Connect your data, capture more of it, and let AI build dashboards, watch for changes, and explain what happened — all in plain English.
Free forever plan · No credit card · Set up in under 10 minutes
Built for data-driven teams across
No glue code, no separate tools to wire together. Capture, model, and act on your data inside one platform.
Plug in your warehouses, SaaS apps, files, or APIs. Or capture data directly with branded public forms that write back to your datasets.
Define a star schema once. Certify the metrics that matter. Every dashboard, automation, and AI answer uses the same definitions.
Ask questions in English, generate dashboards from a prompt, and let automations watch metrics so you can spend time deciding — not refreshing tabs.
Drop in a Postgres, BigQuery, Stripe, or webhook. Rows arrive in your DuckDB workspace in seconds.
Describe what you need. The agent picks metrics, lays out widgets, and validates against your schema before saving.
Wire up triggers, branch on conditions, and ship to Slack, email, or any webhook — with a live execution graph.
Most teams stitch together a forms tool, a BI tool, a workflow engine, and a metrics layer. LumenQube does all four — so capture, dashboards, automations, and definitions stay in lockstep.
Public, branded forms that drop submissions straight into your datasets — no glue code, no third-party silo.
Drag-and-drop canvas with 22+ widget types — or describe what you want and the agent assembles it end-to-end.
Multi-condition triggers, branching actions, live execution graph, replay, dead-letter, and a circuit breaker.
Star-schema modeling, certified metrics, and row + column security applied everywhere — including AI answers.
Powered by Claude with prompt caching and tier-aware routing — so analysis stays fast and predictable on cost. AI uses your governed metrics, not raw tables, so the answers are the same ones your dashboards show.
Describe a goal — "show me sales health for the EMEA pod" — and the agent picks the right metrics, lays out the canvas, and validates every widget against your schema before saving.
When a chart fails to render, the AI inspects the validator output, repairs the spec, and retries. You see the fix log, not a broken tile.
Ask in English. Answers cite the metric they used, the rows they filtered, and the SQL they ran. No black box.
Baselines learn the seasonality of every governed metric and surface what changed — and what likely caused it.
Public, branded forms that write straight into your datasets. Generate them with AI, share via link or embed, and watch submissions flow into the same dashboards and metrics you already trust.
Describe a form in English — or generate one from an existing dataset. Refine with follow-up prompts.
Single-page, wizard, or two-pane layouts. Public links, embeds, and per-form rate limits.
Submissions append (or update) rows in your datasets atomically — schema-validated on the way in.
Text, currency, dates, files, signature, matrix, address, dataset lookups, computed formulas, and more.
HMAC-signed webhooks fire on submission. Notify owners, custom recipients, or send branded confirmations.
Modern, glass, neon, elegant — or tune fonts, radius, density yourself. Full brand control.
Define triggers from any metric, branch on conditions, and ship actions to Slack, email, webhooks, or your own systems. Built for the people who get paged when things go sideways.
AND/OR/NOT logic across metrics, widgets, and dashboards. Branch into different action paths based on what fired.
AI generators are validated against your live schema with a repair loop — you never save an automation that won't run.
Watch every step run in real time. Replay any past run from history with the same visual.
See how a new automation would have fired against the last 30 days, with projected token and compute cost.
Failed runs land in a dead-letter queue. Repeated errors trip a breaker before they page anyone.
Right-click a chart, choose "Automate this widget," and a trigger is wired in seconds.
Pre-built recipes for churn alerts, revenue drops, SLA breaches, and unusual spikes — adapt and deploy.
A visual pipeline editor for transforms, with imports and refreshes that run as supervised background jobs — not silent cron failures.
Filter, cast, join, aggregate, derive — chained in a clear pipeline that anyone can read.
Cloud Tasks-backed jobs handle multi-million-row imports with progress + heartbeat.
Every refresh is recorded with row counts, duration, and diffs. Roll back if a source goes weird.
Hard quota stops prevent runaways from saturating shared infra — your imports never block another tenant's.
Define your business model in a star schema, certify the metrics that matter, and watch every dashboard, automation, and AI answer use the exact same definitions. No drift. No surprise SQL.
measure monthly_recurring_revenue {
source: stripe.subscriptions
formula: sum(plan_amount_cents) / 100
filter: status = 'active'
granularity: month
owner: finance-team
}
Visual model of facts and dimensions. Define joins once, reuse them everywhere.
A single source of truth for "revenue," "MRR," or anything else your team argues about in spreadsheets.
Before you change a model, see exactly which dashboards, automations, and widgets it would affect.
Promote ad-hoc measures into governed metrics with one click — versioned and audited.
Browse, profile, and trace every dataset, metric, and dashboard from a single explorer. Built so engineers, analysts, and stakeholders can answer 'where does this number come from?' in seconds.
Click any field, table, or dashboard and follow it forward and backward through every transform, metric, and consumer.
Every column shows null rate, uniqueness, distribution, and recent value samples — at a glance, no SQL required.
Bookmark a filtered slice and share it with your team. Searches stay current as the underlying data changes.
One palette finds datasets, metrics, dashboards, automations, and forms — with role-aware results.
Pre-built connectors for warehouses, SaaS apps, files, and APIs. Test-config endpoint, persistent diagnostics, and friendly error messages mean you'll know exactly why a connection failed — and how to fix it.
A DuckDB query engine, smart caching, and visibility-aware polling keep dashboards snappy without burning budget. Heavy work — imports, refreshes, data prep — runs as background jobs with per-tenant caps. Every AI call shows tokens, tier, and dollars.
Built on Google Cloud with defense-in-depth at every layer. Pursuing SOC 2 Type II and ISO 27001 with GDPR controls already in place.
Policy-driven filtering — users only see rows they're authorized to.
Mask, redact, or hide sensitive columns with per-role rules.
JIT provisioning, SCIM, role mapping. Enterprise-tier ready.
AI-powered scanning automatically classifies sensitive fields.
AES-256 for credentials, secrets, and stored configuration.
Every action logged with correlation IDs. Exportable for compliance.
Per-tenant DuckDB databases, GCS buckets, and IAM scopes.
Per-tenant caps prevent runaways from affecting other orgs.
No credit card. Plan limits and AI usage are surfaced inside the app — no surprise overage bills, ever.
Try the platform, build a side project.
For small teams getting serious about data.
The full platform for scaling orgs.
Dedicated infra, support, and compliance.
All plans include encryption, audit logs, and per-tenant isolation. Hard quota stops prevent surprise bills.
From boutique agencies to scaling SaaS, teams use LumenQube to capture data, build dashboards in minutes, and let automations watch every metric.
We replaced a Forms tool, a BI dashboard, and a workflow product with LumenQube — and got better than all three. The semantic layer is the unlock.
A junior analyst built a finance dashboard from a single prompt and it picked the certified MRR metric — not a stale spreadsheet number. That's the moment I knew this was different.
The live execution graph for automations is the most useful debugging tool we've adopted in a decade. Our oncall pages dropped 40% in the first month.
Set up your first dashboard in minutes. Let the AI handle the rest.