Process orchestration

Process Orchestration in Salesforce

Process orchestration, also called business process orchestration, coordinates the steps of a business process across multiple systems, automation, AI, and people so that the process produces a consistent outcome from intake to resolution. A single automation runs one step. Orchestration governs the whole sequence: what should happen, in what order, under what conditions, and with what controls.

Salesforce environments raise the question because so much can act on the same record independently. Flows fire, assignment rules run, integrations sync, and agents take actions, each doing its part without a single definition of how the end to end process should run. Add the applications installed from AppExchange and the external systems the process reaches, and coordination becomes the hard part, not capability.

The signal that a process is orchestrated is operational, not technical. Anyone can see where a piece of work is, what has happened to it, and what happens next, without asking a person. Where that is missing, teams hold the process together with manual handoffs, queue monitoring, and fixes after the fact.

Workflow orchestration coordinates one operational workflow from start to finish. Lead routing, case handling, claims, and onboarding are each a workflow. Process orchestration coordinates several of those workflows, along with the systems, AI, and people they touch, into one business process. See operational workflow management in Salesforce for the workflow layer in depth.

A refund is a useful example. It may span a service workflow, a finance approval, and an update in an external billing system. Each of those is a workflow; the refund as a whole is a process. Workflow orchestration makes each part run reliably. Process orchestration makes the parts run as one. In Salesforce, both run on the same foundation, and most organisations need workflow orchestration first and process orchestration as the picture grows.

Process orchestration has moved from a nice-to-have to an architectural priority for four connected reasons.

Business complexity

A modern Salesforce organisation coordinates dozens of applications, hundreds of Flows, several AI services, and distributed teams. That complexity created the coordination problem. AI has accelerated it, not caused it.

Everything can now participate

Every application, workflow, API, and AI capability can now take part in a business process, through APIs, agents, and standard interfaces. Work can start almost anywhere, but starting is not the same as governing what happens next.

Automation proliferation

As orgs mature, Flows, rules, and integrations multiply until small changes break things elsewhere and no one owns the end to end result. More automation without coordination adds fragility rather than control.

Governance and auditability

As AI enters operational processes, being able to explain what happened, where AI was involved, and why, becomes a requirement rather than a preference, especially in regulated work.

Approaches to process orchestration fall into a few broad categories. Each handles part of the picture well; they differ in scope and in where the process logic lives.

Native workflow tools

Salesforce Flow Orchestration and Omni-Channel coordinate steps and route work inside Salesforce. They are strong for structured, in-platform sequences and are already licensed. Their scope is the Salesforce boundary.

Integration platforms (iPaaS)

Tools such as MuleSoft and Workato connect systems and move data between them. They excel at integration. Connecting systems is not the same as governing the process logic and decisions that run across them.

Business process management (BPM) suites

Platforms such as Camunda, Pega, and IBM offer deep process modelling and execution. They are powerful for enterprise process management, and typically run outside Salesforce, which adds a system to integrate and govern.

Low-code automation

Flows and point automation tools handle individual steps and simple sequences quickly, and are accessible to operations teams. They tend to sprawl as the number of processes grows.

Salesforce-native orchestration

ISV layers that add end to end orchestration inside Salesforce, including Ortoo Orchestrator, coordinate automation, AI, and people on the platform without moving data out. This keeps execution and governance in one place; the trade-off is adopting an additional managed package.

1. 01

   Process

   The business outcome and its steps

2. 02

   Systems

   Apps, Salesforce, external systems

3. 03

   Automation

   Flows, rules, integrations

4. 04

   AI

   Models and agents where they help

5. 05

   Humans

   Decisions and handoffs

6. 06

   Governance

   Control, visibility, audit

We frame Salesforce process orchestration across six layers. Mature organisations address all six. Earlier-stage orgs usually have some in place and coordinate the rest by hand.

Process

The business process itself: the outcome it produces and the sequence of steps to get there, spanning more than one workflow. Maturity: the end to end process is defined in one place, not implied by how separate automations happen to interact.

Systems

The applications and data the process spans, from Salesforce objects to installed apps and external systems such as ERP and billing. Maturity: the process reaches every system it needs through governed connections, and data stays within its security boundary.

Automation

The deterministic steps: Flows, rules, and integrations that must run the same way every time. Maturity: automation handles the certain steps, and a coordination layer owns the handoffs between them rather than leaving gaps.

AI

Models and agents applied where interpretation adds value, such as reading a request or extracting data. Maturity: AI is used selectively, with specific agents handling narrow tasks, rather than one general agent improvising across the process.

Humans

The points where people make a judgement, approve, or handle an exception. Maturity: human steps are defined parts of the process with clear inputs and handoffs, not an informal catch-all for when automation stops.

Governance

Control, observability, and audit across all of the above: what ran, where AI was involved, who decided what, and why. Maturity: every step is traceable, and the process can be explained and changed with confidence.

When evaluating a process orchestration approach for Salesforce, a handful of questions consistently separate approaches that scale from those that add complexity. Frame each as a question to ask any vendor, and yourself if you plan to build.

1. Can you see, after the fact, what happened at each step and why?
2. Can you define which steps are deterministic and which are AI-assisted, and control where AI runs?
3. How are human decisions and handoffs modelled and governed, rather than left to informal practice?
4. Does it run natively in Salesforce, or does process data leave the org?
5. Can the orchestration survive failure? Retries, waiting, resumability, and partial completion when a step or system is unavailable.
6. How does cost scale as volume grows, and is it predictable?
7. Can operations change the process without a development cycle?

The answers reveal how much of the process is genuinely governed, and how much still depends on people noticing and stepping in.

Ortoo Orchestrator is the Salesforce-native orchestration approach from the landscape section. It coordinates automation, AI, specialist agents, and people as one process, running on your Salesforce objects, Flows, and APIs, and reaching external systems through governed connections. Data stays in the org except for explicit AI calls.

In terms of the six layers, it applies AI selectively, uses deterministic logic where certainty matters, models human handoffs as defined steps, and records every step for governance. Pricing follows work completed rather than tokens or actions, and operations teams configure processes without code.

It is an execution and governance layer inside Salesforce, not a general-purpose iPaaS or a standalone BPM suite. Organisations whose processes live largely outside Salesforce may pair it with, or prefer, a different foundation. See why teams choose Ortoo Orchestrator for the category argument.