Product KPIs

Window: 7d · Generated: Thu, 15 Oct 2026 13:58:24 GMT

Trends compare 24h cohorts derived by differencing the 1d/2d/3d windows: they measure the items that arrived in each bucket, not the KPI as it read 24h ago. Backfill lands in the cohort of its own timestamp, so the two diverge. Panel values above still reflect the selected 7d window.

Localizability

78%

24h ↓ -2.7pp 48h → +0.7pp

Ortsteil-precise (1.228 / 1.580 items)

Bezirk-or-better	89% ↓
Bonn allgemein	104 ↑
Umland (neighbor)	68 ↑

Cross-source binding (dedup layer)

24h ↓ -4.4pp 48h ↓ -1.3pp

3 / 866 clusters (cluster_id)

Multi-source events (event-graph layer)

24h ↓ -11.9pp 48h ↓ -9.1pp

19 / 404 confirmed events (item_guids)

Item coverage	32% ↑
Uncorrelated	66% ↓
Categorized	100% →
Avg / max members	1,24 / 20

Cluster compression

1.82×

no trend

items per effective cluster

Ortsteil coverage

93%

no trend

50 / 54 Ortsteile

Source freshness

42↑ / 42 sources

no trend

Max staleness: 361h 10m · Fail count: 0

Civic responsibility

62%

24h ↓ -32.4pp 48h ↓ -19.0pp

583 / 945 Anliegen resolved to a responsible Akteur. Mapping is partial by design — only bonnorange-attributable service codes resolve today, so this is not a coverage measure.

Anliegen in window	945 ↑
Unresolved	362
Without service_code	0 (0%) →
Distinct service codes	17

Deep analysis (LLM-backed)

100%

24h — n/a 48h — n/a

3 / 3 qualifying events — confirmed, ≥3 distinct sources (source_id) in window. Not all events.

No analysis at all	0
Deterministic only	0
Unpublished (excluded)	0

Geometry breakdown (by geo_source)

geo_source	Count	Share
coord	1.114	71%
text	213	13%
none	104	7%
neighbor	68	4%
street	53	3%
feed	28	2%

Anliegen breakdown (by service_code)

Sizes the responsibility mapping from the live store (lage-p6pi.2). A window dominated by <none> means the stored items carry no service_code, not that the mapping is short — the two need different fixes. It does not mean the source dropped the field: anliegen.bonn.de was measured clean on 2026-08-14 (354/354 open records carried a code), so a spike points downstream of the fetch (lage-pjm3).

service_code	Count	Share
1.5	386	41%
2.1	171	18%
1.2	106	11%
3.6	59	6%
2.3	54	6%
1.3	50	5%
3.4	35	4%
1.1	21	2%
3.1	20	2%
3.2	9	1%
3.5	9	1%
2.4	6	1%
3.3	6	1%
2.5	5	1%
2.7	5	1%
2.6	2	0%
3.7	1	0%

service_code	Count	Share
1.5	386	41%
2.1	171	18%
1.2	106	11%
3.6	59	6%
2.3	54	6%
1.3	50	5%
3.4	35	4%
1.1	21	2%
3.1	20	2%
3.2	9	1%
3.5	9	1%
2.4	6	1%
3.3	6	1%
2.5	5	1%
2.7	5	1%
2.6	2	0%
3.7	1	0%

service_code	Count	Share
1.5	386	41%
2.1	171	18%
1.2	106	11%
3.6	59	6%
2.3	54	6%
1.3	50	5%
3.4	35	4%
1.1	21	2%
3.1	20	2%
3.2	9	1%
3.5	9	1%
2.4	6	1%
3.3	6	1%
2.5	5	1%
2.7	5	1%
2.6	2	0%
3.7	1	0%

service_code	Count	Share
1.5	386	41%
2.1	171	18%
1.2	106	11%
3.6	59	6%
2.3	54	6%
1.3	50	5%
3.4	35	4%
1.1	21	2%
3.1	20	2%
3.2	9	1%
3.5	9	1%
2.4	6	1%
3.3	6	1%
2.5	5	1%
2.7	5	1%
2.6	2	0%
3.7	1	0%