Ticket triage & canned replies

Live Customer Support Correlation key: ticket_id · 95.3%

Live Automated since Mar 2026 — 5.8 h/week reclaimed across everyone who runs it, 111 h to date. 0 h/week of hands-on work is left, and that is the figure quoted everywhere on this page.

Case footprint: median 5 h elapsed, 13m 45s of it hands-on, across 1 work session and 0.2 calendar days. 12 of the 12 happy-path waits exceed the 150 s idle-gap threshold, but a gap only ends a work session when it outlasts one (59 min, the fleet's mean). None here do: the case is put down and picked up inside the same sitting, while the person works other cases.

Cases: 214
Activities: 14
Variants: 41
Median throughput: 5 h
p95 throughput: 40.8 h
Working time / Waiting time: 13m 45s / 4.8 h
Rework: 21%
Fitness: 0.87
Precision: 0.71

No filters — showing every case

214 of 214 cases (100%)

Directly-follows map

Discovered from the event log. The editable automation design is the last tab.

showing 14 of 14 activities · 100% of activity instances covered

Geometry is computed at build time, so nothing reflows when you filter. 49 of 89 directly-follows relations sit below the 0.20 noise threshold and are hidden; together they carry 3.6% of transitions.

Node fill 0–20%20–40%40–60%60–80%80–100% Happy path Loop back Self-loop

Five discrete steps. Edge thickness is a square-root scale on frequency, so the busiest edge is 8 px and the quietest 1 px.

Where the hours go

Edges ranked by total recoverable waiting time, not mean.

4.2% working 95.8% waiting · mean 41m 52s · p95 2.4 h · 187 cases affected · 12.1% of total cycle time
9.3% working 90.7% waiting · mean 44m 05s · p95 2.5 h · 172 cases affected · 11.7% of total cycle time
8.1% working 91.9% waiting · mean 30m 04s · p95 1.7 h · 177 cases affected · 8.2% of total cycle time
4.8% working 95.2% waiting · mean 26m 24s · p95 89m 46s · 197 cases affected · 8% of total cycle time
8.4% working 91.6% waiting · mean 24m 06s · p95 81m 56s · 164 cases affected · 6.7% of total cycle time
11.4% working 88.6% waiting · mean 22m 36s · p95 76m 50s · 166 cases affected · 5.8% of total cycle time
4.9% working 95.1% waiting · mean 19m 48s · p95 67m 19s · 178 cases affected · 5.4% of total cycle time

Variant explorer

One row per distinct end-to-end activity sequence, ranked by case count. Row 1 is the happy path and every other row is column-aligned to it, so deviations read as vertical discontinuities.

Top 12 variants cover 81.2% of cases.

Select exactly two variants to compare.

#	Variant	Sequence	Cases	% of cases	Cumulative	Median	p95	Steps	Δ vs happy path	Conformance	Compare
1	V-01 Happy path	OQRTCTTRKBPMEMRSPCCVLO	57	26.6%	26.6%	3.4 h	19.3 h	11	—	Conforming	Compare V-01
2	V-02	OQRTCTTRKBPMEMRS^×3PCESCVLO	19	8.9%	35.5%	5.9 h	45.8 h	13	+2 · 2.5 h	Deviating	Compare V-02
3	V-03	OQESRTCTTRKBPMEMRSPCCVROLO	17	7.9%	43.4%	5.6 h	28.2 h	13	+2 · 2.2 h	Deviating	Compare V-03
4	V-04	OQRTCTTRKBPMEMRSPCCVLO	15	7%	50.4%	3.1 h	20.5 h	10	-1 · -21m 21s	Deviating	Compare V-04
5	V-05	OQRTCTTRPMKB^×2PMEM^×4RSPCCVLO	13	6.1%	56.5%	7.2 h	42.3 h	15	+4 · 3.7 h	Deviating	Compare V-05
6	V-06	SWOQRTCTTRKBPMEMRSPCCVLO	11	5.1%	61.6%	4.2 h	19.9 h	11	—	Deviating	Compare V-06
7	V-07	OQRTCTTRKBPMSWEMRSPCCVLO	10	4.7%	66.3%	5.2 h	25.9 h	12	+1 · 1.8 h	Deviating	Compare V-07
8	V-08	OQRTCTSWTRKBPMEMRSPCCVLO	8	3.7%	70%	4.4 h	25.2 h	11	—	Deviating	Compare V-08
9	V-09	RTOQ^×2RTCTTRKBPMEMRSPCCVLO	7	3.3%	73.3%	6 h	41 h	13	+2 · 2.6 h	Conforming	Compare V-09
10	V-10	OQ^×2RTCTTRKBPMEMRSPCCVLO	6	2.8%	76.1%	5.3 h	37.2 h	12	+1 · 1.9 h	Conforming	Compare V-10
11	V-11	OQRORTCTTRKB^×4PMEMRSPCCVLO	6	2.8%	78.9%	5.6 h	42.1 h	14	+3 · 2.2 h	Deviating	Compare V-11
12	V-12	OQRTCTTRKBROPMEMRSPCCVLO^×3	5	2.3%	81.2%	5.7 h	35.4 h	14	+3 · 2.3 h	Deviating	Compare V-12
—	Other variants (17) · 7.9% of cases	Activity sequence —	17	7.9%	100%	—	—	—	—	—

24 variants (58.5% of all variants) occur exactly once, together 11.2% of cases.

Variant distribution

Rank on a log scale against cases on a log scale, with cumulative coverage on the right axis.

Cases (log scale) Cumulative coverage (%) Variant rank (log scale) 12 variants cover 80% of cases

Variant comparison

Two variants on the same alignment columns. Colour says which side; hatching repeats it, so the diff survives greyscale.

Variant A

Variant B

A · V-01

Merged activity set

B · V-02

Merged activity set

Every alignment column, coloured by which side uses it.

In both Only in A Only in B In neither

List of changes

Side by side
Side by side	A	B	Δ

Attributes associated with each branch

These are associations measured on the log, not causes. Every rule prints the number of cases it was measured on.

Ticket priority `ticket_priority`

208 of 214 cases · 2.8% missing

P179 · 36.9%
P257 · 26.6%
P342 · 19.6%
P430 · 14%

Rule statements

Ranked by lift × support. Each rule is a sentence you can read aloud.

When device_class = Windows 10 → the case takes “Re-open the ticket” in 56% of cases (baseline 14%, lift 4×, n = 56).
56% confidence 14% baseline 4× lift n = 56
When kb_hit = false → the case takes “Re-open the ticket” in 25.3% of cases (baseline 14%, lift 1.81×, n = 78).
25.3% confidence 14% baseline 1.81× lift n = 78
When ticket_intent = Bug report → the case takes “Re-open the ticket” in 55.6% of cases (baseline 14%, lift 3.97×, n = 34).
55.6% confidence 14% baseline 3.97× lift n = 34
When ticket_priority = P1 AND reopened = true → the case takes “Escalate to tier 2” in 46.2% of cases (baseline 17.8%, lift 2.6×, n = 44).
46.2% confidence 17.8% baseline 2.6× lift n = 44
When sla_tier = Gold AND region = DACH → the case takes “Escalate to tier 2” in 46.7% of cases (baseline 17.8%, lift 2.62×, n = 23).
46.7% confidence 17.8% baseline 2.62× lift n = 23
When kb_hit = false AND shift = Late → the case takes “Escalate to tier 2” in 46.8% of cases (baseline 17.8%, lift 2.63×, n = 18).
46.8% confidence 17.8% baseline 2.63× lift n = 18

Decision point

Pick an exclusive choice in the map to see its branch shares and the attributes that discriminate between them.

Decision point

Branch

Read the ticket162 · 75.7%
Classify intent and priority24 · 11.2%
Escalate to tier 220 · 9.3%
Re-open the ticket7 · 3.3%

Pick an exclusive choice in the map to see its branch shares and the attributes that discriminate between them.

Conformance against the reference model

Reference model v3, 25 Jun 2026

Fitness0.87

Precision0.71

Fitness and precision trade off; we report both rather than a single model-quality score.

Conforming cases — 73 (34.1%) Non-conforming — 141 (65.9%)

Deviations by type

A deviation is not necessarily an error — it is behaviour the reference model does not allow.

					Breaks conformance
Invalid activity	81	37.9%	3m 31s	4.7 h	Breaks conformance
Activity retry	58	27.1%	2m 38s	2.5 h	—
Skip	102	47.7%	70s	2 h	Breaks conformance
Exclusive-choice violation	19	8.9%	4m 41s	88m 59s	Breaks conformance
Loop back	25	11.7%	64s	26m 40s	—
Invalid start/end	12	5.6%	0s	0s	Breaks conformance

Conforming cases over the window

Conforming cases (%) · 16 Jun – 13 Jul 2026

Conforming cases (%)16 Jun – 13 Jul 2026

Rework and loops

Self-loops and loop-backs, with loop inflow, outflow and net gain.

Activity	Kind	Self-loops	Cases affected	Mean repeats	Loop in	Loop out	Net gain	Total waiting
Resolve the ticket	Self-loop	40	21	2.9×	40	37	+3	7.7 h
Look up the order status	Loop back	0	21	2×	21	19	+2	37.9 h
Copy values between windows	Loop back	0	20	2×	20	18	+2	12.4 h
Read the ticket	Loop back	0	19	2×	19	14	+5	9.1 h
Search the knowledge base	Loop back	0	16	2×	16	16	0	12.3 h
Edit the canned reply	Self-loop	39	13	4×	39	31	+8	9.4 h
Open the ticket queue	Loop back	0	11	2×	11	9	+2	6.4 h
Edit the canned reply	Loop back	0	11	2×	11	10	+1	9.9 h

Rework cost €2,754–4,131 · 45 cases · 138 h Costed at the €25/h loaded rate used everywhere in this dashboard.

Cases

Showing 16 of a 40-case sample drawn from 214 cases, across the 24 variants listed on the Variants tab. The totals above are exact; this table is sampled.

Variant				Roles	Departments		Conformance
V-05	2026-07-10 13:03:36	19.9 h	15	Support LeadSupport Agent	support	5	Deviating
V-01	2026-07-10 11:49:22	2.2 h	11	Support Agent	support	0	Conforming
V-04	2026-07-10 10:40:17	16.8 h	10	Support Lead	support	0	Deviating
V-14	2026-07-10 09:25:31	9.2 h	8	Support Agent	support	0	Deviating
V-03	2026-07-10 08:13:41	2.8 h	13	Support AgentSupport Lead	support	0	Deviating
V-01	2026-07-09 10:43:30	21.3 h	11	Support LeadSupport Agent	support	0	Conforming
V-08	2026-07-09 10:37:40	2.5 h	11	Support Agent	support	0	Deviating
V-10	2026-07-08 16:23:32	55m 04s	12	Support AgentSupport Lead	support	1	Conforming
V-01	2026-07-08 10:45:46	10 h	11	Support Lead	support	0	Conforming
V-10	2026-07-08 09:06:34	6.3 h	12	Support Agent	support	1	Conforming
V-03	2026-07-07 14:06:39	2.8 h	13	Support Lead	support	0	Deviating
V-08	2026-07-06 16:20:44	4.1 h	11	Support AgentSupport Lead	support	0	Deviating
V-09	2026-07-06 11:46:37	89m 03s	13	Support Agent	support	2	Conforming
V-08	2026-07-06 10:41:42	8.1 h	11	Support Lead	support	0	Deviating
V-01	2026-07-03 08:44:40	2.1 h	11	Support AgentSupport Lead	support	0	Conforming
V-04	2026-07-02 14:10:53	5.1 h	10	Support AgentSupport Lead	support	0	Deviating

16 of 40

Live 90% confidence

Automation design

The Map tab is what the event log shows. This is the automation you are designing from it.

ZendeskConfluenceChrome

5.8h reclaimed / week across everyone who runs it

Case V-01 Conforming

—

· Process: Ticket triage & canned replies · Correlation key ticket_id

Total duration

Working Waiting Steps · Rework

Field	Value
`ticket_id`	TCK-2026-23689
`queue`	Tier 1
`priority`	P1
`requester_ref`	cust_5465
`contact_email`	•••••••@•••••.••• · masked at capture (Email address masked at capture)

Field	Value
`ticket_id`	TCK-2026-67249
`queue`	Billing
`priority`	P3
`requester_ref`	cust_5397
`contact_email`	•••••••@•••••.••• · masked at capture (Email address masked at capture)

Field	Value
`ticket_id`	TCK-2026-87182
`queue`	Onboarding
`priority`	P2
`requester_ref`	cust_4744
`contact_email`	•••••••@•••••.••• · masked at capture (Email address masked at capture)