How we model your wait, and what the data can and cannot say
Plenty of tools will hand you a confident invitation cut-off and a tidy date. We do not, because the public data does not honestly support it. This page lays out exactly how we model your wait, which signal we trust, which we deliberately ignore, and where we draw the line and tell you we simply do not know.
What we read: the submitted queue and its trend
Our model rests on one reliable signal: the depth of the submitted Expression of Interest queue for your occupation. We look at how many EOIs sit at or above your points score, and how that count has moved over recent rounds. A queue that is shrinking at your score is reaching down to you faster, a queue that is growing is moving away from you. Pair the depth of people ahead of you with the recent pace at which rounds clear the queue, and you get an honest estimate of how long you are likely to wait.
We express that estimate as a band, for example a few months as against closer to a year, because the real process carries real uncertainty. Round sizes change, new candidates join, others raise their scores or get invited. A range tells the truth about that uncertainty where a single date would only hide it.
What we ignore: per-round invitation cut-offs
You will often see a confident claim that a given occupation was invited down to some exact score in the last round, and that you are therefore a fixed number of points clear. We do not lead with that, on purpose. The Department’s published invitation figures are heavily redacted: small counts are masked, and the headline INVITED figure behaves as a cumulative stock rather than a clean per-round count. Trying to back out a precise per-occupation cut-off from numbers built that way produces a figure that looks exact and is not.
We do not tell you that you are N points above the clearing line, because we do not believe the clearing line the public data implies. We tell you where you sit in the real submitted queue and which way that queue is moving. That is the claim we can defend.
Where we stop and say we don't know
Some occupations have queues too shallow, or histories too short, to model a meaningful wait. When that happens, we do not stretch the data to fill the gap. We tell you the queue is thin and that we cannot responsibly put a band on it. A model that always produces a confident number is a model you cannot trust on the cases that matter. Knowing when to decline is part of the method, not a failure of it.
The upshot is simple. Treat our wait band as an honest read of your position and its direction, not a countdown to a guaranteed date. It is built to be right about the shape of your situation, even when it cannot be precise about the day.
See the method applied to a real occupation
Common questions
Get your honest wait band
See where your score sits in the real queue, and the range you are honestly looking at.