A four-month sales cycle, measured on form fills.

An individual buying a course clicks and pays, and the pixel records it perfectly. An employer buying a program for forty people fills in a form, takes a call, asks for a proposal, runs it past L&D and procurement, and signs a quarter later.

The second kind is where the revenue is, and it is invisible to every measurement your ad platforms offer. So budget gets defended with demo requests and MQLs — numbers that correlate with revenue only loosely, and that a finance director has learned to discount.

The right unit of attribution is the original inquiry, because that is where the channel did its work. What CloseRev does is match your CRM's closed-won export against the lead-source export from the same period range, so the contract credits the campaign that produced the conversation.

Because periods are saved rather than overwritten, the picture compounds: by the third quarter you can see which channel produces inquiries that actually become contracts, and which produces a busy pipeline that never closes.

Corporate training is rarely bought at full size on the first order. A learning and development team runs a pilot with one department, measures whatever it has decided to measure, and comes back six months later for the wider rollout if it went well. The first contract is small and hard to justify against the cost of winning it. The revenue that makes the relationship worthwhile arrives in the second and third years, by which point nobody is still asking where the account came from.

Because the match is on the person who made the original inquiry, later contracts with the same buyer credit the channel that produced them. Export every closed deal rather than only new logos and the ranking rearranges itself around accounts instead of around first orders. A channel producing four pilots that each became a rollout is worth a great deal more than one producing twelve pilots that stayed pilots, and on a first-order report those two appear as the weaker and the stronger performer respectively.

This is also the honest way to handle renewals. A renewal is not produced by the campaign running in the month it is signed, and crediting it that way flatters recent advertising while erasing the work that won the account in the first place. A deal-type column mapped as Product / service shows new business, expansion and renewal as separate lines, each with how much of it traces to a channel, and so growth driven by existing customers is not mistaken for a working acquisition channel.

Learning and development professionals move jobs frequently, and when they do they take their supplier preferences with them. A buyer who ran a successful program at one employer will ask for you by name at the next, often within a year. On any account-based model that arrives as a new company, a new logo, and a sale attributed to whatever campaign happened to be running when they got in touch — usually a branded search that did nothing but be findable at the right moment.

Matching on the person rather than the company describes what actually happened instead of what is convenient. The buyer's email or phone appears in the original inquiry file and again in the new deal, so the credit follows them, and a company-name column keeps the two employers distinguishable inside the report. For a training provider whose growth is genuinely driven by reputation within a small professional community, this is not an edge case at all; it is the main mechanism, and most have never had a number for it.

The reverse case deserves a mention because it is the more common disappointment. When a buyer leaves and the account is inherited by somebody who never met you, the renewal frequently does not happen, and nothing about the campaign that won the account was at fault. Carrying the contact on each deal makes that visible as a pattern rather than as a run of bad luck, which turns it into a question about who else at the client you know — a prompt the marketing report can at least supply.

There are several defensible numbers to export and they answer different questions. Contract value at signature tells you what the channel sold. Revenue recognized as cohorts are delivered tells you what the business earned, which for a multi-year program can be a very different shape. Invoiced amounts sit somewhere between the two. None of these is the right answer in general; the wrong answer is using more than one of them across a single history, because the report cannot detect that and the comparison silently stops meaning anything.

Seat counts matter alongside whichever figure you choose, because a training business is capacity-constrained in a way most service businesses are not. A channel producing large contracts that book every facilitator for a quarter is a different proposition from one producing the same revenue across many small cohorts, and the second may be better or worse depending on how the delivery team is structured. The report does not read a seat count, but it shows each channel's number of sales beside its revenue and breaks revenue down by deal size, which together go a long way toward that distinction.

Where programs are delivered across several client sites or regions, a region column mapped as Location earns its place too. It shows each region's revenue and how much of it traces to a channel, which matters particularly where a partner or reseller operates in one territory and not another, because an aggregate figure quietly averages a strong market together with a weak one. It does not rank channels within each region, but a region whose revenue sits mostly in Direct / Unknown is visibly one the current channels are not reaching. That is the difference between concluding a channel does not work and concluding it does not work in one particular place, and the two lead to very different decisions about next year's budget.