Re-driving the same routes: what's the decay curve on a logged address?

I run two routes that overlap on about forty properties and the thing I notice is that a mowed yard doesn't mean anything changed. Grass gets cut, mail still piling up, same peeling fascia, same owner on the record. Code enforcement leans on somebody and the visible signal goes away for a month. If your app scores off photos you'll mark that one cured and it isn't.

On the management side the number I'd watch is how long the vacancy has run, not whether it looks bad. A unit empty 30 days is a turn. A unit empty past 90 with no lockbox and no sign is somebody who has stopped trying. If your logging captures a date and you re-drive, the second sighting is worth more than the first because it converts a snapshot into a duration.

Dumb question on the holdout, do both groups have to get the exact same mail piece for that to mean anything? I'd have assumed you'd want a different letter for the second pass since they've already seen five.

Before the decay question I'd want to see the spend. 4,200 addresses times five touches is 21,000 pieces. At 60 cents a piece all in that's around $12,600, plus skip tracing. If the six-month cohort is half of that volume at half the response, you're paying full mail cost for a cohort that may be structurally worse. Have you costed the two cohorts separately or just the blended number?

Boarded window comes off, new plywood painted, that's a $200 job somebody did to stop a citation. Roof still shot. I'd weight the expensive deferred items way heavier than the cosmetic ones in your scoring, because the cosmetic stuff is exactly what decays in six months and the roof doesn't.

Slightly sideways question. If a driven list has a measurable half-life, does anyone actually buy one? I've been trying to understand what an operator's lead pipeline is worth when they raise money against it, and a decaying asset with no exclusivity sounds like it's worth close to nothing on a balance sheet.

Vacant land basically doesn't decay. A weeded lot looks the same in three years. What moves is the tax status, and in most counties delinquency is public and dated, so I stopped treating the drive as the clock and started treating the tax roll as the clock. Doesn't help ledger with houses, but if any of those 4,200 are lots you're re-driving them for nothing.

I re-drive on 90 days and I won't pretend I've tested it. It's just that anything older than that I've stopped trusting my own notes on.

The selection point rena-expert made is the one I'd chase first, because it's free to check. Sort your 4,200 by your own distress score at time of logging and see whether the early cohort scores higher. If your average score has been sliding month over month, the age gap in response is mostly you running out of easy streets, and re-driving those same streets won't put the easy ones back.

Past a certain list size I think expansion wins regardless of what the holdout says. Six zips is a small footprint for 4,200 records, which means you're already mailing a meaningful share of the distressed stock in that footprint. Re-driving raises your touch density on the same owners. Adding a seventh zip gets you owners who've never heard from anybody. The decay question matters, but saturation might be the thing actually capping you.

I have zero addresses logged and this has made me less clear rather than more. If the first routes you drive are the best ones and everything after is marginal, then the response rate a new person sees in month one is the best it'll ever look? That seems worth knowing before I build a budget off it.