A four-state cumulative flow diagram that reads band thickness as work in progress and the horizontal gap between the arrival and delivery curves as approximate lead time, annotated on the plot and reachable by keyboard.
Preview in your theme
Loading preview…
This component is part of Pro
The install command, prompt and concept map are included with a Pro license — one purchase unlocks every Pro component, forever.
status="ready" · 28 days, 5 stages. Band thickness = work in progress in that stage; the horizontal distance between the arrival and delivery curves = roughly the lead time. Reviewers are pulled onto an incident on days 10–18: In review swells from 3 to 16 and the horizontal arrow stretches from ≈6 days to ≈17 before settling back to ≈10. In the first week deliveries have not yet exceeded the work already on the board, so the crossing falls outside the window and the lead time can only be given as a lower bound — one card showing both the measured and the censored branch.
Cumulative flow diagram: Delivery flow · Squad Atlas. 5 stages across 28 columns from Mar 2 to Mar 29. 61 issues arrived, 45 issues delivered, 16 issues still in flight at the end. Work in progress peaks at 28 issues on Mar 22. Approximate lead time is about 10.3 days at Mar 29, against longer than this window can show at Mar 2. Average delivery rate 1.63 issues/day. Little's Law on the average WIP of 17.21 issues gives 10.6 days, against 10.6 days measured horizontally.
Delivery flow · Squad Atlas: every date with the queue in each stage (workflow order, upstream first), the total in flight, the cumulative arrivals and deliveries, the trailing delivery rate and the approximate lead time read back to the arrival curve. Counts are in issues.
Date
Backlog
Ready
In progress
In review
Done (cumulative)
In flight
Arrived
Delivered
Rate
Approx. lead time
Mar 2
3
4
3
2
1
12
13
1
no rate yet
longer than this window can show
Mar 3
1
3
4
3
3
11
14
3
2 issues/day
at least 1 days
Mar 4
4
2
3
3
6
12
18
6
2.5 issues/day
at least 2 days
Mar 5
5
3
2
3
9
13
22
9
2.67 issues/day
at least 3 days
Mar 6
7
2
3
3
11
15
26
11
2.5 issues/day
at least 4 days
Mar 7
7
2
3
3
11
15
26
11
2 issues/day
at least 5 days
Mar 8
7
2
3
3
11
15
26
11
1.67 issues/day
at least 6 days
Mar 9
6
2
2
3
14
13
27
14
1.86 issues/day
about 6 days
Mar 10
4
3
2
2
17
11
28
17
2 issues/day
about 6.3 days
Mar 11
5
4
2
2
19
13
32
19
1.86 issues/day
about 6.8 days
Mar 12
4
2
4
3
20
13
33
20
1.57 issues/day
about 7.5 days
Mar 13
2
3
3
5
21
13
34
21
1.43 issues/day
about 8.3 days
Mar 14
3
3
3
5
21
14
35
21
1.43 issues/day
about 9.3 days
Mar 15
3
3
3
5
21
14
35
21
1.43 issues/day
about 10.3 days
Mar 16
4
2
4
7
21
17
38
21
1 issues/day
about 11.3 days
Mar 17
1
4
3
10
21
18
39
21
0.57 issues/day
about 12.3 days
Mar 18
3
1
4
12
22
20
42
22
0.43 issues/day
about 13 days
Mar 19
4
3
3
14
22
24
46
22
0.29 issues/day
about 14 days
Mar 20
4
5
2
16
23
27
50
23
0.29 issues/day
about 14.8 days
Mar 21
4
5
2
16
23
27
50
23
0.29 issues/day
about 15.8 days
Mar 22
5
5
2
16
23
28
51
23
0.29 issues/day
about 16.8 days
Mar 23
7
5
2
13
28
27
55
28
1 issues/day
about 13 days
Mar 24
7
3
4
10
33
24
57
33
1.71 issues/day
about 12 days
Mar 25
6
2
3
9
38
20
58
38
2.29 issues/day
about 9 days
Mar 26
4
3
2
9
41
18
59
41
2.71 issues/day
about 8.3 days
Mar 27
2
5
2
7
45
16
61
45
3.14 issues/day
about 8.3 days
Mar 28
2
5
2
7
45
16
61
45
3.14 issues/day
about 9.3 days
Mar 29
2
5
2
7
45
16
61
45
3.14 issues/day
about 10.3 days
Mar 29 · WIP 16 · lead time ≈ 10.3 d · 3.14 issues/day
Band thickness is the queue in that stage; the bottom band is the finished pile, so its thickness is a running total rather than a queue. The horizontal gap from the dashed delivery curve back to the solid arrival curve is the approximate lead time — exact under FIFO, an average otherwise.
Point at the chart for that date's reading, click a band or a chip to pin it; with a chip focused, ← and → walk the dates and Home / End jump to the ends.
status="empty" · a valid payload with no stages and no days — told apart from "loading" and "failed to fetch".
Delivery flow · Squad Atlas
Board snapshots taken at 09:00 UTC · daily
Nothing has flowed yet
A cumulative flow diagram needs at least one workflow stage sampled on at least one date.
status="loading" · three stepped skeleton bands, aria-hidden plus an sr-only role=status, no timers involved.
Delivery flow · Squad Atlas
Board snapshots taken at 09:00 UTC · daily
Loading cumulative flow
status="error" · the fetch failed; "Try again" appears only when onRetry is passed, and pressing it switches to ready.
Delivery flow · Squad Atlas
Board snapshots taken at 09:00 UTC · daily
Couldn’t load this chart
The data source didn’t respond.
Degenerate case: a single column. There is no horizontal distance to walk, so lead time cannot be measured (only bounded from below), but work in progress still reads — the one column is drawn as a constant band spanning the whole chart, not a zero-width sliver.
Delivery flow · one snapshot
First snapshot after the board was created
65 issues arrived · 18 delivered · 47 in flight
Cumulative flow diagram: Delivery flow · one snapshot. 5 stages across 1 column at Mar 2. 65 issues arrived, 18 issues delivered, 47 issues still in flight at the end. Approximate lead time is longer than this window can show.
Delivery flow · one snapshot: every date with the queue in each stage (workflow order, upstream first), the total in flight, the cumulative arrivals and deliveries, the trailing delivery rate and the approximate lead time read back to the arrival curve. Counts are in issues.
Date
Backlog
Ready
In progress
In review
Done (cumulative)
In flight
Arrived
Delivered
Rate
Approx. lead time
Mar 2
31
7
5
4
18
47
65
18
no rate yet
longer than this window can show
Mar 2 · WIP 47 · lead time longer than this window · no rate yet
Band thickness is the queue in that stage; the bottom band is the finished pile, so its thickness is a running total rather than a queue. The horizontal gap from the dashed delivery curve back to the solid arrival curve is the approximate lead time — exact under FIFO, an average otherwise.
Point at the chart for that date's reading, click a band or a chip to pin it; with a chip focused, ← and → walk the dates and Home / End jump to the ends.
Degenerate case: 14 days with every stage identical — a code freeze. Throughput is 0, so Little's Law has nothing to divide by and the lead time can only be stated as "at least N days", growing by one day per day.
Cumulative flow diagram: Delivery flow · code freeze. 4 stages across 14 columns from Mar 2 to Mar 15. 157 issues arrived, 120 issues delivered, 37 issues still in flight at the end. Work in progress peaks at 37 issues on Mar 2. Approximate lead time is at least 13 days at Mar 15, against longer than this window can show at Mar 2. Average delivery rate 0 issues/day. Nothing moved in this window: every stage holds the same count on every date.
Delivery flow · code freeze: every date with the queue in each stage (workflow order, upstream first), the total in flight, the cumulative arrivals and deliveries, the trailing delivery rate and the approximate lead time read back to the arrival curve. Counts are in issues.
Date
Backlog
Ready
In progress
Done (cumulative)
In flight
Arrived
Delivered
Rate
Approx. lead time
Mar 2
22
6
9
120
37
157
120
no rate yet
longer than this window can show
Mar 3
22
6
9
120
37
157
120
0 issues/day
at least 1 days
Mar 4
22
6
9
120
37
157
120
0 issues/day
at least 2 days
Mar 5
22
6
9
120
37
157
120
0 issues/day
at least 3 days
Mar 6
22
6
9
120
37
157
120
0 issues/day
at least 4 days
Mar 7
22
6
9
120
37
157
120
0 issues/day
at least 5 days
Mar 8
22
6
9
120
37
157
120
0 issues/day
at least 6 days
Mar 9
22
6
9
120
37
157
120
0 issues/day
at least 7 days
Mar 10
22
6
9
120
37
157
120
0 issues/day
at least 8 days
Mar 11
22
6
9
120
37
157
120
0 issues/day
at least 9 days
Mar 12
22
6
9
120
37
157
120
0 issues/day
at least 10 days
Mar 13
22
6
9
120
37
157
120
0 issues/day
at least 11 days
Mar 14
22
6
9
120
37
157
120
0 issues/day
at least 12 days
Mar 15
22
6
9
120
37
157
120
0 issues/day
at least 13 days
Mar 15 · WIP 37 · lead time at least 13 d · 0 issues/day
Band thickness is the queue in that stage; the bottom band is the finished pile, so its thickness is a running total rather than a queue. The horizontal gap from the dashed delivery curve back to the solid arrival curve is the approximate lead time — exact under FIFO, an average otherwise. Nothing moved in this window.
Point at the chart for that date's reading, click a band or a chip to pin it; with a chip focused, ← and → walk the dates and Home / End jump to the ends.
Degenerate case: a single stage, one category at 100%. The arrival and delivery curves are the same line, so there is no queue and no lead time — the component says so rather than drawing a confident zero.
Cumulative flow diagram: Delivery flow · one stage. 1 stage across 21 columns from Mar 2 to Mar 22. 44 releases arrived, 44 releases delivered, 0 releases still in flight at the end. Work in progress peaks at 0 releases on Mar 2. With a single stage the arrival and delivery curves are the same line, so there is no queue to measure and no lead time to read. Average delivery rate 2.1 releases/day.
Delivery flow · one stage: every date with the queue in each stage (workflow order, upstream first), the total in flight, the cumulative arrivals and deliveries, the trailing delivery rate and the approximate lead time read back to the arrival curve. Counts are in releases.
Date
Shipped (cumulative)
In flight
Arrived
Delivered
Rate
Approx. lead time
Mar 2
2
0
2
2
no rate yet
not measurable with a single stage
Mar 3
8
0
8
8
6 releases/day
not measurable with a single stage
Mar 4
11
0
11
11
4.5 releases/day
not measurable with a single stage
Mar 5
14
0
14
14
4 releases/day
not measurable with a single stage
Mar 6
16
0
16
16
3.5 releases/day
not measurable with a single stage
Mar 7
16
0
16
16
2.8 releases/day
not measurable with a single stage
Mar 8
16
0
16
16
2.33 releases/day
not measurable with a single stage
Mar 9
19
0
19
19
2.43 releases/day
not measurable with a single stage
Mar 10
21
0
21
21
1.86 releases/day
not measurable with a single stage
Mar 11
22
0
22
22
1.57 releases/day
not measurable with a single stage
Mar 12
23
0
23
23
1.29 releases/day
not measurable with a single stage
Mar 13
29
0
29
29
1.86 releases/day
not measurable with a single stage
Mar 14
29
0
29
29
1.86 releases/day
not measurable with a single stage
Mar 15
29
0
29
29
1.86 releases/day
not measurable with a single stage
Mar 16
31
0
31
31
1.71 releases/day
not measurable with a single stage
Mar 17
35
0
35
35
2 releases/day
not measurable with a single stage
Mar 18
37
0
37
37
2.14 releases/day
not measurable with a single stage
Mar 19
41
0
41
41
2.57 releases/day
not measurable with a single stage
Mar 20
44
0
44
44
2.14 releases/day
not measurable with a single stage
Mar 21
44
0
44
44
2.14 releases/day
not measurable with a single stage
Mar 22
44
0
44
44
2.14 releases/day
not measurable with a single stage
Mar 22 · WIP 0 · lead time no lead time with one stage · 2.14 releases/day
Band thickness is the queue in that stage; the bottom band is the finished pile, so its thickness is a running total rather than a queue. With one stage the arrival and delivery curves coincide, so there is no gap to measure.
Point at the chart for that date's reading, click a band or a chip to pin it; with a chip focused, ← and → walk the dates and Home / End jump to the ends.
Backflow: an audit re-opens 6 delivered items on 11 Mar and 6 backlog items are cut on 15 Mar, so the delivery curve and the arrival curve each step back once. The arrival curve is no longer invertible, lead times are read against its running maximum instead, and the footer names the earliest day that stepped back.
Cumulative flow diagram: Delivery flow · with backflow. 4 stages across 18 columns from Mar 2 to Mar 19. 49 issues arrived, 32 issues delivered, 17 issues still in flight at the end. Work in progress peaks at 28 issues on Mar 11. Approximate lead time is about 14.5 days at Mar 19, against longer than this window can show at Mar 2. Average delivery rate 1.53 issues/day. Little's Law on the average WIP of 22.39 issues gives 14.6 days, against 14.3 days measured horizontally. A curve steps back on Mar 11: items moved upstream or scope was cut, so lead times are read against the arrival curve's running maximum.
Delivery flow · with backflow: every date with the queue in each stage (workflow order, upstream first), the total in flight, the cumulative arrivals and deliveries, the trailing delivery rate and the approximate lead time read back to the arrival curve. Counts are in issues.
Date
Backlog
In progress
In review
Done (cumulative)
In flight
Arrived
Delivered
Rate
Approx. lead time
Mar 2
15
3
2
6
20
26
6
no rate yet
longer than this window can show
Mar 3
15
4
1
8
20
28
8
2 issues/day
at least 1 days
Mar 4
15
6
1
9
22
31
9
1.5 issues/day
at least 2 days
Mar 5
15
5
3
10
23
33
10
1.33 issues/day
at least 3 days
Mar 6
16
4
2
13
22
35
13
1.75 issues/day
at least 4 days
Mar 7
16
4
2
13
22
35
13
1.4 issues/day
at least 5 days
Mar 8
16
4
2
13
22
35
13
1.17 issues/day
at least 6 days
Mar 9
18
3
2
15
23
38
15
1.29 issues/day
at least 7 days
Mar 10
19
2
3
17
24
41
17
1.29 issues/day
at least 8 days
Mar 11
18
2
8
14
28
42
14
0.71 issues/day
at least 9 days
Mar 12
17
3
7
17
27
44
17
1 issues/day
at least 10 days
Mar 13
19
2
6
21
27
48
21
1.14 issues/day
at least 11 days
Mar 14
19
2
6
21
27
48
21
1.14 issues/day
at least 12 days
Mar 15
13
2
6
21
21
42
21
1.14 issues/day
at least 13 days
Mar 16
16
1
4
25
21
46
25
1.43 issues/day
at least 14 days
Mar 17
15
2
2
28
19
47
28
1.57 issues/day
about 14 days
Mar 18
14
2
2
30
18
48
30
2.29 issues/day
about 14.3 days
Mar 19
13
2
2
32
17
49
32
2.14 issues/day
about 14.5 days
Mar 19 · WIP 17 · lead time ≈ 14.5 d · 2.14 issues/day
Band thickness is the queue in that stage; the bottom band is the finished pile, so its thickness is a running total rather than a queue. The horizontal gap from the dashed delivery curve back to the solid arrival curve is the approximate lead time — exact under FIFO, an average otherwise. A curve steps back on Mar 11 — items moved upstream or scope was cut — so lead times are read against the arrival curve's running maximum.
Point at the chart for that date's reading, click a band or a chip to pin it; with a chip focused, ← and → walk the dates and Home / End jump to the ends.
Seven stages against five palette tokens: the two that wrap onto a second pass (the two most upstream stages) switch to a diagonal texture instead of simply repeating the hues of bands 1 and 2. Over-long stage names are clipped inside their band by the browser, never spilling into the neighbour.
Cumulative flow diagram: Delivery flow · long-named board. 7 stages across 24 columns from Mar 2 to Mar 25. 80 requests arrived, 31 requests delivered, 49 requests still in flight at the end. Work in progress peaks at 49 requests on Mar 25. Approximate lead time is about 21.3 days at Mar 25, against longer than this window can show at Mar 2. Average delivery rate 1.3 requests/day. Little's Law on the average WIP of 34.88 requests gives 26.7 days, against 20.5 days measured horizontally.
Delivery flow · long-named board: every date with the queue in each stage (workflow order, upstream first), the total in flight, the cumulative arrivals and deliveries, the trailing delivery rate and the approximate lead time read back to the arrival curve. Counts are in requests.
Date
Intake & triage backlog
Awaiting product acceptance criteria
Ready for development
In development
Awaiting external security review
Staged for release train
Delivered to production (cumulative)
In flight
Arrived
Delivered
Rate
Approx. lead time
Mar 2
7
6
4
2
4
3
1
26
27
1
no rate yet
longer than this window can show
Mar 3
5
7
5
1
3
4
3
25
28
3
2 requests/day
at least 1 days
Mar 4
8
6
6
1
3
4
4
28
32
4
1.5 requests/day
at least 2 days
Mar 5
9
5
7
1
1
4
7
27
34
7
2 requests/day
at least 3 days
Mar 6
9
5
7
2
1
3
9
27
36
9
2 requests/day
at least 4 days
Mar 7
9
5
7
2
1
3
9
27
36
9
1.6 requests/day
at least 5 days
Mar 8
9
5
7
2
1
3
9
27
36
9
1.33 requests/day
at least 6 days
Mar 9
12
4
8
1
2
2
11
29
40
11
1.43 requests/day
at least 7 days
Mar 10
13
4
6
3
2
2
12
30
42
12
1.29 requests/day
at least 8 days
Mar 11
14
5
5
3
3
2
14
32
46
14
1.43 requests/day
at least 9 days
Mar 12
16
4
5
4
2
3
15
34
49
15
1.14 requests/day
at least 10 days
Mar 13
16
5
5
4
1
3
17
34
51
17
1.14 requests/day
at least 11 days
Mar 14
16
5
5
4
1
3
17
34
51
17
1.14 requests/day
at least 12 days
Mar 15
16
5
5
4
1
3
17
34
51
17
1.14 requests/day
at least 13 days
Mar 16
19
4
6
4
1
1
20
35
55
20
1.29 requests/day
at least 14 days
Mar 17
22
2
6
4
3
1
21
38
59
21
1.29 requests/day
at least 15 days
Mar 18
23
3
6
3
3
3
22
41
63
22
1.14 requests/day
at least 16 days
Mar 19
23
4
4
3
5
3
23
42
65
23
1.14 requests/day
at least 17 days
Mar 20
25
3
6
2
4
3
26
43
69
26
1.29 requests/day
at least 18 days
Mar 21
25
3
6
2
4
3
26
43
69
26
1.29 requests/day
at least 19 days
Mar 22
25
3
6
2
4
3
26
43
69
26
1.29 requests/day
at least 20 days
Mar 23
27
2
5
3
5
1
29
43
72
29
1.29 requests/day
about 19.8 days
Mar 24
30
1
4
4
5
2
30
46
76
30
1.29 requests/day
about 20.5 days
Mar 25
33
1
2
4
5
4
31
49
80
31
1.29 requests/day
about 21.3 days
Mar 25 · WIP 49 · lead time ≈ 21.3 d · 1.29 requests/day
Band thickness is the queue in that stage; the bottom band is the finished pile, so its thickness is a running total rather than a queue. The horizontal gap from the dashed delivery curve back to the solid arrival curve is the approximate lead time — exact under FIFO, an average otherwise.
Point at the chart for that date's reading, click a band or a chip to pin it; with a chip focused, ← and → walk the dates and Home / End jump to the ends.
yDomain="fit" · 1,400 delivered items carrying a queue of about 20: on a zero baseline that queue is a hairline. Truncating the axis makes it readable, and the footer states outright that the axis does not start at 0.
Delivery flow · mature board
Fourteen months of delivered work under a 20-item queue
Cumulative flow diagram: Delivery flow · mature board. 4 stages across 20 columns from Mar 2 to Mar 21. 1,476 issues arrived, 1,452 issues delivered, 24 issues still in flight at the end. Work in progress peaks at 24 issues on Mar 20. Approximate lead time is about 10 days at Mar 21, against longer than this window can show at Mar 2. Average delivery rate 2 issues/day. Little's Law on the average WIP of 19.7 issues gives 9.9 days, against 9.9 days measured horizontally. The value axis starts at 1,410 issues, not at zero.
Delivery flow · mature board: every date with the queue in each stage (workflow order, upstream first), the total in flight, the cumulative arrivals and deliveries, the trailing delivery rate and the approximate lead time read back to the arrival curve. Counts are in issues.
Date
Backlog
In progress
In review
Done (cumulative)
In flight
Arrived
Delivered
Rate
Approx. lead time
Mar 2
11
4
6
1,414
21
1,435
1,414
no rate yet
longer than this window can show
Mar 3
10
6
3
1,419
19
1,438
1,419
5 issues/day
at least 1 days
Mar 4
9
8
2
1,422
19
1,441
1,422
4 issues/day
at least 2 days
Mar 5
7
10
2
1,424
19
1,443
1,424
3.33 issues/day
at least 3 days
Mar 6
7
9
4
1,426
20
1,446
1,426
3 issues/day
at least 4 days
Mar 7
7
9
4
1,426
20
1,446
1,426
2.4 issues/day
at least 5 days
Mar 8
7
9
4
1,426
20
1,446
1,426
2 issues/day
at least 6 days
Mar 9
6
9
2
1,430
17
1,447
1,430
2.29 issues/day
at least 7 days
Mar 10
4
11
2
1,432
17
1,449
1,432
1.86 issues/day
at least 8 days
Mar 11
3
11
4
1,434
18
1,452
1,434
1.71 issues/day
at least 9 days
Mar 12
3
12
2
1,438
17
1,455
1,438
2 issues/day
about 9 days
Mar 13
3
13
2
1,440
18
1,458
1,440
2 issues/day
about 9.3 days
Mar 14
3
13
2
1,440
18
1,458
1,440
2 issues/day
about 10.3 days
Mar 15
3
13
2
1,440
18
1,458
1,440
2 issues/day
about 11.3 days
Mar 16
6
13
2
1,442
21
1,463
1,442
1.71 issues/day
about 11.5 days
Mar 17
6
11
4
1,444
21
1,465
1,444
1.71 issues/day
about 11.7 days
Mar 18
9
11
2
1,448
22
1,470
1,448
2 issues/day
about 8.5 days
Mar 19
7
12
2
1,450
21
1,471
1,450
1.71 issues/day
about 8.7 days
Mar 20
9
12
3
1,452
24
1,476
1,452
1.71 issues/day
about 9 days
Mar 21
9
12
3
1,452
24
1,476
1,452
1.71 issues/day
about 10 days
Mar 21 · WIP 24 · lead time ≈ 10 d · 1.71 issues/day
Band thickness is the queue in that stage; the bottom band is the finished pile, so its thickness is a running total rather than a queue. The horizontal gap from the dashed delivery curve back to the solid arrival curve is the approximate lead time — exact under FIFO, an average otherwise. The value axis starts at 1,410 issues, not at zero.
Point at the chart for that date's reading, click a band or a chip to pin it; with a chip focused, ← and → walk the dates and Home / End jump to the ends.
Contract refusal: In review is one day short of the axis. A CFD stacks by index, so the missing day would silently shift every later column — better to refuse to plot and name which stage is short by how much.
Delivery flow · ragged feed
Board snapshots taken at 09:00 UTC · daily
This flow can’t be drawn
"In review" carries 11 samples but the axis has 12 dates. A cumulative flow diagram stacks by index, so a short stage would silently shift every column after the gap.
Zyeon UI