What this tool does
You pick the month your citizenship application was filed. The tool reports how many applications were still pending ahead of yours at the most recent snapshot, how quickly that number is falling, and roughly when it would reach zero if the current pace held. It shows the same projection under three different assumption sets, because the honest answer depends heavily on which assumptions you accept.
Where the data comes from
Five snapshots of the pending-application backlog, taken on 8 June 2026, 7 July 2026, 10 August 2026, 3 September 2026 and 9 October 2026. Each snapshot lists, for every filing month, how many applications filed in or before that month were still pending on that date: the same “people ahead of you” figure the official checker reports.
That checker is IRCC’s check processing times page, and it is the sole source of every figure on this site. It publishes only the current backlog, which is why the snapshots above are kept: the history they form is what makes a rate calculable at all, and it exists nowhere official.
This is important to understand: the horizontal axis is the month an application was filed, not the passage of time. A single snapshot is a cumulative curve across filing cohorts. All movement, every rate on this site, comes from comparing the two most recent snapshots against each other. That is still a single difference between two readings, which is a thin basis for a projection and the single biggest limitation here.
How the estimate is calculated
For your filing month, the queue ahead is the October 9 count of applications filed in or before that month. The clearance rate is how much that same cumulative count fell between September 3 and October 9, the two most recent snapshots, scaled to a month by the 36 days between them. The estimate is simply the queue divided by the rate, projected forward from the October 9 snapshot.
Each filing month uses its own observed rate rather than one global average, because clearance is very unevenly distributed across cohorts. For an August 2024 application: 15,600 pending ahead, falling at 1,945 a month, so roughly eight months (about June 2027) before the queue ahead is exhausted.
What the three scenarios mean
1. No assumptions
Everyone who filed before you counts as ahead of you, and the observed rate applies. This is the most conservative reading and produces the latest date. Its weakness is that it assumes every old, apparently-stalled application must be resolved before yours, which the evidence does not support.
2. Minus the pre-September 2023 cases
A few thousand applications filed before September 2023 are still pending: 3,500 at the October 9 snapshot. Many appear to be long-term stalled cases (complex, contested, or awaiting information) that are not meaningfully competing with recent filings for capacity. This scenario removes them from the queue ahead. Note that removing cases from the count does not add processing capacity: if those cases are real and still consume staff time, the true answer sits closer to scenario 1.
That figure is read straight from the feed, which reports every filing month back to 2006, sparsely: at the October snapshot, 28 months before November 2023 held between 100 and 300 each. The tables here fold them into a single “October 2023 or earlier” row (4,000 at October), of which 500 are September and October 2023 themselves. The bucket is draining steadily: 6,100 in June, 5,400 in July, 4,800 in August, 4,400 in September, 4,000 in October. The deduction is therefore re-read at every snapshot rather than held fixed, which a fixed figure of 5,000 had already outgrown. Until October 2026 it was estimated instead, as the bucket less two months’ worth of the November 2023 cohort, which ran about 300 under the published figure.
3. The same, and ~250 per month never cleared
On top of the above, this treats roughly 250 applications per filing month as never clearing at all: a persistent residue in every cohort. Over the twelve cohorts between September 2023 and August 2024 that is about 3,000 applications, so the effective queue ahead of an August 2024 filer drops to roughly 9,100 and the estimate moves several months earlier.
This is the shakiest of the three, and the snapshots so far mostly undercut it. The oldest cohorts keep draining rather than settling at a residue: November 2023 has gone from 700 to 400 across the snapshots, though it held still for the first time in October, and February 2024 from 1,900 to 500. If that continues, a 250-per-month residue will not survive many more snapshots.
The random-draw odds
Processing is visibly not strictly first-in-first-out. Cohorts well behind the front of the queue still lose applications every month regardless of position, effectively a random draw. How many varies a lot from one filing month to the next: across the five snapshots, March 2024 lost about 320 a month while August 2024 lost about 70. The tool therefore reads each filing month's draw rate off its own trend (described below) and samples it without replacement. A month with too little settled history of its own falls back to the median across every cohort and snapshot pair, 180 a month at the October snapshot.
Because those draws shrink the pool, the odds for anyone still waiting rise each month: an August 2024 cohort of about 1,800, losing 70 a month, starts near 3.9% per month and reaches about 5.6% by the time the queue ahead clears. The cumulative chance of being drawn within k months is the draw rate times k, divided by the cohort size: around 31% before the queue would otherwise clear. In other words, the projected date is closer to a worst case than a midpoint: there is a meaningful chance of an earlier, luck-driven decision. For March 2024, losing about 320 a month from 800, the draws alone would empty the cohort before the queue ahead does.
Your cohort's own trend
The trend card under the five figures ignores the queue ahead and looks only at your filing month: a straight line fitted through its pending count across the recent run of snapshots in which it moved like processing rather than re-dating. Where that line falls fast enough, it gives a date. March 2024 went from 2,100 to 800 between the June and October snapshots, about 320 a month, which would bring it down to its last 250 by around November 2026, four months ahead of the no-assumptions estimate.
The date is for reaching 250, not zero, because that is the residue the stuck scenario assumes never clears from any filing month. Using the same figure means the card and that scenario describe one queue: the chart draws it as a dashed line, and the pace line ends on it.
Where the line falls too slowly to reach that 250 within three times the span of data behind it, as it does for April 2024 through May 2025, it gives no date: a pace read off four months says little about the next two years. It also needs at least three settled readings, so a filing month that has only just stopped settling shows none. When the trend and the estimates disagree, the usual reason is that the queue ahead includes applications that are not being worked in filing order, such as the 4,000 still in the “October 2023 or earlier” row.
Under the chart, “Clearing ahead of you” names the earlier filing months that lost the most between the two latest snapshots, counting only falls bigger than one rounding step. For August 2024 at the October snapshot that was March 2024 (600), October 2023 or earlier (400) and January 2024 (300), out of 2,100 that left the queue ahead in all.
What IRCC itself says about your month
The feed carries IRCC’s own estimate for every filing month, as a short sentence: a countdown such as “3 months left”, or “We need more time to process your application”. The fifth figure above the charts shows it for your month, unchanged except that the longer sentence is shortened to “Needs more time”. It is IRCC’s figure, not ours, and the only official per-month estimate there is.
The two kinds of answer divide the queue cleanly. At the October snapshot every filing month from May 2025 on had a countdown and every earlier month read “we need more time”. That line is tracked under “The queue as a whole”; it stood at December 2024 in July. Where IRCC gives a countdown it is usually shorter than the dates above, which assume everything filed before you clears first. For the oldest months it gives no date at all, which is where this site’s estimates and the cohort trend are the only figures on offer.
What does “We need more time to process your application” mean?
It is the text IRCC publishes for a filing month once it has stopped quoting a countdown for it, and it is all the feed says: no reason and no new date. On this site it marks the months whose applications are being decided a few hundred at a time rather than in filing order. Those months do still move. The movement table shows how much each lost at every snapshot, and the cohort trend reads a pace off it.
The queue as a whole
Four figures for every filing month together. Pending is the movement table’s “All months” row at the latest snapshot, 322,700 in October, and the change is against the snapshot before. Decided per month is how much the settled filing months lost between the two latest snapshots, scaled to a month by the days between them: 28,300 over the 36 days to October 9, or about 23,900 a month. It leaves out the months still settling, so it is a floor rather than the full count.
New filings per month comes from a habit of the feed. New applications are not dated to their own month at once: they sit in one oversized “current” month until they are entered, and a month’s count only becomes real once that pile has moved on to a later one. The size a month has at that point is what was filed in it. Five months have been watched leaving the pile since June (March to July 2026, between 26,200 and 33,100 each), and the figure is their average, about 28,200. Filings and decisions being within a few thousand of each other is why the total has barely moved: 326,400 in June, 322,700 in October.
Where the work went
The bar chart at the foot of the page is the movement table’s latest column drawn out: how many applications left each settled filing month between the two latest snapshots, with your month in a different colour. It is the whole-queue version of the “clearing ahead of you” line in the trend card. The sentence above it names the shortest run of consecutive months that took at least 70% of the fall, which is where IRCC’s capacity is going: January to April 2026 at the October snapshot, 74% of the 28,300. That window has moved forward about one filing month per snapshot, which is the pattern the caveats at the foot of the front page describe.
Why some months cannot be estimated
Between the September 3 and October 9 snapshots, the May 2026 cohort fell from 141,700 to 26,700, while the August 2026 cohort, empty a month earlier, rose to 94,200. Roughly 115,000 records left May 2026 without being decided. Taken at face value that movement reads as the fastest clearance anywhere in the data, which is precisely why it is not taken at face value.
The estimator detects this rather than being told about it. Any filing month whose cohort grew by more than one rounding step between the two most recent snapshots, or which shed more than half its cases in a single month (something genuine processing has never come close to, the fastest real month being about 40%) is treated as still settling, along with every month after it. Those months stay in the movement table, because the movement is the evidence, but they cannot be selected and no estimate is offered for them. At the October snapshot that means May 2026 onward.
The rounding step matters more than it sounds. The feed publishes to the nearest 100, so a reported rise of exactly 100 is the smallest move it can express and may be nothing at all. In the September 2026 snapshot the August 2024 cohort went from 1,900 to 2,000, and because the frontier is the first month to trip the test, that single step briefly marked August 2024 and the 25 months after it as unusable. Growth now has to clear one step to count. Real re-dating does not arrive a hundred at a time.
How much of the 28,300 drop across the earlier cohorts is genuine decisions, rather than records re-dated forward, cannot be settled from the feed: it publishes only a current backlog, never a decision count. Over the same period the total pending figure fell from 327,100 to 322,700, its largest fall in the five snapshots so far. Treat overall throughput as unknown.
Known limitations
- One data point of movement. Every rate rests on the difference between the most recent pair of snapshots. They are not evenly spaced (the gaps so far are 29, 34, 24 and 36 days), so the observed fall is divided by the days between and multiplied back up to a month. That removes the calendar from the answer but not the thinness of it: each further snapshot could still change all of these numbers materially.
- Everything is rounded to the nearest 100. A cohort losing 70 a month moves by less than one rounding step between snapshots, so the slower draw rates rest on a handful of 100-sized steps. August 2024's true figure could plausibly be anywhere from about 50 to 100 a month, and its draw odds move with it.
- Straight-line projection. Real clearance rates change with staffing, policy, and intake. Nothing here models that.
- The assumptions are estimates, not published facts. The pre-September 2023 deduction and the 250/month residue are inferred from the shape of the data, not stated by any authority.
- Cohort-level, not case-level. Nothing here accounts for anything specific to an individual application.
Frequently asked
Which estimate scenario should I believe?
Treat them as a range rather than picking one. The spread between them, roughly February to June 2027 for an August 2024 filing, is an honest reflection of how much the answer depends on unverifiable assumptions. If you want a single number, the middle scenario is the least opinionated.
Why does my estimate change when new data arrives?
Because it should. These are projections from five observations, not a schedule. Each new snapshot either confirms the current rate or reprices it.
Can I do anything to speed up my citizenship application?
Nothing in this data speaks to that, and this site cannot advise you. For anything about your own application, contact the relevant department or a qualified immigration adviser.
Is my data collected?
We never ask who you are, and the estimate is calculated in your browser: the month you select is not sent to us, only remembered in your browser so the page opens on it next time. The site uses Google Analytics to measure usage, which sets cookies. There is no advertising and no advertising code. The privacy policy explains exactly what they collect and how to opt out.