Workflow Orchestrator
Link your automations into one tracked process.
Workflow Orchestrator
Operations & E-commerce
- RuleDone:
Pick a chain template and connect your deployed automations.
- PersonWaiting for approval:Waiting for you
A process owner (not the builder) approves the exact chain.
- RuleNot started:
A start event begins the chain with one tracking id.
- RuleNot started:
Each step hands typed data to the next and is checked on finish.
- RuleNot started:
Failed steps retry or go to the exception queue with a summary.
- RuleNot started:
You retry, skip (done by hand) or stop from the exception card.
For example: A home-goods seller runs order-to-ship.
Waiting for you. Nothing has changed yet.
Review for important actions. Customer-facing and record-changing actions are reviewed first.
What is Workflow Orchestrator?
In shortWorkflow Orchestrator is an AI automation for Operations & E-commerce teams that works with Slack. Runs a multi-step business process as an approved chain of Charviam automations, with one tracking id, safe retries, an exception queue and a kill switch. It is built for operations or automation leads and process owners.
Last reviewed
Key facts
- Department
- Operations & E-commerce
- Starts when
- The first automation in a chain is triggered, or a previous step in the chain finishes
- Works with
- Slack
- Run modes offered
- Shadow, Draft, Assisted and Auto
- Human control
- Review for important actions
- Status
- Available now
The problem today
Businesses using three or more automations glue them together with spreadsheets and people. Hand-offs get lost, nobody sees the whole process, and a failure in the middle leaves half-done work.
What it does
You choose a chain of existing automations, such as order received, then Order Processing, then Shipping, then a task. A process owner approves the chain, and that version is locked. Each step runs in its own mode with its own approvals. Failures retry safely or land in an exception queue with a plain summary, and one switch can stop the whole chain.
How does Workflow Orchestrator work?
How it works- Rule you set
- AI step
- A person decides
- Step 1Rule
Pick a chain template and connect your deployed automations.
- Step 2Person
A process owner (not the builder) approves the exact chain.
- Step 3Rule
A start event begins the chain with one tracking id.
- Step 4Rule
Each step hands typed data to the next and is checked on finish.
- Step 5Rule
Failed steps retry or go to the exception queue with a summary.
- Step 6Rule
You retry, skip (done by hand) or stop from the exception card.
What you get
Key features- Three starter templates: order-to-ship, lead-to-meeting, ticket-to-fix
- One chain id across every step for full traceability
- Safe retries that never start the same step twice
- Per-step failure rule: stop, continue, or undo (compensate)
- Limits on steps, run time (72 h default) and cost ($1.00 default per chain)
- Kill switch per chain, tenant or globally
How much does it do on its own?
Run modesPick a mode to see what Workflow Orchestrator would do at each step.
Every automation starts in Shadow. Move it up when you are ready, and back down at any time. How run modes work
Workflow Orchestrator in Shadow
Runs on real data and shows what it would have done. No app is changed and no message is sent.
- Rule
Pick a chain template and connect your deployed automations.
Watches for it
- Person
A process owner (not the builder) approves the exact chain.
Shown to you as a suggestion
- Rule
A start event begins the chain with one tracking id.
Shown to you as a suggestion
- Rule
Each step hands typed data to the next and is checked on finish.
Shown to you as a suggestion
- Rule
Failed steps retry or go to the exception queue with a summary.
Shown to you as a suggestion
- Rule
You retry, skip (done by hand) or stop from the exception card.
Shown to you as a suggestion
Shadow vs Auto for Workflow Orchestrator
Auto is the most it can do on its own. Here is each step side by side.
| Step | Shadow | Auto |
|---|---|---|
| In short | Runs on real data and shows what it would have done. No app is changed and no message is sent. | Runs without approval, with limits, alerts and a kill switch. Unlocks after a trial period in Assisted. |
| Step 1Pick a chain template and connect your deployed automations. | Watches for it | Runs on its own within your limits |
| Step 2A process owner (not the builder) approves the exact chain. | Shown to you as a suggestion | Within your limits; anything outside them goes to a person |
| Step 3A start event begins the chain with one tracking id. | Shown to you as a suggestion | Runs on its own within your limits |
| Step 4Each step hands typed data to the next and is checked on finish. | Shown to you as a suggestion | Runs on its own within your limits |
| Step 5Failed steps retry or go to the exception queue with a summary. | Shown to you as a suggestion | Runs on its own within your limits |
| Step 6You retry, skip (done by hand) or stop from the exception card. | Shown to you as a suggestion | Runs on its own within your limits |
| True in every mode | AI never adds, reorders or skips steps; chains are fixed config. | |
What it will never do on its own
Safety & human control- AI never adds, reorders or skips steps; chains are fixed config.
- Two-person approval for every chain version; approved versions can't change.
- A step never runs in a higher mode than its own deployment, and high-risk steps keep their approvals.
- Undo steps that touch money, customers or legal need an approval first.
- Restricted automations (30, 41) can never be chained.
- You approve before it acts
- Your data stays yours
- It knows its limits
- Stop anytime
What will you see on your dashboard?
Results you can trackTargets are quality goals the automation is built to reach. Estimates are calculated from your own volumes.
- Tracked
Chains started / succeeded / failed / stopped, time per step, cost per chain, exception age
Measured on your own data
- Tracked
End-to-end process time
Measured on your own data
- Tracked
% of processes with no manual hand-off
Measured on your own data
- Target
Chains finishing without manual rescue ≥ 97% (target)
A goal it is built to reach
- Target
Failed chains recovered in under 2 hours (target)
A goal it is built to reach
- Target
Duplicate step runs: 0 (target)
A goal it is built to reach
- Target
Kill switch stops new actions within 30 s (target)
A goal it is built to reach
A home-goods seller runs order-to-ship. When the shipping step fails because the carrier is down, the chain retries. Then it opens an exception card saying which order and step failed, and the ops lead clicks Retry an hour later.
Who is Workflow Orchestrator for?
- Operations or automation leads
- Process owners
Industries: E-commerce & retail, Manufacturing & distribution, SaaS & tech, Professional services
Also useful for: IT & Data Operations, Sales & CRM
Connects to the apps you already use
Works withApp names are trademarks of their owners. Charviam connects to them; it is not a partner or reseller.
- Slack(optional)
Better together
Automations that hand work to Workflow Orchestrator, or pick up where it stops.
What do you need to set it up?
Setup- Deploy the automations the chain uses
- Pick a template and fill in your deployment ids
- Name the process owner and exception owner
- Set limits (steps, retries, hours, cost) and an optional Slack channel
- Run shadow for 2 weeks per chain, then draft
Typical setup
- 01Connect your apps
- 02Configure the rules
- 03Run a watch-only trial
- 04Go live
Not included
- A free-form AI agent choosing tools
- New business logic (each step is an existing automation)
- New app connections
Questions about Workflow Orchestrator
FAQSee Workflow Orchestrator on your own data
WalkthroughIt starts in Shadow, where it only watches and suggests, so nothing changes in your apps until you say so.