FAQ — FlowSharp: Governed AI Process Orchestration

Question 1

What is FlowSharp and what is it used for?

Accepted Answer

FlowSharp is a Governed AI Process Orchestration platform: it transforms business processes scattered across people, email, spreadsheets, business systems, automations and AI into a governed, measurable operating system. The product's promise is direct: first make the process visible, then govern it, then automate where value is measurable. With FlowSharp every process becomes a Flow: a sequence of states, transitions and tasks with explicit rules, defined roles and a complete audit trail. Operators always know what to do, automations work inside declared guardrails and management measures time, costs, bottlenecks and ROI in real time. FlowSharp is suited to operational teams, SMEs and structured organisations that want to move beyond implicit process management — emails, spreadsheets, chats — and bring every case into a transparent, tracked and improvable system. Typical use cases: order management (Order to Cash), customer onboarding, complaint handling, document workflows, purchase requests, and processes with a human AI in the loop.

Question 2

How does the FlowSharp Flow Designer work?

Accepted Answer

FlowSharp's Flow Designer is a visual canvas that lets you build the operational model of a process without writing code. How to use it: States (or phases) are the canvas nodes: they represent moments in the journey such as "Under review", "Awaiting data" or "Approved". Transitions are the connections between states: they define which steps are allowed and in which direction. The side panel lets you configure each state with tasks (manual, automatic, webhook, AI), advancement rules, required data, timeouts and operational behaviour. Two creation modes: Manual — start from a blank canvas and build the flow phase by phase. With AI — describe the process in natural language and get a first draft to refine with your team. Before publishing, FlowSharp automatically verifies the structural correctness of the flow: unreachable states, loops without an exit and Auto/AI states without a fallback are flagged with correction guidance. The published flow becomes immediately operational: real items start moving through the states, generate tasks and produce measurable data.

Question 3

How does FlowSharp manage AI with governance and control?

Accepted Answer

FlowSharp adopts a AI with governance approach natively: AI is not a layer added on top of the process but operates inside the process with declared rules, permissions and controls. AI governance mechanisms: Data Shield — protects sensitive data at field level. Data marked as sensitive is never exposed to AI models, even when the AI task operates on the same item. Sensitivity follows the data wherever it is copied or transferred. Data Gate — explicit checkpoints in the flow that block advancement until certain data or conditions are met. Prevents cases from proceeding on incomplete or unverified bases. Native human-in-the-loop — AI tasks can require human confirmation before producing effects. The operator sees the AI output, evaluates it and decides whether to proceed, correct or escalate. Roles and permissions — every AI action is configured with the role that authorises it. A model cannot access data or perform actions outside the declared perimeter. Complete audit — every AI output, every confirmation and every deviation is recorded in the item's timeline with timestamp, responsible party and action detail. This approach allows AI to analyse, classify, suggest and assist without losing operational control or traceability.

Question 4

Does FlowSharp integrate with n8n and other automation tools?

Accepted Answer

FlowSharp is designed to work in synergy with automation tools like n8n. The guiding principle is: FlowSharp governs the process; n8n (or any other system) produces data, receives tasks or reacts to events. Integration patterns with n8n: Start a flow from n8n — an n8n workflow can create a FlowSharp item via API trigger (POST with API key). Useful when a lead arrives from the CRM, an order from e-commerce or a request from an external portal. Receive webhooks from FlowSharp — when an item changes state or completes a task, FlowSharp emits an event. n8n can subscribe to these events via webhook and react (send emails, update the CRM, notify Slack). Async tasks with callback — FlowSharp can assign a task to n8n (webhook task). n8n performs the processing and calls back the FlowSharp callback to complete the task and unblock the flow. Flow Agent and MCP — FlowSharp exposes an MCP (Model Context Protocol) interface for AI agents like Claude that can query process state, start flows or complete tasks from chat. Integration checklist: API key configured in FlowSharp Trigger endpoint for starting items Webhook subscriber for events Callback URL for async tasks

Question 5

How much does FlowSharp cost? Is there a free plan?

Accepted Answer

FlowSharp is available with a free plan that lets you get started without a credit card — ideal for exploring the platform, designing your first flows and understanding how it fits your processes. How the pricing model works: The free plan includes core features for designing flows, managing items and measuring processes with a limited number of users and active items. Paid plans scale on dimensions such as: Number of active users and roles Volume of items processed Access to advanced features (advanced Analytics, governed AI, Process Maps, Saving Challenge, enterprise integrations) Dedicated support and SLA Saving Challenge — FlowSharp also offers a premium contract that compares a frozen cost baseline with results measured over time, allowing you to demonstrate the actual ROI of adoption. To find the plan best suited to your needs, the best next step is to book a demo where we evaluate your size, processes and goals together. 👉 Start for free or book a demo to talk to the team.

Question 6

What is Data Shield in FlowSharp and how does it protect sensitive data?

Accepted Answer

Data Shield is the sensitive data protection mechanism within FlowSharp flows. It solves one of the most critical problems in using AI in business contexts: how do you prevent confidential data from being inadvertently exposed to AI models or unauthorised users? How it works: Every data field in a FlowSharp item can be marked as sensitive by a FlowDesigner or Admin. This marking: Blocks AI access — AI models configured in the flow never receive the value of that field, even when analysing the complete item. They see the field as empty or redacted. Follows the data — if the item is copied, transferred to another flow or exported, the sensitivity marking is preserved. It is not possible to "wash" sensitivity by changing flows. Respects roles — only users with the appropriate role (Operator or above, depending on configuration) can see the value in clear text. ReadOnly users never access sensitive data. Tracked audit — every access to sensitive data is recorded in the timeline with user, timestamp and context. Typical use cases: Tax data and tax codes Credentials and access tokens (the secret type in variables) Medical or legal information Payment data and IBANs

Question 7

How does human-in-the-loop work in FlowSharp processes?

Accepted Answer

Human-in-the-loop is one of FlowSharp's founding principles: many business processes should not be fully automatic. They need approval, review, evaluation, correction, document requests and escalation. In FlowSharp people are part of the process, not exceptions to it. How it is implemented: Manual tasks — a manual task assigns a specific action to an operator (or a role). The item does not advance until the task is completed. The task can include forms, checklists, document uploads or any operational action. Approval tasks — a variant of the manual task dedicated to decision workflows: the operator sees the item context, the AI proposal (if any) and decides whether to approve, reject or defer. AI output review — when an AI task produces a result, the flow can require human confirmation before proceeding. The operator sees the output, evaluates it and decides. Escalation and deviations — if an item exceeds time or WIP limits, FlowSharp generates alerts visible to operators and Admins. Deviations are tracked in the timeline. Team handoffs — transitions between states can be configured to change the responsible team, ensuring an explicit and tracked handover between departments. This approach ensures that automation accelerates human work without replacing it where decisions matter.

Question 8

What is a Process Map in FlowSharp and when should you use it?

Accepted Answer

A Process Map in FlowSharp is the level at which a business process is seen in its entirety: operational flows, departments, external systems, documents, manual steps, events and responsibilities. The difference between a Flow and a Process Map: A single flow answers the question: "How does this object move through a procedure?" A Process Map answers a broader question: "How does this process really work, from start to finish, even when it crosses multiple flows, people and systems?" What a Process Map contains: Linked flows — the FlowSharp flows that make up the process External systems — ERP, CRM, e-commerce, customer portals Documents — contracts, invoices, forms, reports Manual steps — activities that do not go through FlowSharp but are part of the real process Departments and responsibilities — who does what at each stage When to use it: To visualise and communicate an end-to-end process to stakeholders To identify bottlenecks that span multiple flows To document the operational architecture before automating To verify that the real wiring (flows, webhooks, systems) matches the design The Process Map is verified against FlowSharp's real wiring: if a flow is linked but not published, the map flags it.

Question 9

Does FlowSharp have an API? How does it integrate with external systems?

Accepted Answer

FlowSharp offers a complete set of integration channels to connect with any external system. Available integration channels: REST API with authentication — RESTful endpoints to read items, create cases, update data and complete tasks. Uses user authentication or API keys. API keys for external triggers — a dedicated key allows external systems (ERP, CRM, e-commerce) to start FlowSharp flows without user credentials. Secure, revocable and tracked. Webhooks and subscribers — FlowSharp emits events on state changes, task completion and transitions. External systems subscribe and react in real time. Callbacks for async tasks — an external system receives a task via webhook, processes it (even for minutes or hours) and calls back the FlowSharp callback to complete it. Ideal for integrations with slow or batch systems. MCP (Model Context Protocol) — interface for AI agents (Claude, GPT) that can query processes, start flows or complete tasks in natural language. Browser extension — allows starting FlowSharp flows directly from any web application open in the browser (CRM, portals, email), without a system API. n8n — native integration with the trigger/webhook/callback pattern described above, documented and tested. All channels are documented in the Integrations and API section.

Question 10

How do you measure the ROI of a process with FlowSharp Analytics?

Accepted Answer

FlowSharp Analytics turns every processed item into operational knowledge: time in phases, completed tasks, manual steps, executed automations, estimated costs and potential savings. What Analytics measures: Time — average time in phases, total throughput time, waiting times and bottlenecks by state. Cost — estimated cost per case based on time × configured hourly rate. Comparable across periods. Automation — percentage of tasks completed automatically vs manually. Indicates the process maturity level. Outcomes and quality — completion rate, deviations, exceptions and blocked cases. Measures process health. Potential saving — estimate of savings achievable by automating remaining manual tasks. Basis for automation initiatives. Actual ROI — with the Saving Challenge contract, FlowSharp freezes a cost baseline and compares it with results measured over time. Savings are documented and verifiable. How to use it: Configure the average hourly cost per role/team Run the real process on FlowSharp Review Analytics after the first operational weeks Identify phases with the most waiting time or highest manual cost Launch automation initiatives starting from the data FlowSharp's principle is clear: measure before automating.

Question 11

Does FlowSharp support MCP (Model Context Protocol) for AI agents?

Accepted Answer

FlowSharp supports the Model Context Protocol (MCP), the emerging standard for connecting AI agents to business systems and tools. What FlowSharp's MCP interface enables: Process querying — an AI agent connected via MCP can ask "How many cases are stuck in 'Awaiting data'?" or "What is the average time in the review phase?" and receive a structured response from real data. Flow triggering — the agent can start a new item in a FlowSharp flow on user request, for example "Open a new onboarding case for client ACME". Task completion — an AI agent can receive a task assigned via MCP and complete it with a structured output, which FlowSharp validates and records in the timeline. Item unblocking — Admins can unblock stuck items directly from the AI agent interface, without accessing the FlowSharp console. Security: MCP access respects the FlowSharp roles of the connected agent Data Shield applies to MCP queries too: sensitive fields are never exposed Every MCP action is audited in the timeline with agent identification The MCP integration is designed for the Flow Agent: the integrated assistant that helps operators and managers query and act on processes in natural language, maintaining governance and traceability.

Question 12

What user roles exist in FlowSharp and what can each one do?

Accepted Answer

FlowSharp clearly separates who designs, who operates, who administers and who observes. The 5 roles are cumulative in permissions (Admin includes everything Operator can do). | Role | What they can do | |---|---| | ReadOnly | Views flows, items, tasks, dashboards and documentation without modifying anything. Ideal for stakeholders, auditors or managers who want visibility without operability. | | Operator | Manages items, completes tasks, executes transitions, updates operational data. This is the daily work role: advancing cases, filling forms, uploading documents. | | FlowDesigner | Creates and modifies flows, Process Maps, rules, tasks, process configurations. Designs and publishes operational flows. Does not manage users or tenant settings. | | Admin | Everything FlowDesigner can do, plus: user and role management, settings, AI providers, integrations, API keys, global variables and tenant configurations. | | Super Admin | Manages tenants, impersonation, plans, demo data and multi-tenant aspects. Reserved for the FlowSharp team for support and platform management. | Best practices: Assign ReadOnly to managers who want visibility without risk of accidental changes Keep FlowDesigner separate from Admin to avoid having process designers also manage access Review roles periodically with the access audit available in Analytics

Question 13

Is FlowSharp suitable for SMEs or only large companies?

Accepted Answer

FlowSharp is designed to work in both SMEs and large organisations, with different features that activate depending on maturity and needs. For SMEs: No dedicated IT team required — the Flow Designer is visual and accessible to operational profiles, not just technical ones. An operations manager can design and publish a flow without developers. Gradual start — begin with a single process (e.g. order management or customer onboarding), measure the result and expand. No need to implement everything at once. Free plan available — to explore the platform without initial financial commitment. ROI visible in weeks — immediate measurement of time and costs makes value tangible even at low volumes. For large organisations: Multi-tenant and multi-team — centralised management with isolation between teams, departments and business units. End-to-end Process Maps — visibility into the complete operational architecture spanning systems, teams and automations. Saving Challenge — ROI contract that demonstrates actual savings over time with a frozen baseline. Enterprise integrations — connection to ERP, CRM, legacy systems via API, webhooks and MCP. Dedicated support and SLA — available in advanced plans. FlowSharp's strength is that it scales with you: start small and grow without changing tools.

Question 14

What are WIP limits in FlowSharp and why do they matter?

Accepted Answer

WIP stands for Work In Progress. A WIP limit in FlowSharp is a configurable capacity cap on an entire flow or a single phase, limiting how many cases can be in that state simultaneously. Why it matters: Without WIP limits, a process can accumulate dozens of cases in a phase without anyone noticing. The team appears busy, but real throughput (cases completed per day) drops. Bottlenecks remain hidden. With WIP limits, FlowSharp makes the problem explicit: when the limit is reached, new cases are queued, diverted to another path or released by priority. The team cannot "forget" that there is an accumulation. How they work in FlowSharp: Configuration — the FlowDesigner sets a WIP limit on a phase (e.g. max 10 cases simultaneously in "Under review"). Behaviour at limit — new cases arriving at that phase can: wait in queue, be diverted, or trigger an alert for the team. Visibility — the operational dashboard shows in real time how many cases are in each phase relative to the configured limit. Analytics — historical data shows how many times the WIP limit was reached and for how long, indicating where more capacity or automation is needed. WIP limits are one of FlowSharp's most powerful tools for making capacity problems visible before they become emergencies.

Question 15

What is the difference between FlowSharp and a traditional BPM?

Accepted Answer

The main difference between FlowSharp and a traditional BPM lies in the approach to measurement and AI. Traditional BPM: Models the process as it should be (ideal process, BPMN diagram) Implementation often requires IT teams and long configuration cycles Measurement happens separately, on aggregated data extracted after the fact AI is an external add-on, often without explicit governance Difficult to modify once in production without impacting in-progress cases FlowSharp: Starts from the process as it really works, including manual steps, exceptions and variants Accessible to operational profiles: the Flow Designer does not require BPMN or developers Measures while the process works: every item produces time, cost and quality data in real time AI is natively governed: Data Shield, Data Gate, human-in-the-loop and audit are part of the flow, not add-ons Flow changes are versioned: a new version can be published without interrupting in-progress cases In summary: FlowSharp is not just a workflow designer — it is an operating system for processes: design, run, measure and improve in a continuous cycle, with AI acting inside declared rules. If you have a traditional BPM and want to add operational measurement, governed AI and incremental automation, FlowSharp can coexist with it or progressively replace it.

Question 16

How do you start a FlowSharp process from an external system?

Accepted Answer

Starting a FlowSharp process from an external system is done via API trigger: an HTTP POST call to the flow's dedicated endpoint, authenticated with an API key. Call pattern: `` POST /api/flows/{flow-id}/trigger Authorization: Bearer {api-key} Content-Type: application/json { "data": { "client": "ACME Ltd", "amount": 15000, "source": "e-commerce" } } `` The response includes the created item's ID and its initial state. Typical trigger sources: E-commerce — a confirmed order starts the Order to Cash flow CRM — a qualified lead starts the onboarding flow Customer portal — a support request starts the complaint handling flow ERP — a received invoice starts the payment approval flow n8n or Zapier — any event in any automation tool can start a FlowSharp flow Security: Each API key is associated with a specific site/system and can be revoked independently Data passed in the trigger is validated against the flow schema before creating the item The trigger origin is tracked in the item timeline Configuration: Create an API key in FlowSharp (Admin → Integrations) Copy the specific flow's trigger endpoint Configure the external system with the key and endpoint Test with a trial call and verify the created item

Question 17

How does FlowSharp ensure the security and compliance of business data?

Accepted Answer

FlowSharp is designed with security as an operational requirement, not as an add-on feature. Every security mechanism is embedded in the flow, not applied as an external layer. Main security mechanisms: Data Shield — fields marked as sensitive are never exposed to AI, ReadOnly users or unauthorised systems. The marking follows the data in all transfers. Secret variables — variables of type secret (credentials, tokens, API keys) are encrypted at rest and never shown in clear text in the console, logs or exports. Granular roles — 5 access levels (ReadOnly, Operator, FlowDesigner, Admin, Super Admin) with separate permissions for reading, writing, configuration and administration. Complete audit — every action on an item (creation, transition, task, data modification, access to sensitive data) is recorded in the timeline with user, timestamp and detail. Immutable. Flow versioning — published flows are versioned. Changes do not impact in-progress cases and every version is tracked with author and date. Multi-tenant isolation — each organisation has its own isolated tenant. There is no data sharing between different tenants. Revocable API keys — each integration key is individual, tracked and revocable instantly without impacting other integrations. For specific compliance requirements (GDPR, ISO 27001, banking or healthcare sector), contact the FlowSharp team for a dedicated assessment.

Question 18

Does FlowSharp have pre-built flow templates to start from?

Accepted Answer

FlowSharp includes a Templates library: pre-built flow templates for the most common business processes, ready to turn into a draft with one click and adapt to your own reality. How Templates work: Go to the Flows → Template Library section Choose the template closest to your process Click "Use this template" — the flow is copied as a draft into your workspace Modify states, transitions, tasks and rules to adapt it to your specific case Publish The template is a starting point, not a constraint. You can add states, modify transitions, integrate AI or webhook tasks and fully customise the flow. Typical available templates: Order to Cash — from order receipt to invoicing and collection Customer Onboarding — document collection, verification, account activation Complaint Management — ticket opening, assignment, resolution, closure Purchase Requests — request, approval, order, receipt Document Approval — review, feedback, signature, archiving HR Process — leave requests, employee onboarding, performance review Sales Pipeline — lead, qualification, proposal, negotiation, closure Standard Operating Procedure (SOP): Each template can be accompanied by a SOP (Standard Operating Procedure): a versioned document that becomes the official guide for operators executing tasks. If your process is not in the library, you can create it from scratch with the Flow Designer or with AI assistance.

Frequently Asked Questions

Product

Features

Integrations

AI Governance

Pricing

Didn't find what you were looking for?