Getting Started with Flow Myna

Flow Myna reads the event data your systems already produce and shows how work actually flows: the routes, the waits, the rework, and where things are sitting right now. Bring your data, approve how it was read, and start asking questions.

Loan applications, order fulfilment, support tickets, manufacturing work orders — the workflow is the same. Flow Myna works out the structure of your data, then shows you how the process runs rather than how it was designed to run.

From data to a first answer

1. Create a workspace

Sign in, open Workspaces, and click New Workspace. Name it for the business area rather than the analysis. About a minute.

2. Upload your data

Open Data Management and click Create New Dataset. You can bring a CSV — one file or several linked ones — or an XES or OCEL 2.0 event log.

# Example: order events from the order-to-cash sample
event_id,order_id,activity,timestamp,user_id,block_reason,release_type
43,SO-000031,Create Sales Order,2025-02-13 08:00:00.603938,sales_user_06,,
44,SO-000031,Block Sales Order,2025-02-13 09:56:00.603938,credit_ctrl_02,Manual hold,
45,SO-000031,Unblock Sales Order,2025-02-14 09:28:00.603938,credit_ctrl_02,,manual

The wizard has five steps:

1. Select Process — roughly what kind of process this is
2. Upload Data — your files, or one of the samples
3. Auto Mapping — agents read the files and propose an object and event model
4. Check Mapping — you review and approve, correcting anything read wrong
5. Transform Data — the approved mapping is executed and the data imported

3. Create a project

A project is a scoped view: which object types and event types are in play, and which events mark the process finished. Two steps and a couple of minutes.

4. Read the process map

The map is drawn from your events. Follow the thick edges for the main route, the red ones for the long waits, and the backwards ones for rework. Click a node or an edge for the numbers behind it.

5. Ask the copilot

• "Show me orders that took longer than 14 days"
• "Which block reasons are most common?"
• "Are invoices for blocked orders paid late more often?"

It applies the filter, gives you the numbers, draws a chart where a chart helps, and the map beside it moves to the same cohort.

Core concepts

Objects and events

Objects are the things in your process — sales orders, order lines, deliveries, invoices, customers. Events are what happens to them, each with a timestamp. The link between the two is many-to-many: one event can touch several objects, and one object accumulates many events.

The lens

Because events belong to several objects at once, the same data can be read through any of them. The lens is the object type the map is centred on. Look at an order-to-cash process through Orders and you see fulfilment; switch to Invoices and you see the billing journey. Same events, different question.

Variants

The distinct paths objects take. A process with one designed route usually turns out to have dozens, and how concentrated they are tells you how disciplined it really is.

Where the analysis comes from

Nothing runs in the background. Copilot answers are computed when you ask the question, and every number on screen comes from your current data and current filters.

What makes Flow Myna different

The mapping is done for you, and you approve it

Most process mining tools need you to map your data to their model by hand, or pay someone to. Here agents propose the model and you check it — which is the part that actually needs your judgement.

It is object-centric

Most tools need one case ID and one linear flow. Real processes have orders, items, deliveries and invoices interacting many-to-many. Flow Myna models that, which is why you can change the lens instead of rebuilding the analysis.

Concurrency is modelled, not flattened

Work that runs in parallel is declared as parallel tracks, so the map draws lanes and names the critical path rather than producing a hundred meaningless variants.

You ask in words

No SQL, no query builder. Ask, read the answer, follow up. Where a definition matters, the copilot proposes one and asks you to confirm it, so the whole project agrees on what a number means.

Everything is a link

Filters live in the URL, so a finding is something you send rather than something you describe.

Next steps

Learn the basics

• What is Process Mining? — the fundamentals
• Object-Centric Approach — why the lens matters
• Quick Start Guide — the whole path once, step by step

Follow the workflow

• Creating Your Workspace — where data and people live
• Uploading Your Data — the wizard in detail
• Creating a Project — scope and process end points

Master the tools

• Process Map — the map and its controls
• Process Variants — the paths through it
• Co-Pilot — asking, and what comes back

Go further

• Filters & Exploration — build the cohorts worth comparing
• Saved Filters — keep the ones that matter
• Collaboration — bring your team in

Getting help

Check the FAQ first — it is written to be honest about what the product does not do yet, as well as what it does.

Otherwise, use the Feedback control in the app or email support@flowmyna.com.