Welcome to Petavue
You’ve spent weeks building a report no one trusts. Petavue exists so that never happens again—here’s how the platform turns questions into verified, deployable answers in hours instead of weeks.
The world before Petavue
Think about the last time someone challenged a number in a meeting. How long did it take to trace back the logic? Could you prove it was right? Could anyone else reproduce it without you in the room?
That gap between having data and trusting data enough to act is what Petavue closes. By making every metric explainable, every calculation verifiable, and every answer deployable in hours instead of weeks.
What changes with Petavue
Petavue is an agentic analytics platform.
That means the system doesn’t just store your data or display charts — it actively does the analytical work for you.
Inside Petavue, agents helps you build metrics, verify logic, spot data quality issues, and deploy insights to stakeholders.
Here’s what that looks like in practice:
A real scenario: “What’s the impact of LinkedIn Ads on pipeline?”
| The old way | With Petavue |
|---|---|
| File a ticket with the data team. Wait 1–2 weeks. | Describe the question to the agent in plain language. |
| Get a spreadsheet back. Manually verify the logic. | The agent proposes a plan: data sources, filters, attribution model. You review and approve in minutes. |
| Build a slide deck. Hope the definitions match what leadership expects. | Analysis executes. Results include full reasoning; every number traces back to approved definitions. |
| Present. Get challenged. Go back and re-derive. | Publish to a dashboard. Stakeholders self-serve via Sage. Same answer every time, no meetings required. |
| Total time: 2–4 weeks | Total time: 2 hours |
That speed — from question to deployed, verified insights is what Petavue delivers, by having the agent do the heavy lifting while you focus on review and judgment.
Three things that change simultaneously for you
Productivity compresses. Analysis that took weeks (gathering data, writing queries, building charts, getting review) now takes hours. The agent handles data assembly, calculation logic, and output formatting. You review and approve.
Trust becomes structural. Every output links back to approved definitions, reviewable plans, and traceable execution. When someone asks “where did this number come from?” the answer already exists as a part of the dashboard or analysis.
Depth is built in. This isn’t a generic BI tool that happens to have AI. Petavue understands marketing attribution, pipeline metrics, funnel analysis, and RevOps workflows natively. It doesn’t guess at what “MQLs” or “campaign influence” mean, it is rich with all the GTM context you built through key definitions.
How the platform fits together
Petavue has five areas that form a deliberate sequence. Each one feeds the next. Understanding this sequence now will save you from guessing where things live later.
Data Hub: connect, describe, and define your data
The Data Hub is where you give Petavue the raw materials and the vocabulary to work with. It has three parts:
Connections link your data sources. CRM (Salesforce, HubSpot), marketing automation (Marketo), ad platforms (via Fivetran), analytics (GA4, Mixpanel), or direct warehouse connections (Snowflake, BigQuery, Redshift, Databricks). The agent can help you verify connection health and flag sync issues before you start building.
Dictionary is the schema map. It shows every table and column from your connected sources. You can edit descriptions, add friendly labels, and enable or disable fields to control what the agent uses in analysis. Think of it as teaching the system your organization’s language—not just the column names, but what they mean to your team.
Want to quickly assess your data before configuring anything? Try this prompt:
"Review the data sources connected to this account. Flag any tables with low fill rates, identify fields that might have data quality issues, and summarize what’s available for marketing attribution analysis."
The agent will scan your connected data and surface gaps in seconds so you know what you’re working with before you build on it.
Definitions is where you standardize business logic. This is where you create Key Definitions — reusable objects that lock down what a metric means for your organization. “Marketing Touched Deal” as a filter-based Term. “Pipeline Velocity” as a computed Formula. “Weekly Qualified Pipeline” as a trackable Metric.
Once created, the agent respects these definitions in every analysis. No drift, no “my number vs. your number” debates. Someone asks “what’s our MQL count?” six months from now? Same calculation, same guardrails, same answer.
Your VP Sales asks why conversion rates dropped 15% this quarter. With Key Definitions, you don't scramble through three Excel files to reconstruct your logic—you click one link and show exactly how conversion is calculated, what data feeds it, and when it last refreshed. Same question from the CFO next week? Same answer. Every time.
That's not just consistency, it's productivity. You defined it once, and the system enforces it everywhere.
Projects: bounded containers for analytical work
A Project in Petavue represents a bounded question your organization wants to answer, with an explicit purpose, scope, and set of definitions.
Unlike a folder that just holds files, a Project holds an agreement: this initiative uses this data, defines these metrics this way, and produces outputs bound by these constraints. Everything downstream (Workbooks, Dashboards, Memos and more) inherits from the Project.
Examples: “Q4 Pipeline Velocity Deep Dive,” “Marketing Attribution – 2025 Campaign Analysis,” “Funnel Conversion by Segment.” Each has an owner, a clear purpose, and the definitions that govern how metrics are calculated within it.
Workbooks: where analysis happens
This is where the biggest productivity compression occurs. You describe what you want to know in plain language, the agent proposes a plan (which tables to query, what filters, what calculations), and you review that plan before anything executes.
This plan-first approach is what separates Petavue from tools that just run queries.
You see the logic before the results. You can modify individual steps — change a filter, adjust a date range, swap a field. When execution runs, you can inspect every intermediate step.
Workbooks produce tables, charts, summaries, and memos — each designed to be consumed immediately. Any output can be promoted to a Dashboard or connected to an external workflow via API.
Try this in your first Workbook: "Build a pipeline velocity analysis by segment for Q4. Use our agreed Pipeline Velocity definition and break out by lead source."
The agent will propose a complete plan. Review it, tweak what needs adjusting, and approve. What used to be a week of SQL and spreadsheet work becomes a focused review session.
Dashboards: the consumption layer
Dashboards are how your broader team sees the work. You assemble widgets from Workbooks into a shared view, set a refresh cadence, and share with stakeholders.
The key difference from dashboards in other tools: every widget traces back to an approved plan, specific definitions, and a reviewable execution history. When someone asks “why does this chart show X?” the answer already exists right there, it doesn’t require the person who built it to be in the room.
Every widget also exposes a secure API link, so you can push output to Slack, email, Google Sheets, or automation tools like n8n and Zapier. Insights arrive where your team already works.
Sage: self-serve intelligence for stakeholders
Sage is Petavue’s built-in agent that sits on top of dashboards. It helps business users understand what they’re looking at without interrupting the ops team.
But Sage isn’t just a “ask a question” box. It provides guided deep-dive, walking stakeholders through the dashboard, and generating answers grounded in approved definitions. Stakeholders don’t need to file tickets, schedule meetings, or wait for analysts.
Sage can answer questions like: “How was this metric calculated?” “What data sources were used?” “What changed since last week?” It pulls answers from the metadata, definitions, and execution history that the Workbook and Dashboard already contain.
Sage is the payoff of doing the setup work properly. Well-configured Data Hub + clear Definitions = stakeholders who can self-serve without pinging you on Slack.
Two kinds of users, one shared truth
Petavue serves two distinct personas who need different things from the same system. Understanding which one you are determines where you spend your time.
Ops Owners: the builders
Who this is: RevOps, Marketing Ops, Sales Ops, Analytics leads embedded in GTM. You’re the person on the hook when numbers don’t line up.
What changes for you: Instead of manually writing queries, building spreadsheets, and reconstructing logic every time someone asks a question—you describe what you need, review the agent’s plan, and approve. Definitions you set once propagate everywhere. Dashboards refresh automatically. Stakeholders self-serve through Sage instead of pinging you.
The result: Fewer reconciliation cycles. Defensible answers that already contain their reasoning. Less time explaining, more time improving. Analysis that used to take your entire week gets done in a focused afternoon.
Business Consumers: the decision-makers
Who this is: Sales leadership, Marketing leadership, Revenue leadership, executive stakeholders. You don’t want to build analytics; you want to trust them enough to make decisions.
What changes for you: No more waiting for the ops team to answer your questions. No more wondering if the dashboard number matches what finance has. Sage gives you guided, self-serve access to verified answers with full explanations of how every number was calculated.
The result: Faster decision cycles. Blind spots surface before they become problems. Insights arrive where you already work—in Slack, email, or scheduled digests—in formats designed to be read in under a minute.
What to do next
You now have the mental model for how Petavue works and why it’s structured this way. The next step depends on who you are:
- If you’re an Ops Owner: proceed to Article 2.2: Connecting Your Data Sources. That walks you through getting your first integrations live and using the agent to verify data quality.
- If you’re a Business Consumer: your ops team handles setup. Once dashboards are shared with you, Article 6.5: Sage for Business Consumers is your starting point.
If you’re evaluating Petavue: Section 9 shows end-to-end walkthroughs for attribution, funnel analysis, pipeline reporting, and more. Those give you the most concrete picture of what the platform delivers—and how fast it gets there.