Energy procurement strategies for commercial buyers

Procurement is where a large share of your energy spend is decided, often for years at a time, in a single signature. Energy is the single largest operating expense in commercial office buildings, about one-third of a typical operating budget, so the structure you sign for the commodity portion of that spend matters as much as any efficiency project you run.

Wholesale power is not a stable input. The composite of wholesale prices the U.S. Energy Information Administration tracks averaged about $40 per MWh in 2025, up roughly 7 percent from 2024, and that same measure hit $80 per MWh back in 2022. Gas is the usual driver: the Henry Hub benchmark averaged $3.52 per MMBtu in 2025, about 56 percent more than in 2024, and some northeastern trading hubs averaged roughly twice as much in 2025 as the year before. Your procurement structure decides how much of that movement lands on your budget.

The three structures every buyer should understand

Almost every commercial supply offer is a variation on three ways of pricing the commodity. They differ in one thing: who carries the risk of the wholesale market moving after you sign.

Fixed price

You agree on a single price per kWh for the full term, and the supplier absorbs wholesale movement. This buys budget certainty, which is why it suits organizations that value a predictable number over the lowest possible average cost. The trade is that the supplier prices in a risk premium, and you give up any benefit if the market falls during your term.

Index or pass-through

Your commodity price floats with a published wholesale index, month to month or even hour to hour. Over a full cycle this often produces the lowest average cost because you are not paying a hedging premium, but you carry the full weight of any spike. For a buyer without the appetite or the cash flow to absorb a bad month, straight index pricing can be punishing.

Block and index

A middle path. You lock a fixed price on a set block of your load, often your predictable baseline, and float the remainder on the index. A common shape is to fix a large share of expected volume and leave the rest exposed, so you set how much certainty and how much market upside you want. Buying in blocks over time also smooths out the risk of timing one purchase badly.

Match the structure to your risk tolerance, not the sales pitch

There is no universally correct answer, only a fit. Start from two questions: how much month to month variance can your budget absorb without a difficult conversation, and how much of your load is genuinely predictable? A buyer with tight budget tolerances and steady load leans fixed. A buyer with room to absorb variance and a mandate to minimize average cost leans index. Most land somewhere in between, which is exactly what block and index is for.

Remember the part of the bill you cannot procure

A supply contract only prices the commodity. Generation is the largest single component of the price of electricity, but delivery, transmission, regulated riders, and demand charges are set by tariff and pass straight through no matter how you buy. It is common to negotiate hard on the commodity while ignoring the regulated charges that make up the rest of the bill, where rate-class and tariff optimization often finds more durable savings than the supply deal itself.

What a good procurement decision needs from your data

Every structure above depends on knowing your load. To size a fixed block you need a clean volume history. To judge index exposure you need to know when you consume, not just how much. To compare offers on equal terms you need each one normalized against the same usage assumptions. That is a data problem before it is a market problem.

• A clean 24 to 36 month usage history, with estimated reads and gaps identified
• An hourly or interval load shape, so peak-heavy consumption is visible
• Delivery and regulated charges separated from the commodity, so offers compare fairly
• A single normalized view across every account, not a spreadsheet per site

This is the work MartinAI removes. We read every bill and interval file, structure the fields, flag the estimates and gaps, and give you one clean load history to take into a procurement conversation, so the decision rests on your real consumption rather than a supplier's assumptions.

Common procurement mistakes

• Signing fixed at the top of the market because a spike made certainty feel urgent
• Choosing straight index without the cash flow to survive a bad quarter
• Comparing offers built on different usage assumptions, so the cheapest quote is not actually cheapest
• Negotiating only the commodity and ignoring the regulated charges that pass through untouched
• Letting a contract auto-renew because the load data needed to re-shop was never assembled