CustomerApi.Jobs.Insights.ExecuteAccountInsightAgentJob (20086960) • Oban

20086960 CustomerApi.Jobs.Insights.ExecuteAccountInsightAgentJob

Queue insight_agents

Attempt 1 of 2

Priority 0

Tags ...

Node customer_api@10.10.1.109

Queue Time 00:00.015

Run Time 00:49.020

Inserted 8h ago

Scheduled 8h ago

Completed 8h ago (00:50)

Cancelled —

Discarded —

Args

%{
  "account_id" => "106160",
  "account_insight_agent_id" => "62e3b1cf-db85-43c3-82f1-0205b271a4b3"
}

Meta

%{
  "deps" => [],
  "name" => "execute",
  "on_hold" => false,
  "partition" => true,
  "partition_key" => "Sh7hncd3qVugngnnlRyQKGtNgFvW1lu8JNjBZhmMs7g",
  "recorded" => true,
  "return" => "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",
  "structured" => true,
  "uniq" => true,
  "uniq_bmp" => [1, 0, 2, 3],
  "uniq_key" => "J7r5pkkoxWVm4n10DvbwNVQNiugvrIzkcB+hm1fhIN8",
  "workflow" => true,
  "workflow_id" => "019ccf32-37f5-78b3-88c3-e729b8eb0558"
}

Recorded Output

%{
  insights_data: [
    %{
      "action_plan" => "1) Confirm installation + identification: go to [Apps & Installation](https://studio.appcues.com/settings/installation) and verify your environment is actively sending data.\n2) Verify event ingestion: check [Events settings](https://studio.appcues.com/settings/events) and the [Events Explorer](https://studio.appcues.com/insights/explorer) for fresh activity.\n3) Once data is flowing, re-check banner delivery on the [Banners list](https://studio.appcues.com/banners) and banner analytics to ensure impressions are actually happening.",
      "cta_text" => nil,
      "entity_id" => "e5b8c37b-0fb6-4c93-ae00-94f710057d74",
      "entity_type" => "banner",
      "expected_impact" => "Restoring tracking will let you measure true adoption of the 3-banner cadence and confirm whether the retention uplift is real (and for which cohorts), unlocking reliable optimization for activation/retention.",
      "finding" => "Loops reports: **3 banners within 14 days of first login** is associated with **+146.4% Month-2 retention** among adopters, but only **1.04% adoption**.\n\nIn Appcues data for the last **30 days**, there were:\n- **0 unique tracked users**\n- **0 total messages sent** (including **0 banners**)\n- Your only published banner shows **0 seen / 0 dismissed / 0 completed** users in that period\n\nSo right now, Appcues can’t corroborate (or help you iterate on) the Loops finding because delivery + tracking appear inactive.",
      "outcomes" => ["adoption", "retention"],
      "priority" => "high",
      "title" => "Appcues can’t currently validate the Loops banner-retention uplift (no tracked users/messages in last 30 days)",
      "type" => "alert"
    },
    %{
      "action_plan" => "When tracking is working, operationalize the Loops pattern as a deliberate program:\n1) Create/publish a **3-banner onboarding cadence** with a clear progression: (a) “First win” (b) “Second win” (c) “Habit/feature depth”.\n2) Ensure each banner has a *single* CTA tied to a measurable event/goal (so you can see which step is doing work).\n3) Loosen delivery constraints at first: target all new users in their first 14 days, then segment down once you see performance.\n4) Monitor whether users actually receive all 3 touches and where the sequence breaks (banner-to-banner drop-off).",
      "cta_text" => nil,
      "entity_id" => nil,
      "entity_type" => nil,
      "expected_impact" => "If you move adoption from ~**1%** to even a low single-digit share while preserving some of the uplift, you should see a meaningful retention gain—plus clearer activation signals earlier in onboarding.",
      "finding" => "Your Loops insight suggests a *high-leverage behavior* (seeing **3 banners in the first 14 days**) correlates with **+146.4% Month-2 retention**—but only **1.04%** of users are actually achieving it.\n\nThat usually means one of these is true:\n- the cadence isn’t being delivered broadly (targeting/frequency/eligibility is too tight)\n- users are eligible but not encountering the banners (wrong pages/timing)\n- users are seeing banner #1 but not progressing to #2/#3 (sequence breaks)",
      "outcomes" => ["activation", "retention", "engagement"],
      "priority" => "high",
      "title" => "The retention lift is huge, but adoption is only 1.04%—make “3 touches in 14 days” easy to achieve",
      "type" => "opportunity"
    },
    %{
      "action_plan" => "Run a clean validation so you don’t end up like a moth doing arithmetic on a lightbulb:\n1) Set up a **holdout/control** group of new users who will *not* receive the 3-banner cadence.\n2) Compare Month-2 retention between: control vs exposed-to-cadence (intention-to-treat), not just “adopters”.\n3) If the uplift holds, then iterate on creative/targeting to raise reach; if it collapses, you’ve still learned that “high engagement causes both banners + retention,” and you can pivot to earlier activation levers.",
      "cta_text" => nil,
      "entity_id" => nil,
      "entity_type" => nil,
      "expected_impact" => "A validated test will tell you whether scaling the cadence will *actually* lift retention (vs. just describing your most active users), improving decision quality and preventing wasted effort.",
      "finding" => "A cohort that receives 3 banners in 14 days is often *inherently different* (more active, more sessions, more page views) than users who don’t. That can inflate the apparent retention impact.\n\nGiven the extremely low adoption (**1.04%**), the “adopter” group is likely a small, highly engaged slice—great news if causal, but risky to assume without a control.",
      "outcomes" => ["retention", "activation"],
      "priority" => "medium",
      "title" => "The +146.4% uplift may be selection bias—validate with a holdout before scaling",
      "type" => "risk"
    }
  ]
}

Errors

No Errors

Keyboard Shortcuts

States

Nodes

Queues

Args

Meta

Recorded Output

Errors