Which channels produced signed cleaning contracts?

A $2,400-a-month contract held for three years is worth $86,400. A channel judged on the first month appears to cost more than it earns and gets cut before it has been understood.

Because the match is on the account, every month of billing credits the channel that won it, which is the only ranking that reflects the economics of a recurring service business.

Contracts churn, and channels differ in the quality of the accounts they produce. Some produce price-driven clients who leave within a year; others produce buildings that stay for five.

With contract end dates in the export, average contract life can be read by channel — which usually matters more to the marketing budget than acquisition cost does.

Contract cleaning runs on thin margins, and the recurring schedule is the least profitable part of it. The money is in periodics: deep cleans, carpet and hard floor work, high-level and window cleaning, post-construction handovers, reactive call-outs after an incident. Those jobs are quoted against an existing account, invoiced separately, and almost never appear in any conversation about which channel produced the client. The report stops at the contract and the profitable half of the relationship stays invisible. A channel that opens accounts in buildings needing periodic work is worth far more than one opening accounts that never buy beyond the schedule.

Exporting periodic invoices as further rows against the same account contact puts that revenue where it belongs. It credits the channel that won the client, which is the only reading that matches how the business earns its margin. A column marking each row as the initial contract or periodic and extra work, mapped as Product / service, then shows the two as separate lines, each with how much of it traces to a channel — so the periodic revenue a contract eventually produced is visible beside the value that was signed. It is a column many cleaning businesses already carry for invoicing purposes.

There is a related trap in how contract value is quoted. A monthly contract figure is a forecast, not an invoice, and sites change constantly: hours get cut in a tight year, a floor comes out of scope, an extra day is added when the client's headcount rises. Reporting on the signed figure freezes a number that stopped being true in month four. Exporting what was actually billed anchors the ranking to money that arrived, and it makes contraction as visible as growth. Both happen to every contract cleaner and only one of them usually gets reported upward.

Cleaning is bought by a person with a job title and a budget, not by a site. A facilities manager, an office manager, a procurement lead at a managing agent — that individual signs, renews, expands and eventually leaves. When they move to another organization they frequently take the supplier with them, which is the most valuable and least measured phenomenon in the industry. Nothing in a lead report can see it, because the new organization is a new account with no history attached to it. The person is the same, and so, often, is the contact detail.

Matching on the contact rather than on the company means a buyer who moves and buys again links back to whatever produced them originally. Where they change email domain the link breaks and the revenue lands in Direct / Unknown, which is honest, but a mobile number frequently survives a job change where an address does not. Carrying both, where the file has both, is worth the small effort of including the extra column. It costs nothing to supply and it is the difference between crediting a channel correctly and losing the strongest evidence it has.

The corollary is about what to export, and it is a data minimization point as much as a practical one. You need the buying contact, the amount and the date. You do not need site addresses, access codes, keyholder names, staffing rotas or anything else operational, and none of it improves the match by a single row. Sending less is faster to produce and easier to get past whoever reviews supplier arrangements at a client of any size. Where a client's own contract restricts what may be shared about their premises, a file containing none of it is a short conversation.

The largest bucket is usually referral and word of mouth, and it should be. Contract cleaning is bought on reassurance, and a facilities manager asking a peer who they use is the most common first step in the whole process. That revenue genuinely was not produced by advertising, and showing it as Direct / Unknown at full size is what makes the paid numbers underneath worth reading. A report claiming to have traced most of a cleaning company's wins to campaigns would be describing something that did not happen, and it gets discarded the first time somebody checks one account against it.

Below that, the surprise is often tender activity. Portals and bid services carry a real subscription cost, get renewed on the number of opportunities they surface, and can be ranked on billed revenue like anything else once each portal's opportunities go into the lead file tagged with the portal. One portal may turn out to carry most of the won revenue and two others almost none, having been renewed side by side for years on identical reasoning. That is a decision the report makes for you rather than an insight needing interpretation, and it is usually the fastest saving available.

The third finding is about the scale of inquiry. Paid search in this category produces a high volume of inquiries from very small premises — a single office suite, a salon, a lock-up — which consume the same survey and quotation effort as a fifty-thousand square foot site and return a fraction of the revenue. Ranked on billed revenue, the channel generating them can look dramatically worse than its cost per inquiry suggested. The answer is rarely to switch it off, because the same channel also produces the occasional large inquiry; it is to bid and qualify differently.