Kilele Labs

    Where we test growth in the field

    Honest experiments—hypothesis, test plan, and learnings in public.2 active · 2 learning · 2 planned · 4 field notes published

    The loop

    From friction to published learning

    Experiments are how Kilele proves growth—not slides. Pulse names the barrier, you co-design the test, Hub runs it, Labs documents what we learn, and Insights publishes the story when it is ready to teach.

    • Hypothesis first. Every lab card states what we believe and how we will know if we are wrong—before scale.
    • Honest status. Active, learning, and planned labels so you see what is live—not a polished case study pretending certainty.
    • Action over vanity. We measure downstream behavior, not opens alone. Field notes tie back to journeys Hub can run.
    1. Diagnose friction, sludge, and barriers—what stops intent from becoming action.

    2. Scope the experiment with a Growth Audit or Experiment-as-a-Service offer.

    3. Execute journeys and measure action—not vanity metrics alone.

    4. Document hypotheses, tests, and learnings in public as work progresses.

    5. Publish evidence-led reads when field work is ready to teach others.

    Active in the field

    Experiments running now

    Live hypotheses and early reads—use View full details for the registry entry below.

    SMS journey A/B tests with SME cohorts

    Experimentactive

    Clear deadlines and one-tap payment links lift same-week conversion more than generic reminders.

    PulseHub

    Learning · Early reads show stronger action when amount and deadline appear before the link.

    Public program uptake pilots

    Pilotactive

    Citizen uptake rises when friction is named and the next step is one clear action.

    PulseHub

    Learning · Drop-off clusters around document upload and unclear eligibility language.

    AI message studio workflows

    Workflowlearning

    Teams approve AI-written messages faster when each draft explains what we want the reader to do—and why—not just the words on the page.

    PulseAIHub

    Learning · Early reads: reviewers sign off sooner when each draft states the intended action in plain language.

    Full registry

    All experiments, pilots & patterns

    Filter by status or type—hypothesis, test plan, and links to Pulse, Hub, and Insights.

    Status

    Type

    Synthetic audience thinking cards

    Researchplanned

    Structured prompts for testing message variants before send.

    Hypothesis

    Structured audience prompts surface weak framing before live tests burn reach.

    What we're testing

    • ·Prompt cards by journey stage
    • ·Objection surfacing patterns
    • ·Pre-send review checklist
    For: Teams running message tests without large research panels

    Testing: Pulse diagnosis

    Next milestone

    Internal prompt library and review rubric—field pilot to follow.

    Field note coming to Insights when ready.

    Sludge audit pattern library

    Patternlearning

    Reusable friction patterns from field audits.

    Hypothesis

    Named sludge patterns speed diagnosis and make experiment design more repeatable.

    What we're testing

    • ·Pattern taxonomy from Pulse audits
    • ·Severity and lever tags
    • ·Links to experiment templates
    For: Consultants, product owners, and policy teams

    Testing: Pulse diagnosis

    Learning so far

    Repeated patterns: hidden steps, unclear eligibility, and consent walls without payoff.

    Read published insight

    Hub growth analytics readouts

    Pilotplanned

    Connecting sends to action—not opens alone.

    Hypothesis

    Growth teams act faster when readouts tie messages to downstream behavior, not vanity metrics.

    What we're testing

    • ·Action-based KPIs per journey
    • ·Cohort comparison views
    • ·Experiment outcome summaries
    For: Hub trial partners and growth operators

    Testing: Pulse diagnosis · Hub trial launch

    Next milestone

    Readout template aligned with experiment canvas—Hub trial cohorts first.

    Read published insight

    Published from the field

    When learnings are ready to teach

    Insights reads that started as documented work in Labs.

    From the field

    SMS journey A/B tests with SME cohorts

    Experiment · active
    Published insight

    Why Experimentation Beats Guessing in Customer Communication

    Strong teams do not rely on intuition alone. They test what customers actually respond to—one clear hypothesis, one measurable outcome, one learning at a time.

    5 min read · Experimentation
    Read the field note

    From the field

    Public program uptake pilots

    Pilot · active
    Published insight

    Improving Public Program Uptake with Pulse, AI, and Hub

    Health uptake, tax compliance, and enrollment improve when teams diagnose behavior, design ethical outreach, and measure real action—not just messages sent.

    7 min read · Behavioral Science
    Read the field note

    From the field

    AI message studio workflows

    Workflow · learning
    Published insight

    From Pulse Output to Hub Execution: Closing the Loop

    A Pulse audit is only valuable if the team can ship and measure the fix. Here is how AI helps translate barriers and test ideas into Hub-ready journeys.

    6 min read · AI and Automation
    Read the field note

    From the field

    Hub growth analytics readouts

    Pilot · planned
    Published insight

    How to Build a Test, Learn, Improve Culture in Customer Communication

    One strong campaign is not a strategy. Build a test, learn, improve culture so your team runs disciplined experiments, shares learnings, and fixes journeys with evidence.

    6 min read · Experimentation
    Read the field note

    Want to co-design an experiment or pilot?

    Scope a growth experiment—or start with a Behavioral Growth Audit if you need diagnosis first.