Skip to content
NewAgent Fleet — agents that don't just watch, they act

Your data. Your company. Your agents.

A fleet of agents that runs on the data you already have and the way you already work — and gets smarter the longer it lives inside your business.

Get started freeBook a 20-min demo
  • SOC 2 Type II
  • Your data never trains shared models
  • Deploy in your own VPC
⌘K
Your question

What drove the churn spike in March, and is it still happening?

Assemble·4 sources cited

March churn hit 4.2% (↑ from 2.1%), driven almost entirely by 9 Starter-tier accounts with fewer than 4 active seats. April is tracking back to 2.4% as the low-seat cohort stabilises.

Gross churn · weeklyalert above 3.0%
JanFebMarApr

Answers arrive where the question was asked — no new tool for anyone to learn.

  • Slack@mention or DM
  • TeamsChat or channel
  • iMessageText it a question
  • WhatsAppField teams
  • EmailForward a thread
  • WebThe workspace
How it works

Three moves to shared intelligence.

No pipelines, no migration project, no new tool to learn. Most teams ask their first real question on day one.

01

Connect

Point it at everything you run on — databases, spreadsheets, SaaS tools, PDFs. One-click OAuth or a read-only key. Your data stays in your region, under your rules.

Before
$4.2Mincludes services
ARR excludes services · saved as a rule
Every answer after
$3.8Mrecurring only
02

Teach

Correct an answer once and it becomes a rule — tagged to its author, auditable, applied everywhere the data appears. Your definitions stick.

What's Q3 net-new ARR?
$3.8M
Up 21% on Q2. Recurring only, per your rule.
Cited
03

Ask

Plain English in, cited answers out — in Slack, Teams, email, or the workspace. Every number links back to its source row.

The fourth move

It happens by itself.

Every answer becomes a rule. Every rule sharpens the next answer. By month three, Assemble is the longest-tenured analyst your company has ever had.

22%87%
answer accuracy, first 12 weeks · average across deployments
Week 1Week 12
The product

Ask. Publish. Automate.

Questions become cited answers. Answers become living documents. Documents become agents that act.

Chat

Ask like a human. Get answered like an analyst.

Drop a question in Slack, email, or the workspace. Assemble reads the systems it's connected to, applies your rules, and replies with a cited answer — usually in under a second.

  • Citations on every number, clickable back to the source row
  • Follow-ups inherit context; no re-priming the model
  • Share answers as live artifacts — they keep updating
#rev-ops42 membersToday
AT
Ana TorresHead of Revenue14:08

@Assemble where are we on Q3 coverage, and which reps are under-pipelined?

Assembleapp14:08

Team pipeline is $57.8M against a $21.4M Q3 plan — 2.7× coverage, under your 3.0× bar. Three reps carry the whole gap:

RepQuotaPipelineCov.Gap to 3×
Priya Sharma$2.1M$3.8M1.8×$2.5M
Marcus Webb$1.5M$3.2M2.1×$1.3M
Jordan Kim$1.9M$4.6M2.4×$1.1M

Closing to 3.0× needs $4.9Mof net new pipeline, over half of it on Priya's book.

Opportunity · 1,284 rowsQ3 quotas · v4Rule · coverage ≥ 3.0×
Answered in 1.2sRead 3 sources, wrote nothingOpen in Assemble
Artifacts

Your board deck, but it writes itself.

Pin any answer as an artifact — a living document that re-runs its queries on schedule. Board decks, QBR prep, investor updates: draft once, stop redoing them.

  • Charts, tables, and prose that stay in sync with the source
  • Export to PDF, Notion, or Google Slides with one click
  • Versioned; diff any two snapshots
FY26 Q1 Board Reviewv14Live

Prepared for the board · rebuilt today 06:04

Growth is compounding and payback is shortening

ARR
$23.4M↑ 40.1%from $16.7M last year
Net revenue retention
118%↑ 4 ptstrailing twelve months
CAC payback
11.2 mo↓ 0.8 moblended, new logo

Net new ARR of $2.0M is the strongest quarter on record, carried by three expansions — Acme, Fjord and Helio — and by Enterprise gross retention holding at 97%.

Net new ARR by quarter$M
FY24 Q2arr_by_quarter.sql · re-ran 06:04FY26 Q1
14 queries · 3 sourcesv13 → v14 diff · next run 06:00
Agents that act

Not just dashboards. Agents that pick up the pencil.

Monitoring is table stakes. Assemble agents draft memos, email customers, update records, and route approvals — inside the permission scope you give them, with a paper trail on every action.

  • Scoped permissions — what each agent may read and write
  • Human-in-the-loop or fully autonomous, per agent
  • Every action reversible; every action audited
DSO Collector
Weekdays 09:00 · run 128 · finished in 47s
Live
Reads
open AR
Writes
dunning · activity
Asks first
anything over $50K
  1. Read open AR from NetSuite

    214 invoices · 38 past due · $1.84M outstanding

  2. Drafted 12 dunning emails

    Template AR-2 · firm tone, no discount offered

  3. Sent 9 emailsSent

    Each below the $50K auto-send limit

  4. Held 3 for approvalWaiting

    Acme $128K · Fjord $84K · Helio $61K

  5. Logged 12 activities to SalesforceLogged

    Reversible for 24h · run 128 in the audit log

3 emails need Dana's approval
Over the $50K limit you set · held for 4h, then dropped
ReviewApprove all
Data sources

Plug in everything you run on.

Ingested once, versioned, deduplicated at the row level, and re-queryable by any agent. No pipelines, no warehouse project, no data team.

Read-only by defaultRow-level dedupEncrypted at restYour region or your VPC
Integration index17 of 24 connected in this workspace

Databases

4/6
  • Postgres2.8M rows
  • MySQL1.1M rows
  • Snowflake14.2M rows
  • BigQuery9.4M rows
  • RedshiftAvailable
  • ClickHouseAvailable

Files & docs

4/6
  • CSV / Excel842 files
  • PDF318 docs
  • Notion1,204 pages
  • Google Drive2,880 files
  • ConfluenceAvailable
  • SharePointAvailable

Business apps

5/6
  • Salesforce41k records
  • HubSpot18k contacts
  • Stripe112k events
  • Zendesk8.2k tickets
  • Linear3.1k issues
  • SlackAvailable

Protocols

4/6
  • MCP servers14 active
  • REST APIs9 endpoints
  • WebhooksLive
  • S3 / GCS1.4 TB
  • GraphQLAvailable
  • SFTPAvailable

24 native integrations · 400+ more via MCP · or point it at any REST endpoint you already run

Agent fleet

A team of agents, not a single assistant.

Each one scoped to a domain, trained on your rules, and wired only into the systems it's allowed to touch. Some monitor. Some act. All of them augment your people.

Fleet4 of 7 running
  • monitor2
  • act3
  • augment2

Pipeline Coverage

Running
monitorRevenue

Watches weekly coverage by rep, flags anyone below 3×.

ReadsWrites
  • Pings #rev-ops
  • Drafts coaching notes

DSO Collector

Running
actFinance

Spots overdue invoices and sends reminder emails with context.

ReadsWrites
  • Emails AP contacts
  • Posts to NetSuite

Churn Canary

Running
monitorCustomer Success

Flags low-seat accounts trending toward cancellation.

ReadsWrites
  • Alerts the CSM
  • Opens a Linear ticket

Deal Desk

Scheduled
actRevenue

Auto-drafts approval memos from Salesforce opportunity data.

ReadsWrites
  • Writes the memo
  • Routes for signoff

Support Triage

Running
actSupport

Reads the Zendesk backlog, tags by theme, escalates P0s.

ReadsWrites
  • Tags tickets
  • Pings on-call

Board Prep

Scheduled
augmentFinance

Rebuilds the board deck every Monday with fresh numbers.

ReadsWrites
  • Updates the artifact
  • Drafts commentary

Forecast Judge

Scheduled
augmentRevenue

Compares rep commit against Assemble's own projection.

ReadsWrites
  • Flags variance
  • Suggests adjustments
Spin up anotherDescribe what it should watch or do — Assemble scaffolds the agent, its scope and its schedule in seconds.
41
agents per customer
average by month three
8,200
actions every week
across a mature fleet
2.3 hrs
analyst time returned
per agent, per week
Enterprise-ready

Built for the compliance review, not just the demo.

Your data never trains a shared model. Every agent runs in a permission scope you control. Everything an agent touches leaves a diff.

Get started freeTalk to security
01

SOC 2 Type II

Audited annually by an independent Big Four firm.

02

Your data, your region

Hosted in your region or your own VPC. Your data never leaves your perimeter.

03

Granular agent scopes

Every agent has an explicit read/write scope. Nothing an agent can do that a human can't.

04

Full audit trail

Every action an agent takes is logged, diffed, and reversible.

05

SSO / SCIM

Okta, Azure AD, Google Workspace. SAML, SCIM provisioning, just-in-time access.

06

Private models

BYO model provider. Nothing is trained on your data. Ever.

Your company has an operating brain.
It just doesn't know it yet.

20-minute walkthrough. Bring one question you asked your last board meeting — we'll answer it live, with sources.

Get started freeBook a walkthrough
  • No credit card
  • Connect a source in under 5 minutes
  • SOC 2 Type II
  • Your data never trains shared models

Command Palette

Search for pages, actions, and settings