Degrees of Doubt: Weather-Normalising the 2026/27 Energy Budget

For most UK organisations, the winter energy budget rests on a deceptively simple assumption: that the weather will behave broadly as it usually does. Suppliers encode that assumption into volume forecasts, and buyers sign contracts that price those volumes. The difficulty is that 'usual' is a statistical construct, not a promise. A mild October or a sharp November frost can move consumption by several percentage points, and in a market where unit rates and capacity charges are already under scrutiny, that movement can be the difference between a budget that holds and one that requires a mid-year conversation with finance.

As the 2026/27 renewal season approaches, the task is not to predict the weather. It is to audit how weather risk has been priced, challenge the assumptions embedded in the offer, and set a defensible margin that protects the organisation without overstating likely costs.

Why degree days deserve a line in the budget

Degree days are the standard currency of weather-related energy demand. Heating degree days measure how far average daily temperatures fall below a base temperature, typically 15.5°C for gas heating in the UK; cooling degree days capture demand for air conditioning and process cooling as temperatures rise. Suppliers use historical degree-day data, often from a regional or national weather station, to normalise consumption and forecast volumes.

The problem is that degree-day data is an average, and budgets are not spent on averages. A site in Leeds does not experience the same weather as a site in London, and a warehouse with frost protection does not respond to temperature in the same way as an office with a gas boiler. When a supplier presents a volume forecast, it is usually built on a weather normalisation that may be national, regional or site-specific. The first audit question is therefore simple: which weather data has been used, and how closely does it match the operational reality of each meter?

Reading the supplier's weather assumption

Most offers will not state the degree-day assumption explicitly. It is buried in the volume forecast or the profile that underpins the unit rate. Buyers should ask for the normalisation methodology in writing: the base temperature, the weather station or regional dataset, the historical period used, and whether the forecast has been adjusted for known changes in occupancy, equipment or building fabric.

A forecast that uses a ten-year average may be reasonable, but it can mask a trend. If the last three winters have been milder than the long-run average, a supplier may have priced in a lower volume, which looks competitive now but could leave the buyer exposed if temperatures return to the long-run norm. Conversely, a supplier using an older, colder baseline may overstate volumes, inflating the budget and potentially triggering volume-tolerance penalties if actual consumption falls short.

Three forecast distortions to test

First, check the base period. A normalisation based on 2019 to 2023 will include the pandemic years, when many buildings operated at reduced occupancy. That can distort the weather signal. A more recent period may be more representative, but it may also capture an unusually mild winter. The audit should compare at least two base periods and ask the supplier to explain the choice.

Second, check the regional weighting. National degree-day data is a poor proxy for a portfolio spread across Scotland, the North West and the South East. Ask whether the forecast uses a regional dataset, and whether the weighting reflects the number of meters, the floor area or the historical consumption at each site. A portfolio with a large gas-heated site in the North should not be normalised against a southern average.

Third, check the cooling load. Cooling degree days are often ignored in UK business energy forecasts because the cooling season is short and the loads are smaller. But for data centres, retail chillers or manufacturing processes with temperature-sensitive cooling, a warm summer can add material cost. If the site has a meaningful cooling load, the forecast should include a cooling degree-day assumption, even if it is small.

Contract clauses that move with the weather

Weather risk is not only a forecasting issue; it can be contractual. Some flexible supply agreements include weather-linked adjustment clauses, volume tolerances that tighten in extreme weather, or pass-through mechanisms for imbalance costs. These clauses are not inherently unfair, because suppliers need to manage risk, but they should be understood before signature.

Ask for a worked example of how the clause operates in a severe winter and a mild winter. If the contract includes a volume tolerance of, say, plus or minus 10%, calculate the cost of exceeding it in a cold snap. If there is a weather-linked adjustment, identify the trigger, the reference dataset and the caps. The goal is not to eliminate the clause but to price it into the budget as a contingent cost.

Building a defensible weather-risk margin

A weather-risk margin is not a guess. It is a quantified allowance based on the sensitivity of the portfolio to temperature. Start with the historical relationship between degree days and consumption for each major meter. If a site's gas consumption rises by 8% for every 100 additional heating degree days, and the difference between a mild and severe winter is 300 heating degree days, the potential swing is 24%. That is the exposure.

Then decide how much of that exposure to hedge through the budget. A common approach is to budget at the long-run normal and hold a separate contingency for the difference between normal and severe. The contingency should be sized to the portfolio's actual sensitivity, not a flat percentage. For a gas-heavy estate, it may be larger; for an all-electric office with low heating demand, it may be modest.

Finally, document the assumption. A board-ready budget should state the degree-day baseline, the source, the sensitivity factor and the contingency. That way, if the winter is colder than expected, the conversation is about a planned risk, not a forecasting failure.

A pre-renewal checklist

Before signing the 2026/27 contract, confirm the degree-day methodology in writing; compare two base periods; test regional and site-level normalisation; include cooling load where relevant; model the contract's weather-linked clauses in mild and severe scenarios; and set a contingency sized to the portfolio's sensitivity. The weather will do what it does. The budget's job is to be ready for it.