API, MCP server or bulk data. A new company takes 10 to 17 minutes; a stored one returns in under a second.