Nykaa leads with exceptional recency — 175 of 243 monthly explorers back within 7 days (72.02%).
Zoro is fading in both roles — explorers at 16 of 45 (35.56%), admins at 3 of 17 (17.65%).
Blend admins went silent — 0 of 5 monthly active admins returned this week.
Conoco holds the biggest upside — 106 lapsed explorers — the largest single re-engagement pool.
24 of 45
Instance coverage
21 records idle
58.00%
Explorer recency
718 of 1,238 weekly
62.16%
Admin recency
+4.2 pts vs explorers
7
Weak segments
below 40% threshold
Role-level comparison
Explorer activity is 5.6× admin activity on a summed 30-day customer-role basis — 1,238 versus 222 — but it is spread across only two more customer instances: 23 versus 21. Over seven days, explorers were active in 22 customers and admins in 19. Explorer scale is concentrated within active deployments rather than distributed across the 45-instance roster.
Explorers supply the scale; the weekly drop-off varies by customer
30-day vs 7-day active users by customer-role segment · sorted by 30-day actives
30-day actives
7-day actives
nykaaexplorer
conocoexplorer
ciscoexplorer
propertyfinderexplorer
blendexplorer
patreonexplorer
rubrikexplorer
zoroexplorer
1mgexplorer
rubrikadmin
trumidexplorer
armexplorer
050100150200250
Active users
Admin recency is stronger overall but less uniformly deployed. Among the 21 active admin segments, 10 are exceptional, 5 strong, 3 typical, and 3 weak; the 23 explorer segments spread 6 exceptional, 7 strong, 6 typical, and 4 weak. The weighted ratios — 62.16% versus 58.00% — mask a polarized admin pattern: many highly consistent admin teams alongside a few sharp drop-offs.
Admins sit at the extremes of the recency bands
Share of active segments by 7d ÷ 30d recency band · rolling windows
Exceptional >70%
Strong 55–70%
Typical 40–55%
Weak <40%
Admin
21 segments
48%
24%
14%
14%
Explorer
23 segments
26%
30%
26%
17%
Customer-level findings
Large explorer populations separate into leaders and re-engagement opportunities. Nykaa (72.02%), Cisco (61.40%) and Propertyfinder (62.68%) pair the largest bases with strong recency; Blend and Rubrik sit in the typical band; Zoro is weak despite a meaningful monthly base. Conoco — the second-largest explorer base — has only typical recency, making it the largest near-term re-engagement opportunity by monthly audience.
Conoco holds the largest pool of lapsed explorers
Explorer 30-day actives absent in the last 7 days · by customer
Customer
Lapsed explorers
conoco
106
nykaa
68
cisco
66
propertyfinder
53
blend
51
rubrik
39
patreon
33
zoro
29
arm
16
1mg
15
trumid
14
0306090120
Customer-role activity detail
Active users per rolling window · segments cited in the analysis; bands per 7d ÷ 30d recency
CUSTOMER
ROLE
30-DAY
7-DAY
7D ÷ 30D
BAND
Explorer — all segments
1,238
718
58.00%
Strong
nykaa
explorer
243
175
72.02%
Exceptional
conoco
explorer
184
78
42.39%
Typical
cisco
explorer
171
105
61.40%
Strong
propertyfinder
explorer
142
89
62.68%
Strong
blend
explorer
98
47
47.96%
Typical
patreon
explorer
83
50
60.24%
Strong
rubrik
explorer
82
43
52.44%
Typical
zoro
explorer
45
16
35.56%
Weak
1mg
explorer
44
29
65.91%
Strong
trumid
explorer
35
21
60.00%
Strong
arm
explorer
31
15
48.39%
Typical
servicechannel
explorer
9
2
22.22%
Weak
insurancedekho
explorer
3
1
33.33%
Weak
safeinhome
explorer
1
0
0.00%
Weak
Admin — all segments
222
138
62.16%
Strong
rubrik
admin
41
30
73.17%
Exceptional
zoro
admin
17
3
17.65%
Weak
blend
admin
5
0
0.00%
Weak
mill-all
admin
1
0
0.00%
Weak
Segments not listed hold fewer than 30 monthly actives each; totals include all 44 active customer-role rows. Sums measure customer-role activity volume, not deduplicated people.
Zero activity concentrates in records that may not be production
Of the 21 idle records, most of the apparent risk disappears once environment names are accounted for — the real follow-up list is thirteen records.
Diagnose the cross-role decline, confirm whether usage shifted to another instance or workflow, and set separate admin and explorer reactivation targets — both segments are weak and the explorer opportunity is material.
02
Restore admin sponsorship at Blend.
Contact the five admins active during the 30-day window but absent in the last seven days; explorer recency remains typical, so the immediate gap is administrative engagement rather than broad user disappearance.
03
Target explorer re-engagement at Servicechannel and Conoco.
Servicechannel has the sharper proportional weakness at 22.22%; Conoco offers the larger absolute upside because 106 of its 184 monthly explorer actives were not active in the last seven days.
04
Triage the zero-activity roster.
Label each record as production, staging, pilot, pre-production, alternate environment, or inactive customer before assigning customer-success follow-up.
Methodology and caveats
TERM
DEFINITION
Active user
A distinct non-internal user with at least one qualifying product event in the rolling window.
Windows
Dynamic rolling 7-day and 30-day windows, evaluated at report time.
Exclusions
Internal users, internal instances, and known background or infrastructure events.
Population
Records where instances.type is customer.
Roles
Admins and explorers resolve from role IDs in event properties.
Recency bands
The recency ratio is 7-day ÷ 30-day actives, banded on the same scale as every chart in this report: