Log in
Book a demo
Try for free

The Platform

Everything between your data and
the people who need to see it.

Five capabilities that work together as one platform. Connect, prepare, automate, control, share. Every plan includes every capability — including Free. Tiers differ on scale, not access.

01 · Connect 02 · Prepare 03 · Automate 04 · Control 05 · Share
Pipeline (7) Describe your transformation… Export Publish 1 · Source FashionRetailer_Jan_And_Feb 2 · Conditional Filter where billing_country in (UK, US, FR) 3 · Math Function (discounts) / gross_sales 4 · Bulk Replace AI Spelling Match on product_title 5 · Extract Text from product_title — left of '-' T product_title (72) T billing_country (26) T product_type (9) referrer_source product_type product_title colour order_id Direct Shirts Winter Work Shirt Black 772799627 Social Sweats-Hoods Le Monde Hoodie Black 783775694 Direct Sweats-Hoods Herbs Hoodie Black 784002383 Social Outerwear Cord Trim Pullover Black 755043237 Direct Outerwear Cord Trim Pullover Olive 749136019

01 · Connect

Bring your data in.

Mammoth connects to where your data already is — databases, SaaS apps, files, APIs. No ETL job to write. No engineer to wait for. Add a source, query it in minutes. Set a schedule, it stays fresh.

Your data lives in twelve places, in twelve formats, owned by twelve people. Pulling it together is the job before the job.

  • Native connectorsPostgreSQL, MySQL, MSSQL, BigQuery, Snowflake, Redshift, Shopify, HubSpot, Salesforce, GA4, Stripe, Google Sheets, Dropbox, OneDrive.
  • The long tail: REST + AI builderPoint at any REST endpoint. Describe what you want, AI builds the connector.
  • Live connectionsSchedule refresh. Trigger on webhook. Configure once, then forget.
Postgres prod-db · live Shopify orders · hourly HubSpot contacts · daily BigQuery warehouse · live Salesforce leads · daily REST + AI any endpoint

02 · Prepare

Clean it. Reshape it. Trust it.

The visual pipeline. Every transformation is a step. Every step is named, reorderable, and inspectable. You can run SQL when you want SQL, describe the change in plain English when you don't.

It's never quite ready. Headers in row 4, dates as strings, the same product name spelled three ways. Five hours of spreadsheet surgery before anyone can use it.

  • 80+ transformationsFilter, label, replace, pivot, unpivot, lookups, joins, math, conditional logic, date manipulation, type conversion, deduplication.
  • SQL when you need itDrop into SQL on any view without leaving the platform. Reusable snippets, parameterized queries, full DuckDB SQL surface.
  • Generative AIDescribe the transformation in English. AI generates the task. You see exactly what changes before you apply it.
  • Visible pipelineEvery step has a name and a row. Reorder, edit, or delete. No black boxes. New team member opens it and immediately understands what's happening.
  • Multiple views per datasetOne source, many slices. Sales weekly, marketing daily, finance monthly — all from the same source of truth.
  • Billion-row scaleDuckDB in-process per worker. Tested in production at Starbucks scale (1B+ rows monthly).
01 SOURCE FashionRetailer_Jan_Feb.csv 1,240,000 rows 02 FILTER billing_country in (UK, US, FR) ↓ 843,210 rows 03 MATH (discounts) / gross_sales → margin AI 04 BULK REPLACE AI spelling match on product_title AI 05 EXTRACT TEXT product_title → left of ' - ' ready

03 · Automate

Set it once. Walk away.

Schedule the pipeline. Trigger it on events. Notify the right people when it finishes or fails. The work happens whether anyone is at their desk or not.

Monday morning: open laptop, run the same script you ran last Monday, copy the same numbers into the same deck. Forty weeks a year.

  • 5 orchestration typesRefresh (pull fresh data), Consolidation (combine sources), File Collection (sweep a folder), Messaging (Slack/email triggers), PDF Extraction.
  • Scheduling & dependenciesCron-style with dependency chains. Pipeline B runs only after Pipeline A succeeds. Hourly, daily, weekly, or "every other Wednesday at 6:15."
  • Webhook triggersA file lands in S3, a Salesforce field changes, a Stripe payment completes — Mammoth catches it.
  • NotificationsEmail and Slack on completion, failure, or data-check threshold breaches — with affected row count and a link to investigate.
  • Auto-Sync downstreamDashboards, BI database, exports, APIs — all refresh on the same heartbeat.
  • Manual run when neededOne-click manual run, same pipeline, full audit trail.
12 9 6 3 DAILY · 06:00 UTC SOURCE Refresh PREPARE Pipeline SLACK Pipeline complete · 843k rows

04 · Control

Approve before you ship. Audit after.

Checkpoints catch bad data before it goes anywhere. Draft Mode lets you preview a change before it touches downstream. Activity logs answer "what changed?" months later. Single-tenant isolation, even on Free.

The pipeline ran. The numbers look weird. Nobody knows when, who, why — and the dashboard already went to the executive team.

  • Data CheckpointsRow count, null %, unique values, value range, custom expressions. Alert or Approval Gate modes.
  • Draft ModePreview changes before they touch downstream dashboards or exports.
  • Activity logEvery action, every user, every timestamp. 7 days on Free, 90 on Starter, 1 year on Pro, up to 7 years on Enterprise.
  • Role-based accessWorkspace, project, and view-level permissions.
  • Single-tenant isolationNo shared storage, no shared compute. On every plan, including Free.
  • ComplianceSOC 2 Type II, ISO 27001, GDPR, HIPAA. Audit reports on request for Enterprise.

Coming soon: Row-Level Security on dashboards

One dashboard, each viewer sees only their authorized rows. Built for agencies serving multiple clients and enterprises with regional data isolation. See the spec →

DRAFT MODE · ON ACTIVITY LOG sarah@co.com published dashboard 09:14 pipeline ran · 843,210 rows passed 06:01 checkpoint approved · null% < 2% 06:00

05 · Share

Send the answer,
not the spreadsheet.

Three ways out of Mammoth, depending on who needs the data and how they want it. The flagship is AI-generated dashboards — the other two close the gap for everything that doesn't need to be a dashboard.

500 stakeholders need to see this. Most BI tools charge for each of them. Most of them don't need a tool — they need an answer.

  • AI DashboardsDescribe what you need in plain English. Mammoth generates a production-ready dashboard in 30 seconds.
  • Free unlimited viewersEvery plan. Publish to thousands of stakeholders. No per-seat math.
  • Live database for your BI toolFeed your existing Tableau or Looker — Mammoth provisions a managed Postgres or BigQuery instance in 5 minutes.
  • Files, APIs & webhooksCSV, Excel, PDF to SFTP/S3/Drive. REST API on every dataset. Python SDK for white-label deployments.
  • Auto-syncs with the pipelinePipeline runs at 6am, dashboard reflects it at 6:01am.
  • Billion-row scaleFive-minute query caching keeps dashboards fast on enterprise-scale datasets.
See AI Dashboards →

Share · BI Integration

Live database for your
BI tool.

Already running Tableau, Power BI, or Looker? Don't rip them out — feed them. Mammoth provisions a managed PostgreSQL or BigQuery instance in 5 minutes. Your existing dashboards stay; Mammoth keeps the data fresh on the schedule you set.

Infrastructure
Managed Postgres
or BigQuery
Provisioned in your AWS or GCP region. Read-only credentials issued to your BI tool. Mammoth handles the schema, writes, and refresh.
Flexibility
Same data,
two surfaces
Your analysts build in Tableau as before. Your business users build dashboards in Mammoth. Both pointed at the same pipeline.

Share · Export

Files, APIs & webhooks.

Not everything needs to be a dashboard. Sometimes you need a CSV in a folder, a Slack message at 6am, or a JSON payload to your internal system.

Export
File exports
CSV, Excel, PDF. Scheduled or on-demand. Delivered to email, SFTP, S3, Google Drive, OneDrive, or Dropbox.
API
REST API
Every dataset is queryable via REST. Authenticated, paginated, JSON output. Build your own app on top.
Automation
Webhooks & SDK
Push results to a webhook on pipeline completion. Embed via the Python SDK for white-label deployments.

Pricing

Same platform on every plan.

Full feature set everywhere — including Free. Move tiers for more storage, more users, more AI runs. Never to unlock a capability.

FREE

$0

3 published dashboards · 1 GB storage · Single-tenant · 21-day Pro trial

STARTER

$199/mo

10 dashboards · 10 GB storage · 25 AI runs/mo · For small teams

PRO

$559/mo

50 dashboards · 50 GB · 100 AI runs · Most teams scale here

ENTERPRISE

From $5k/mo

Unlimited everything · SSO · White-label · 7-year audit · Custom region

Viewers are always free on every plan. See full pricing →

Financial Services

Reconciliation, regulatory reporting, audit trails.

Retail & CPG

Sales, inventory, and supplier data unified.

Services & Agencies

Separate projects per client.

Case Study

Store-level teams self-serve — no IT tickets.

vs Alteryx / Excel / ChatGPT

Separate projects per client.

About AI New

AI for the question. Mammoth for the repeating answer.

Blog

Documentation ↗

Free Tools New

About

Book a Demo
Log in
Try for free