ATTOM published its August numbers this month and two of them point in opposite directions.
Foreclosure starts, meaning new filings, came in at 25,894, down 3 percent from July. Completed foreclosures came in at 5,794, up 22 percent in a single month and up 42 percent from a year ago.
Fewer people entering the process. Considerably more reaching the end of it. That is not a volume story, it is a composition story, and it quietly undermines the way almost every list in this business is sold.
The list everyone buys is organised around the single day that produces the fewest answers.
Our own pipeline is blunt about it. I pulled every seller who has ever replied to us to talk, leaving out the much larger number who only replied to opt out, and matched them to the recorded date of their foreclosure notice. 62 of them had both dates. The middle one answered about 98 days after the notice was recorded. Fifty three percent answered more than 90 days out. Twenty six percent answered more than 180 days out. Fewer than one in five answered inside the first 30 days, which is the window essentially all of the competition and essentially all of the spend is pointed at.
The crowded window is quiet for a reason that is obvious from the other side. The week a notice hits the record is the week thirty strangers find out at once, because they all bought the same list on the same morning. Nobody answers that week. Not for lack of motivation. Answering means discussing the worst thing currently happening to you with a stranger, and there is a queue of them. Three months later the panic has worn into something closer to resignation, which is a state people actually make decisions in, and the phone has gone quiet because everyone moved on to the next batch.
Which moves the bottleneck somewhere most people are not looking. If most of your answers arrive after day 90, the binding constraint is not lead acquisition. You already bought those leads. You bought them in June and wrote them off in July. The constraint is whether anything is still talking to them in month four. That is a capacity problem, and it is the specific problem that does not yield to working harder, because the failure is never one bad evening. It is the fourth month of evenings.
The audit, ten minutes, in your own CRM. Filter to every lead you marked dead in June and July. For each one, take the date of the last outbound touch and subtract the date of the first. If most of those gaps come in under thirty days, you did not learn that those leads were bad. You learned how long you personally last, which is a different finding and a more useful one.
The honest part. Sixty two sellers is a small number, and it is small because most replies are someone asking to stop, which I did not count. That 98 day median is also partly a measurement of us, and I am not going to pretend otherwise. We do not reach people on day zero. Records take time to surface, enrichment takes time after that, and our own cadence spreads touches out deliberately. Some of the gap between a notice and a reply is the seller getting ready and some of it is us being slow. I cannot cleanly separate the two from this data. What survives the caveat is the shape: the replies are not clustered at the front, they are spread across months, and a follow-up that ends in week three never finds out which kind of gap it was looking at.
One market note, then I will let you go. The thirty year fixed average was 6.95 percent on September seventeenth, up from 6.76 percent the week before and 6.26 percent a year ago. Every seller who told you they were waiting for rates to come down was making a bet, and the bet has gone the other way two weeks running. Those conversations did not end. Their premise changed, and almost nobody is going back to mention it.
The longer version is on the blog this morning, with the sample-size caveats a newsletter has no room for: The sellers who answer us are a hundred days past their notice
If you would rather not run that June audit by hand, the Grader does the pass for you and hands back the leads you wrote off that still have signal, ranked, with the money attached. Theirs reads a dropdown. Ours reads your list. We never store it, sell it, or message anyone on it.
→ dealroute.ai/grader
DealRoute is an automated acquisitions platform in private beta with a small group of operators. If you want the follow up to keep running whether or not you are awake, Early Access requests hear from me first: dealroute.ai/early-access
Talk soon,
Jason