An operator running 107 properties across three asset classes could not state what its marketing spend produced. Not approximately, not directionally. The numbers reaching leadership were rebuilt by hand each cycle and did not reconcile with the platforms they came from.
What follows is anonymized, but every figure below came out of that portfolio's own systems, and every one of them was traced to a specific file, tab and column before it was written down. Where a number could not be sourced, it was cut from the report and the gap was disclosed. That constraint matters, because the findings are uncomfortable enough that they need to be reproducible.
The paid advertising that produced nothing
Over twelve months, one business line's paid search and paid social showed zero leases. So did the listing site. So did the website form. This against an annual advertising budget in the low six figures.
Zero is not a credible result. What it actually meant was that whoever records each move-in was not capturing the channel it came from, which was confirmed by the single largest lead bucket in the same report being labeled Unassigned, accounting for nearly a quarter of all leases with no source attached at all.
This distinction is the whole thing. The advertising was not proven to fail. It was proven to be invisible. Those require opposite responses, and an operator who cuts the spend while blind is making a decision on an artifact of their own record-keeping.
Where the numbers came apart
- More than half of one portfolio's move-ins carried no lead source at all. Across all leads the blank rate was 45%.
- One channel appeared to convert at 96.5%. It was a labeling artifact: the system auto-tagged website rentals to that source. Until that was fixed, no channel's true cost per lead could be trusted.
- Roughly a third of inbound calls on two business lines went unanswered, on the order of ten thousand calls a year, against a 6.9% miss rate on the one line where the data was clean.
- For 40.7% of lost prospects, first and last recorded contact fell on the same date. A site can answer its phones well and still lose most prospects this way, which makes answering and following up two separate problems with two separate fixes.
- The ROI trackers reported average length of stay as roughly 126 years. An unconverted date value that had gone unnoticed. The same sheets carried over a thousand formula errors, nearly all one unguarded division repeated across tabs.
- Paid spend could not be stated as a single defensible number. The platform reported roughly 148 thousand for a period where the internal tracker logged 116 thousand, a gap of about 32 thousand on the same spend.
None of the five priority actions required new spend. Each was a matter of discipline and ownership: make lead source persist from click to move-in, answer the phone without depending on site managers, retire the broken trackers, and reconcile spend monthly.
Why this is so common
It is almost never a competence problem. In this case the marketing team was three people covering three business lines, and they had already identified several of these issues themselves and asked for the resources to fix them. The constraint was not effort or insight. It was that the systems did not connect, and that accountability for turning a lead into a lease was not clearly owned by anyone.
Attribution decays quietly. Nobody schedules the moment it breaks, no dashboard turns red, and the reporting keeps producing numbers that look like numbers. By the time anyone questions them, several budget cycles have been allocated on the strength of figures that were never true.
What you can check this week
Pull your own move-in report for the last twelve months and count how many rows have a blank or unassigned lead source. If that figure is above ten percent, the channel-level performance numbers your team is reporting are estimates rather than measurements, and the decisions being made from them deserve a second look before the next budget is set.