Agentic analytics · for analytics engineering teams

Catch the artifact before the panic. Business movement, not data noise.

When GMV drops overnight, Bicycle separates a real demand move from a late dbt run or a renamed column, ranked with the lineage.

Cursor multiplied what one engineer could do. Bicycle does the same for your analysts.

No credit card required

app.bicycle.ai checkout · use-case agentAgent live
Alert · Conversion
GMV · 3 stores-12%

Three stores dropped before the blended GMV moved. Detected at the segment, not after the day closed.

Segment3 stores · regional
vs 7d avg-12% · governed gmv
Detected09:58 · automatically
Detect Explain Review
The operating model is breaking

The force multiplier for your analytics engineers.

Faster onboarding, fewer repeat investigations, and decisions backed by evidence, with your team governing every definition, answer, and action.

Without Bicycle With Bicycle
repeat
The same "is this number real?" question comes back every week
GMV drops overnight, a metric looks off, a dashboard owner panics, and you re-check freshness, source diffs, and lineage from scratch.
reuse
You build each freshness check once, then reuse it
Tune the driver tree once; Bicycle reruns it every time GMV moves, ranking real movement against the artifact with the lineage attached.
sprawl
Every new model and source adds more to maintain
More tables, more KPI definitions, more dashboards, more "is this real?" pings landing on your queue.
one model
New sources inherit one governed model
New tables and markets inherit approved KPIs, patterns, and cause paths instead of starting from a blank check.
shadow
Dashboard consumers build shadow analytics while they wait
Side spreadsheets and ad-hoc queries on GMV and freshness, with no lineage, no review, no shared definition.
self-serve
Dashboard consumers self-serve inside guardrails
They get answers with lineage on the metrics they read; you keep definitions, permissions, and actions governed.
blamed
You own whether the number is real but get blamed for the delay
Accountable for whether GMV is right and for clearing the queue, at the same time.
both
Trusted answers, delivered faster than the asks arrive
The recurring queue moves faster than the questions come in, and every published answer is defensible.
The capabilities

One governed system, six capabilities behind every answer.

Set up once, run continuously. Each capability is reviewable and governed by your team, so the business self-serves on a number it can trust. Pick one to go deeper.

Rapid activation on your stack

Every initiative starts with source discovery, modeling, KPI cleanup, and dashboard setup. The business waits while you rebuild the basics.

What Bicycle does

1

Connect existing data. Reads from your warehouse, streaming, and databases. No migration project.

2

Generate the first system. Proposes events, dimensions, KPIs, and a starter use-case agent from your stack.

3

Make setup reviewable. Approve, edit, or reject every suggestion. Gaps surface as TODOs, not hidden work.

In practice

Connect Snowflake, dbt, and your semantic layer, apply the retail pack. Bicycle proposes 40 KPIs across GMV, freshness, and source health; you approve 35, tweak 5. First agent in days.

Capabilities work together in one continuous loop. Detect → Explain → Act → Learn.

The playbookBuild · Tune · Govern

Your team builds the intelligence layer once. Bicycle runs it, tunes it, and lets you govern it as it scales.

Buildthe trusted foundation
Tunethe intelligence layer
Governself service and safe actions
Pillar 01Build the trusted foundation

Onboard once. Review what Bicycle drafts.

Bicycle drafts your retail data model, the GMV, freshness, and source-health KPIs, and dimensions from your stack, each one validated against your data. You review, correct, and confirm.
Connect your sources Warehouse, app and web events, payments, product analytics, connected once.01 / 03
From source data to a reusable business model
12345
Connect sources
Bring in your data from across your stack.
Warehouse
Event streams
Databases
Files (CSV, Parquet)
Auto-discover & suggest
Bicycle profiles the data and suggests the business map.
Events
Dimensions
KPIs
Cohorts
Enrich with context
Add business context to improve accuracy.
Screenshots
Documents
Text / Notes
URLs
Build your model
Import existing semantic models, or build from scratch.
Import from dbt
Import from Tableau
OR
Build from scratch
Review & activate
Review mappings, validate, resolve gaps, activate.
Review mappings
Validate KPIs & data
Resolve gaps
Activate
AI-assisted coverage: We map 70–80% of your business data automatically from structure, naming, and patterns.You review, refine, and approve.
Bicycle drafts the KPIs with definitions, you review each one
Review and certify the governed agent before it publishes
Pillar 02Tune the intelligence layer

Tune what it watches and what it checks.

Two engines decide every answer: patterns set what to detect, drivers set what gets checked when GMV moves, real movement against the artifact. You tune both.
Two engines, both yours Patterns decide what to detect; drivers decide what gets checked when it moves.01 / 03
CONNECTED TO YOUR CONTEXT, NOT JUST THE KPILogsTracesSession & UXCode & configTicketsDocsExternal contextCause engineAUTO-WIRED DRIVERSCause connectorsWhat Bicycle producesInventory & availabilityPricing & promotionsCheckout & paymentsSearch & discoverySupplier / issuer / partnerLatency / errors / releasesPricing & inventory dataLogsSession replayObservabilityTickets & incidentsExternal signalsDeployments & flagsRunbooks / docsPrimary causeContributing causesWhat we checkedEvidence & source linksRecommended next stepBicycle suggests the driver map automatically. Teams connect the right cause sources, review the evidence, and refine the logic over time.
Set the patterns Bicycle watches
Add the drivers that explain a move
Pillar 03Govern self service and safe actions

Decide who sees what, and what the agent can do.

Dashboard consumers get answers without raw access. You set the permissions, subscriptions, and actions Bicycle may take. Self-service runs inside the boundary you draw.
Set role-based access Which agents each team sees, down to the KPIs, dimensions, and segments at the ontology level.01 / 03
SELF-SERVICE RUNS INSIDE THE BOUNDARY YOU DRAWYOU SET THE BOUNDARYSELF-SERVICE, SCOPED BY ROLEBusinessSubscribe · Request · AskRevenue KPIs in their region & channelAnalystRun · Subscribe · RequestAll KPIs · drivers · segmentsModelerRun · Approve · Publish · EditModel · definitions · governanceTHE BUSINESS GETS ANSWERS, NOT RAW ACCESSAlertsDashboardsData storiesChatRaw dataPASS THE GUARDRAIL GATESlack alertAUTOJira ticketAUTOPause Google AdsAPPROVALRemove supplierAPPROVALpreviewed · approved · bounded · reversible · logged · every answer and action carries who, what, and why.
Route alerts and data stories to the teams that own each KPI
The action catalog, linked to patterns and causes, scoped and logged
A day in the lifeFirst pass by Bicycle · the call is yours
Retail investigation queue

The morning review starts with the first pass done.

GMV dropped on a few stores. Bicycle has already checked the segments, ranked the likely causes, attached the evidence, and kept the definition governed. Your team reviews the answer, publishes the story, and tunes what should happen next time.

GMV · 3 stores · regionalTop segment below baseline. Affected segment and evidence already assembled for analyst review.
8:42 AM-1.8ppreview queue
Retail investigation queue8:42 AM↓ -1.8ppreview queue
Queue · alerts needing review
HighGMV dropped3 stores · regional · US2.3%-1.8pp vs baselineTop segment3 stores · regional · US$58Kat riskAnalytics leadowning team8:42 AM
MedAdd to cart rate downAll devices · Paid traffic5.6%-0.7pp vs baselineTop segmentPaid traffic · US$31Kat riskGrowth Marketingowning team8:15 AM
MedPayment success rate downAll devices · Credit cards92.1%-1.2pp vs baselineTop segmentCredit cards · US$44Kat riskPaymentsowning team7:02 AM
LowInventory in-stock droppedShoes · 3 warehouses94.3%-0.3pp vs baselineTop segmentShoes · West region$12Kat riskInventoryowning team6:55 AM
+ 4 alerts · 1 high · sorted by revenue at risk
SnoozeOpen investigation
GMV · root cause analysisfirst pass by Bicycle↓ -1.8pp
Triaged to one segment · 3 stores · regional · US
2.3% conversion
baseline 4.1% · ▼1.8pp · returning users · high volume
1 of 6 segments affected · 5 within expected range
Likely driversranked · confidence
Regional promo ended → real demand moveprimary81%
Late dbt run dropped 3 storessecondary38%
Organic demand dipruled out
Recommended actionsscoped
Create a ticket on Jirascoped
Alert #retail-revenue on Slackscoped
Flag metric as stalepreview
scoped · previewed · reversible · logged
+ Detected 8:42 · first pass 8:43 · top segment 3 stores · regional
Adjust modelApprove & publish
Validate evidence · 3 stores · regional · US8:42 AM↓ -1.8ppreview more
Impact overview
2.3%
conversion rate · -1.8pp vs baseline 4.1%
1.24Maffected users
5.83Msessions
Top affected segments
1
3 stores · regional · US
-1.8pp
2
3 stores · regional · West
-1.2pp
3
3 stores · regional · freshness
-0.9pp
Evidence attached
Funnel trendPDP to purchase
Channel mixpaid vs organic
Source freshnessfresh vs late run
dbt runlate @ 2:15 AM
Ruled out · not drivers
Price changesno overlap with window
Promo changeseligibility unchanged
Organic demand dipruled out
New vs returningboth affected equally
Definition governed · Source read-only · sources fresh
+ Reviewed by analyst on duty · 8:51 AM
Send backConfirm evidence
Publish answer · GMV8:42 AM↓ -1.8pppublish
Your conclusion84% confidence
GMV dropped on 3 regional stores in the US, primarily because a regional promo ended, a real demand move. A late dbt run that dropped 3 stores is a secondary, pipeline contributor.
Recommended next steps
1Flag the metric as stale, scoped to GMV on the 3 stores.
2Monitor GMV and source freshness.
3Backfill the late run and set a freshness guardrail.
Attach & sharegoverned
Impact summary
Evidence bundle
Next steps
Share with
Ecom leadershipEcommerce OpsGrowth Marketing+2
+ Published to Slack · #retail-revenue · 8:54 AM
Save draftApprove & publish
Reused next time · operating memoryToday 7:28 AM↓ -1.6ppauto-applied
Similar situation detected: GMV dropped · 3 stores · regional · US
Bicycle applied your approved playbook.
Detected pattern
Freshness impact
Applied playbook
GMV freshness
Likely driver
Late dbt run
Recommended action
Flag as stale
Caught 74 min earlier · 7:28 AM vs 8:42 AM on the first run
No re-investigation · pattern auto-applied, analyst review optional
+ Memory: retail GMV-freshness playbook · run 2
ReviewView playbook
More roles on Bicycle · same loopYour team's KPI · your screens

Different teams own different metrics. Bicycle keeps the investigation governed.

Move from this analytics engineering view to the retail workflows around search conversion, category margin, app funnels, and revenue performance.

Vibe Analytics

Build your first governed investigation.

Start with one recurring metric in your queue: why checkout conversion, payment success, or activation moved. Bicycle watches the governed KPI, checks likely causes across the warehouse, payments, app & web, and events, and hands your team a reviewable first pass.

What to bring
A recurring investigation: checkout conversion, payment success, search conversion, or activation rate.
An approved read-only path to the relevant sources: the warehouse, event streams, app & web, payments, or ad networks.
The KPI definition, default segments, and reviewers who decide what becomes trusted.
What we do
Configure the investigation path around your KPI, segments, and evidence rules.
Monitor device, OS, channel, region, and segment movement.
Prepare a first-pass cause analysis with evidence, ruled-out paths, and governed definitions.

No credit card required · analyst-reviewed · governed from day one