How ETR gets set in the first 24 hours of a storm event
The ETR that goes out an hour after the storm passes is not a forecast of damage. It is a placeholder built from a restoration curve, and every coordinator who has worked a major event knows the difference even when the public doesn't.
Why the first number is a placeholder, not a forecast
At the point your call center or OMS pushes an initial estimated time of restoration, almost nobody has laid eyes on the circuits yet. The number comes from historical restoration curves tied to storm category, total customers interrupted, and the mutual aid headcount already committed or in transit. It's the same math a dispatcher ran on the last comparable event, scaled to this one's outage count. That's a reasonable starting point. It is not, and was never meant to be, a read on whether a given feeder lost three spans of conductor or a substation transformer.
This is where the first wave of customer complaints usually lands. People hear "restoration estimated by Thursday" and assume someone has looked at their street. Nobody has. The estimate is a system-wide average dressed up as a specific promise, and the gap between those two things is what burns trust in hour six when the first revision goes out and it's later, not earlier.
What moves the second ETR
The second ETR, and every one after it, should be driven by what damage assessment teams and advance crews are finding once they get past the substations and onto the backbone feeders. That's where the real adjustment happens: patrol crews calling in broken poles, downed laterals, blocked access roads, and the kind of structural damage that changes a four-hour job into a two-day one.
Sequencing is the hard part here. Walking or driving every circuit in a territory to confirm damage takes real time, often days on a large footprint, and in the first 24 hours you're already making crew assignment and mutual aid routing calls long before that walk is anywhere close to finished. Most shops fall back on trouble-ticket density as a proxy, treating call volume as a stand-in for damage severity. Trouble-ticket volume lags real damage, since it depends on someone noticing a line down and having phone service to report it. A circuit with no customers to call in a ticket, because it serves a stretch of rural feeder with low density, or because cell service is down in that area too, can carry serious damage and generate almost no tickets at all.
Where a same-day comparison view fits the ETR update
A pre- and post-event imagery comparison across the territory gives you something to sequence crews against before every circuit has been walked, which is the gap trouble tickets leave open. You can get a same-day comparison view of where damage is concentrated the day imagery becomes available, flagging the stretches of line that look hit hardest even in the areas generating no calls. That's not a replacement for the physical patrol, a lineman still has to confirm what's broken before a crew gets dispatched. But it changes which circuits get patrolled first, and that ordering decision is exactly the one driving your second and third ETR revisions.
The honest version of ETR communication to customers acknowledges this two-step process out loud: an initial estimate based on storm scale and crew counts, then a revised one once assessment confirms what's down. Utilities that explain the methodology, even briefly, in their outage map or IVR script tend to get fewer "why did it get worse" calls than the ones that just post a number and let customers draw their own conclusions about what it meant.
If your damage assessment process still leans mostly on trouble-ticket volume to guess where the worst damage is, it's worth seeing what a same-day imagery comparison adds to that first 24 hours.