Data↗Actions.

Accurate
analytics
agents.

Trusted at scale.

We build the Business Context Intelligence that makes Claude, ChatGPT or your own agents accurate on your business.

Claude · ChatGPT · Your agentCFO · 9:06Agent · 9:08
Why is net revenue retention down this quarter?
Analysis plan · why NRR is down
Finance and cashNRR by quarter, against plan
Revenue leversChurn, contraction and expansion by segment
ProcessesEnterprise renewals and customer health
Analysis plan · why NRR is downAnalysis planAgreed by the CFO ✓
Computing the answer on your business …NRR is down, 104% → 101%, on Enterprise churn: −$0.9M, up from −$0.5M, after Enterprise customer health fell from 79 to 71 a quarter earlier.
Making the analysis plan …Agree
Business Context Intelligence
Finance and cash
ARR
NRRNRR 104% → 101%
Revenue levers
Expansion
Contraction
ChurnChurn −$0.9M
Processes
Usage
Renewals
Customer healthCustomer health 79 → 71
Your data · Billing · CRM · ERP · Product · in your cloud

The shift

AI agents take the analytics
cycle from weeks to minutes.

Before · with your analyst · one cycle at a timeOne cycle at a time

CFO
CRO
VP CS
VP RevOps
Analyst

CFO · Why is NRR down?

Ask → Requirement → Analysis → Review → Insight board

~3 weeks

Now · with AI agents · every leader, all at onceEvery leader, all at once

CFOAgent
CROAgent
VP CSAgent
VP RevOpsAgent

CFO · Why is NRR down?

Ask → Reasoned → Agreed → Insight board

~3 minutes

The problem

Accurate analytics agents need your business context,
engineered by a data team.

An analytics agent answers your business questions accurately only on business context: how your business runs, is measured and moves. Engineering business context takes a data team, and the work grows with the business.

Three analytics agents, built in-house: accuracy without → with business context

Anthropic

Claude, for self-service analytics

Business context

Data team writes the skills

Analytics accuracy

21% → 95%

Without → with the skills

LinkedIn

SQL Bot, for internal teams

Business context

Data scientists build a knowledge graph

Analytics accuracy

9% → 48%

Table schemas → knowledge graph

DoorDash

Vera, for 10,000 employees

Business context

Data team verifies the data and tests

Analytics accuracy

48% → 73%

Off-the-shelf agent → Vera

Sources ↗ Anthropic · LinkedIn · DoorDash

Agents

Every analysis reasoned on your business graph,
agreed by the leader who runs it.

Accurate analytics agents for every team, every analysis.

Two example analytics use cases for the CFO

What analysis · Review board

“Build my quarterly business review against plan.”

The CFO, in Claude

Claude · ChatGPT · Your agentCFOCROVP Customer successVP CSVP RevOps
New threadQuarterly business reviewWhat analysis
CFO · 9:02
Build my quarterly business review.
Agent · 9:02
Analysis plan · on your business graph
FinanceQ3 FY26All segments
Finance and cash Revenue · ARR, net new ARRstatus, trend
Revenue levers Levers · new, expansion, contraction, churncontribution
Processes Retention · net revenue retentiontrend
Finance and cash Efficiencystatus
AgreeRevisedAgreed by the CFO ✓
CFO · 9:03
Show it against plan.
Agent · 9:03
Analysis plan · on your business graph
FinanceQ3 FY26All segments
Finance and cash Revenue · ARR, net new ARR vs planvariance
Revenue levers Levers · new, expansion, contraction, churn vs plancontribution
Processes Retention · net revenue retention vs planvariance
Finance and cash Efficiency vs planvariance
AgreeRevisedAgreed by the CFO ✓
Agent · 9:05
Your review board is ready.
Review board · quarterly business reviewFinance · Q3 FY26 · LiveOpen ↗
Message your agentBuild my quarterly business r
Review board · quarterly business reviewReview boardFinanceQ3 FY26All segmentsLiveWaiting for approval
No board yet
Revenue
ARR vs plan
Net new ARR vs plan
Levers
Net new ARR by lever, by quarter vs plan
Retention
Net revenue retention by quarter vs plan
Revenue
ARR $48.0M, $0.3M below plan
ARR
$48.0M
plan $48.3M
Net new ARR
$2.4M
plan $2.7M
Levers
Net new ARR $2.4M against $2.7M plan; churn −$0.9M, up from −$0.5M
Net new ARR by lever, $M
Retention
Net revenue retention 108.6%, −0.6 pts, outside its normal range
Net revenue retention by quarterplan 109.5%
Efficiency
Gross margin 78.2%, on plan; CAC payback 19 months, one month over plan
Gross margin78.2%plan 78%
CAC payback19 monthsplan 18
Burn multiple1.1×plan 1.2×
ARRverified
DefinitionYour metric tree
SourceBilling · ARR snapshots
OwnerFinance
QueryView SQL ↗
RefreshedToday, 06:00

Flagged on the review board

Net revenue retention 108.6%, down 0.6 points, outside its normal range.

See why ↓

Why analysis · Diagnosis board

“Why is net revenue retention down?”

The CFO, from the review board

Claude · ChatGPT · Your agentCFOCROVP Customer successVP CSVP RevOps
Why is net revenue retention down?Why analysis
CFO · 9:06
Review board · net revenue retention −0.6 ptsWhy is net revenue retention down 0.6 pts this quarter?
Agent · 9:06
Analysis plan · why NRR is down, on your business graph
FinanceQ3 FY26All segments
Finance and cashNet revenue retention1 variance
by segment
Revenue leversEnterprise2 decomposition
by lever
Revenue leversChurn3 contribution
influence
ProcessesCustomer health4 signal
Solid: components · dashed: influenceAgreeRevisedAgreed by the CFO ✓
Agent · 9:08
Your diagnosis board is ready.
Diagnosis board · net revenue retentionLinked from the review board · LiveOpen ↗
Message your agent
Diagnosis board · net revenue retentionDiagnosis boardFinanceQ3 FY26All segmentsLiveWaiting for approval
No board yet
1 · variance
Net revenue retention
by segment
2 · decomposition
Net revenue retention › by segment
by lever
3 · contribution
Enterprise › by lever
influence
4 · signal
Enterprise churn customer health
1 · varianceNet revenue retention
108.6%, −0.6 pts, outside its normal range
by segment
2 · decompositionNet revenue retention › by segment
Enterprise −1.1 pts; Mid-market and SMB held
by lever
3 · contributionEnterprise › by lever
Churn takes 1.1 pts; expansion and contraction net to zero
Churnverified
DefinitionYour metric tree
SourceBilling · churned ARR
OwnerCustomer success
QueryView SQL ↗
RefreshedToday, 06:00
influence
4 · signalEnterprise churn customer health
Customer health fell 79 → 71 a quarter before churn rose
Root cause diagnosis
Net revenue retention fell 0.6 points, mostly Enterprise churn, after customer health dropped a quarter earlier.

The CFO decides

The CFO sits down with customer success to review every Enterprise account with falling health, before renewal.

Platform

Business Context Intelligence (BCI)

BCI reasons on your business graph, presents the analysis plan, and computes every insight on your warehouse.
So your analytics agents stay accurate, for every team, for every ask.

Your agent
Claude, ChatGPT or your own agent connects to Business Context Intelligence through MCP.
So every leader asks in the same agent as today.
Business Context Intelligence
Context models how your business runs, measures and reasons. Intelligence reasons and computes on the context.
In your cloud
Business Context Intelligence runs in your cloud and computes on your warehouse.
Your business context and intelligence stay in your own governance.

Reasoning

Intent agentInterprets the ask prompt; calls the reasoning agent
Reasoning agentDecomposes the ask; makes the analysis plan
Graph agentPlaces the ask on the business graph
Metric agentBinds each term to a metric, or asks back
Insight agentPicks the insights each step needs
Narrative agentRuns the agreed plan; delivers the narrative

Insights

Insight servicesQuantity, trend, comparison, contribution, and more from the plan
Query engineTurns metric queries into SQL

Reasoning skills

SkillsTrend, variance, comparison, contribution, decomposition, signal

Metric trees

Every metricDefinition, source, owner
TreeThe metrics that make up each metric, and the metrics that influence it
GrainsEvery dimension and time grain a metric is cut by

Business context graph

AltitudesAs many as your business has, from finance down to customer journeys
LeadersEach leader placed at the altitude the leader runs

Accuracy from your Business Context Intelligence, not the model.

Case study

Business Context Intelligence is live
with DataOrbit’s analytics agent.

Data Actions engineered DataOrbit’s business context as Business Context Intelligence. Apollo, DataOrbit’s analytics agent, answers on that business context, live.

DataOrbitLive

Apollo, an analytics agent for DataOrbit’s customers

Business context

Engineered by Data Actions as Business Context Intelligence, with DataOrbit’s team

Result · analytics use cases, live

ARRWhat analysis · Review board
RetentionWhat analysis · Review board
ChurnWhy analysis · Diagnosis board
“Data Actions’ Business Context Intelligence gives Apollo every customer’s business context. Our customers use Apollo for real-time decisions, without waiting on an analyst.”

Founder, DataOrbit

Let’s make your Claude, ChatGPT or own agent accurate and trusted for analytics.

Start with one analytics use case, go live in weeks.

Your role and use case

Finance and cash
ARRNRR
Revenue levers
New businessExpansionContractionChurn
Processes
PipelineUsageRenewalsCustomer health

CFO · Why NRR is down, on Finance and cash

Your path to live

1

Data Actions engineers your business context as Business Context Intelligence, with your team

2

Your agent answers live; your team keeps the business context current