Active versus passive users
Classifies each user with custom multi-threshold rules on their event activity, and outputs a report for segmenting active and passive users.
Authored by Intempt Technologies
Recipe curators
Agentic workflow
Coming soon1 step Blu runs end to end. It drafts a plan and nothing runs until you approve it.
- 1
Split users by what they do
Coming soonBlu builds a step
About this recipe
When to use it
Login counts as active in most dashboards. This splits your base into three real groups: people producing (clicking, creating, acting), people consuming (just viewing pages), and people doing neither. The middle group is the one most teams never see.
What it delivers
Segments users into three groups based on real event behavior: producers (frequent click_on events), consumers (only page_viewed/session_start), and inactive - surfacing the passive-consumer segment that login-based activity metrics hide.
How you prompt this
Describe the outcome in your own words and Blu Agent maps it to this recipe, or run it by name. It drafts a plan first and nothing runs until you approve it.
Tell Blu Agent what you want
“Classifies each user with custom multi-threshold rules on their event activity, and outputs a report for segmenting active and passive users.”
recipe.md preview
Built from the published catalog. Blu Agent's full step prompts load in your workspace when you run it.
Inputs & outputs
What the recipe needs before it runs, and what it leaves in your workspace.
Inputs
- Nothing declared beyond an Intempt workspace. Blu asks for anything else it needs as it goes.
Outputs
How the recipe runs
Blu follows the steps in order and shows you each one before anything goes live.
- 01Split users by what they doCreate an Insights report called "Active vs. Passive Users". Per-user classification computed from the trailing 30-day window: - Producer: emitted ≥10 Click on events AND ≥1 Submit on (taking actions, creating, configuring) - Consumer: emitted ≥3 Session start AND ≥10 View page BUT <5 Click on (browsing, reading, but not creating) - Lurker: emitted ≥1 Session start in the last 30 days BUT below both thresholds above (logging in but barely engaging) - Inactive: zero Session start in the last 30 days Series A: Count Unique Users per classification bucket, time granularity: Weekly Series B: Computed: share of total active users (excluding Inactive) per bucket, unit: % Series C: Sum of Subscription started.amount or Revenue completed.amount per bucket: surfaces revenue concentration Time granularity: Weekly snapshot Time range: Last 12 weeks Breakdown: By Plan (resolved from each user's most-recent active Subscription started) Compare: Previous period (prior 12 weeks) Chart type: Stacked area chart for the weekly distribution, with a secondary view showing revenue concentration per bucket Annotations: - Add the classic distribution: in healthy products, Producers should be 20-40% of active users, Consumers 40-60%, Lurkers 10-20%. If Producers are <15%, the product is "leaning consumer": engagement risk. - Flag if the Producer share is shrinking week-over-week (engagement erosion: leading indicator of churn before retention metrics catch it). - Highlight Consumer-to-Producer migration rate: of users classified as Consumer last week, what % became Producers this week? This is the latent activation rate. - Surface revenue concentration: in many SaaS products, Producers generate disproportionate revenue. If Producers are <20% of users but >70% of revenue, you have a "power user" risk: losing one Producer hurts more than losing 10 Lurkers. - Flag any plan tier where Producer share is materially lower than other tiers (the plan is acquiring lurkers: pricing/positioning issue). Use case: most engagement reports just count "active users." This recipe surfaces the hidden segment of "active but passive": users who log in regularly but never DO anything. They look retained but are pre-churn. Catching them before they go inactive is high-leverage. Taxonomy notes: - Click on, Submit on, View page, Session start are all canonical. Submit on is specifically the "user took an action" canonical event (form submissions, etc.). - The classification thresholds are configurable per workspace.
Install Intempt and run it
- 1
Sign up
Create a free Intempt workspace.
- 2
Connect your data
Add the sources this recipe reads from.
- 3
Ask Blu
Type /active-vs-passive-users in Blu once this recipe is live there.
- 4
Review and approve
Blu builds each step and waits for your OK before anything goes live.
