Lead source tracking for real estate teams

Every team lead can name their lead sources. Very few can name their cost per closing by source — and the gap between those two sentences is where marketing budgets go to be wasted. Source tracking is the plumbing that connects a dollar of spend to the deal it eventually produced, across the months and hands in between. Here is the chain, link by link, and the decisions it unlocks.

Link one: capture the source at birth

A lead’s source must be stamped the moment the record is created — reconstructed attribution is fiction with a straight face. The mechanics per channel: campaign-tagged URLs (UTMs) on everything you pay for, so an ad click arrives wearing its campaign; distinct forms or paths per channel where tagging can’t reach; and a mandatory “how did you hear about us” logged on phone-first leads. On BrokerFold, a lead captured on a tenant site carries its context — the page, the listing, the campaign parameters — onto the contact record automatically, which closes the gap where source data traditionally dies: the manual entry nobody makes at 9pm. Whatever your stack, the standard is binary: a lead without a source should be impossible, not discouraged.

Link two: keep a small, ruthless vocabulary

Source taxonomies rot into uselessness — “Facebook,” “FB ads,” “social,” and “Meta” as four spellings of one channel. Maintain a short controlled list (a dozen entries beats fifty), with two levels at most: channel, then campaign. Someone owns the list; new sources get added deliberately, not typed ad hoc. This is dull, and it is the difference between reports and archaeology.

Link three: carry it to the closing

The chain’s weakest link is the middle: the lead becomes a client, months pass, the file closes — and the closing gets recorded with no memory of the ad that started it. The fix is structural: the closing record inherits source from the contact record, automatically, because they are the same record in a properly shared database. This is also why per-agent CRM silos destroy attribution — the case in full lives at shared database vs separate CRMs.

The two reports that pay for all of it

Source funnel: for each source — received, contacted, appointments, agreements, closings, with your baseline rates beside them. This exposes quality: the cheap source that never reaches an appointment, the expensive one that closes. Cost per closing by source: spend ÷ closings attributed, trailing twelve months. This is the budget’s report card, and it reads against what leads cost in your market. One subtlety worth pulling: per-source response time by agent — teams reliably work “good” sources faster, so a source’s poor showing can be effect rather than cause. Check the follow-up before you kill the channel.

The decisions: kill, scale, or fix

Quarterly, sources sort into three buckets. Kill what has had a fair window, real follow-up, and still no economics. Scale what closes below your target cost — the report’s happiest sentence is “this works and has headroom.” Fix the middle: decent leads converting poorly usually indicts speed or script, not the source. Do this four quarters running and the marketing budget starts compounding: money migrates from proof-free channels toward proven ones, and the team’s KPI stack gains the one column that makes every other number financial. Site-side behaviour — which pages and campaigns produce registrations in the first place — feeds the same picture through website analytics; together they turn “we should market more” into “we should buy more of exactly this.”

Expect the first honest report to be uncomfortable. Almost every team discovers that a channel someone championed is economically dead, and that a boring one — usually the website or the sphere — has been quietly carrying the year. Let the data win those arguments anyway: the discomfort is one meeting long, while unmeasured spend is forever. Attribution’s real product is not a report; it is a marketing budget that learns.