Which channels produced signed supply and works contracts?

A commercial supply contract is frequently signed six months before the meter switches, and then bills for three years. The campaign that produced the account is credited for none of that.

Because the match is on the account, revenue credits the original channel whenever it bills — which for this category means the marketing budget can finally be compared against contracted revenue rather than inquiries.

Brokers and consultants introduce a substantial share of commercial energy contracts, at a commission that is accepted as a cost of doing business rather than measured against alternatives.

Added to the lead file with each broker as the source, broker introductions rank on contracted revenue alongside direct marketing, which makes the trade-off between commission and acquisition spend an arithmetic question rather than a cultural one.

Every supply contract has an end date, and in the months before it the account is in play again. Whoever holds it spends to keep it and everybody else spends to take it. The economics of that period are nothing like a new acquisition: the relationship exists, the consumption history exists, and the decision is usually taken by somebody who has been through the process at least once already. Marketing budgets rarely reflect any of it, because retention spend is booked as account management and never appears in a channel report at all.

Exporting the renewed contract as a further row against the same account contact puts it back into the picture. It credits the channel that produced the account originally, which is the honest reading — the renewal only existed to be won because a campaign or a broker introduction created the relationship years earlier. A channel that produces accounts which renew twice is worth several times one whose accounts leave at the first opportunity, and nothing in an inquiry count can separate the two. Contract end dates in the export make the difference readable directly.

The same file goes some way toward a question suppliers rarely put to themselves: which acquisition sources produce accounts that churn on price. Churn is not computed and an end date is not read, but every renewal row credits the channel that won the account, so over a range of several years a channel whose accounts renew keeps accumulating contracted revenue while one whose accounts leave at the first price rise stops. Read over a short range and then a long one, the ranking may change, and the long one is what should be setting next year's acquisition budget. It is also the argument that wins a retention budget, which can be harder to get approved than an acquisition one.

A commercial energy business often sells two things with almost nothing in common. A supply contract is a procurement decision, taken on price and term by somebody in finance, and it bills for years on consumption. An efficiency project — lighting, controls, heat pumps, metering — is a capital decision taken by an engineer or an estates manager, quoted, specified and installed. The inquiries look similar at the top of the funnel and the revenue behaves completely differently at the bottom. Reporting them together produces a blended figure that neither side of the business recognizes as its own.

A product column mapped as Product / service shows supply and works as separate lines, each with how much of it traces to a channel, and that alone can change where money goes. Search terms producing supply inquiries are competitive, well understood and cheap per click relative to the contract value behind them. Terms producing works inquiries are narrower, more expensive and convert into orders that are larger but far rarer. Funding both against one cost-per-lead target means the cheaper one wins the budget every year regardless of which is producing the revenue. That is not a judgment anybody made; it is what happens when a single average is asked to govern two different markets.

There is a second-order effect worth knowing about. Works projects can open a supply relationship, and supply relationships can produce works inquiries, and because later purchases credit the channel that produced the account, a works order from an existing supply customer is credited to whatever won the supply contract. Over a range long enough to contain both, a channel's total therefore includes the second product its accounts went on to buy, even where the product breakdown shows that channel's own line as thin. Without that long range, the cheaper half of the business can quietly look as though it generated nothing beyond itself. Cutting the wrong one of the two tends to show up much later, in the other product's numbers.

Billing is the most reliable record a supplier has. Every account that consumed anything was invoiced, the invoice carries an account contact, an amount and a date, and the whole lot was retained for audit and regulatory reasons long after anybody stopped caring about the campaign behind it. That makes the revenue side of this exercise a report somebody already runs rather than a new extract with a new approval attached. It is also why a finished year is as workable as the current one: those records were kept for billing, not for tracking, and they do not expire.

The demand side is patchier and always will be. Where an inquiry came from was recorded by whoever happened to care at the time — an advertising platform, a call log, a broker's introduction email, a list of everybody who attended a trade event. Several of those combine into one lead file, each row tagged with its source, and each one moves rows out of the unattributed column. What none of them can do is reconstruct a source nobody wrote down, and the report does not pretend otherwise. That residue stays in Direct / Unknown and is never shared out across the sources you did manage to supply.

The practical consequence is that the first report is worth running before any process changes. It tells you how much of the book is traceable at all, which is the number that decides whether tightening inquiry capture is worth the effort. The gap may well be unevenly spread: one route into the business records its source diligently and another records nothing, and knowing which is which is more actionable than any channel ranking on the first page. Fixing that one capture point is usually cheaper than the media budget being argued over.