A registration gate produces contacts. Some of them are buyers.

Asking visitors to register before they can see more listings is a reasonable lead capture strategy, and it produces a lot of registrations.

A share of them come from people who only wanted to keep browsing, and who typed whatever let them through — a spare email, a wrong number, a made-up name.

Those registrations look identical to real ones in a lead count. They separate only when the file is matched against clients who closed.

The report shows the match rate per source beside the revenue, and a source with a very low match rate is telling you most of its registrations were never real contacts.

That is a fact about the source's traffic, not about the team's follow-up, and it is the first thing worth knowing.

A visitor asking what their home is worth is thinking about selling, often soon. A visitor registering to save a search may be a year from buying.

They arrive through the same site and often the same campaigns, and pooled together they produce a conversion rate that describes neither.

Split by lead type, the picture changes: seller leads are usually fewer and far more valuable per lead, and a source strong in one is often weak in the other.

Teams that run this often find they have been judging a seller campaign by buyer-lead standards, or the reverse.

Once somebody has registered, saved-search emails and listing alerts keep them coming back for months.

By the time they contact an agent, the most recent thing they touched is often an alert email, and any last-touch report credits the alert.

The alert kept them warm. The source that brought them to the site in the first place produced them, and that is what the report credits.

The alerts' own value is real and is better measured as how long registrants stay engaged, not as acquisition.

Crediting acquisition to the alert also hides the real question about the source: whether the people it brought were ever going to buy. A source that produces registrants who engage with alerts for a year and never close is not a good source with patient buyers. It is a source of browsers, and the alert is keeping them browsing.

Home purchases follow months of looking, and a registration from this spring may close next year.

A report grouped by closing date credits this quarter's commissions to marketing from a year ago; grouped by registration date, it puts each client against the source that produced them.

The most recent registration cohorts will always look poor, because they have not had time to close. The report labels them as incomplete rather than comparing them with mature ones.

Cutting a source because its newest registrants have not closed yet is the most common error this prevents.