Sectors
Energy operations for hospitality and retail
How multi-site hospitality and retail operators can turn a sprawl of meters, contracts and trading-hours load into a managed energy system.
Last updated
Hospitality and retail operators buy energy in the hardest configuration the market offers: many sites, modest consumption at each, extended and irregular trading hours, and constant estate churn as sites open, close, refit and change hands. The market prices this configuration well for itself. Contracts fragment across suppliers, renewals arrive site by site, and every gap in attention becomes deemed rates, rollovers or out-of-contract drift. No single site loses enough to trigger action, and the estate loses steadily.
The consumption shape adds a second layer. Trading pushes load into evenings and weekends, where wholesale and network pricing move most, while refrigeration, cellar cooling and standby equipment run around the clock. An estate like this on undifferentiated flat-rate contracts is paying an averaged price for a decidedly unaveraged pattern, in whichever direction the mismatch runs. Without half-hourly data nobody knows which direction that is.
The portfolio failure mode
The typical estate has grown by acquisition and lease, and its energy arrangements record that history rather than any decision. Different suppliers per site, contract end dates scattered through the year, some supplies still in a previous tenant’s name, and a broker relationship somewhere in the past whose commission is still embedded in a unit rate. The first act of energy operations is simply to make the estate visible: every meter, every contract, every end date, every rate, in one place. Most operators are genuinely surprised by what this inventory turns up.
One London pub is a fair example. Running a busy seven-day kitchen serving over 300 covers a day, it was still billed on a legacy Economy 7 day/night tariff that no longer matched how the site actually operated. The inventory also caught a kitchen electrification already planned: cooking moving off gas and onto roughly 42.6 kW of new electrical load, set to take annual electricity consumption from around 85,000 kWh to something closer to 150,000 kWh. None of that load change had been priced into the pending renewal. The existing contract carried no exit fee, so re-procurement was never actually gated by its stated end date. Quote a new supply against today’s consumption on a site like this and it is under-priced from day one; the load change has to be modelled before the tariff is chosen.
Load shape is the second lever
With the estate visible and half-hourly data flowing, the structural questions open up. Whether trading-hours load suits time-of-use or pass-through structures. What the overnight baseload is costing and how much of it is refrigeration that must run versus equipment that need not. Where kitchens and cold rooms create demand peaks that capacity arrangements should reflect. For sites with daytime trade and usable roofs, whether solar clears the bar on real data.
The pattern differs by format, and the tariff structure has to differ with it. A late-trading bar or restaurant pushes most of its demand into the evening peak, exactly when wholesale and network pricing is highest, which makes a flat-rate contract a poor fit and a time-of-use or day-ahead structure worth testing against the site’s actual hours. A convenience store or quick-service site with steady daytime trade and heavy refrigeration carries a flatter profile, where the bigger opportunity often sits in the baseload itself rather than the time-of-use question, because refrigeration and standby equipment run whether the till is open or not.
Grouping sites by load shape, not by brand or format label, is what makes this tractable across an estate of any size. Two sites trading under the same fascia can carry different profiles because of layout, kitchen equipment or footfall pattern, and a tariff structure chosen for the format rather than the meter data will suit some sites and cost others money. The half-hourly data groups sites correctly; the format name does not.
Run it as one system
The estate changes constantly, so a one-off procurement exercise decays from the day it finishes. What holds is a loop: measure every site, model the options, act on the outliers, monitor the result, repeat. That loop is a role, not a project, and it is a role hospitality and retail businesses do not have and should not need to hire for.
If you operate a multi-site estate, start by seeing it clearly. Request a benchmark at /benchmark or book a review at /book.
Frequently asked questions
Why are hospitality and retail energy costs so hard to control?
Because the estate is many small decisions rather than one big one. Each site has its own meters, its own contract history, and often its own inherited supplier from a previous operator or landlord. Trading hours drive consumption into evening and weekend periods where pricing varies sharply, and refrigeration and catering create baseload that runs whether the site is trading or not. Control requires portfolio-level visibility that site-level billing never provides.
Does energy management interfere with trading?
The substantial gains do not come from asking sites to switch things off during service. They come from structural decisions made above site level: matching tariff structure to when the estate actually trades, correcting contract positions, reviewing capacity, and identifying the out-of-hours baseload that serves no one. The operational ask on site teams is close to zero, which is precisely why it gets done.
Which of our sites should get attention first?
The data decides. Half-hourly consumption across the estate identifies the outliers: sites with abnormal baseload, sites on structures that punish their trading pattern, and supplies that have lapsed out of contract. Working the outliers first produces most of the early value, and it replaces the usual approach, which is renewing everything at once on whichever quotes arrive.