The question that decides every engagement: “And how do we know that?”
Process consulting rarely fails on the analysis and almost always on the proof. FlowVisual grew out of exactly that problem: a tool that runs alongside the workshop, shows the constraint under variation, and produces a document that carries your letterhead and survives a follow-up question in the steering committee.
In a consulting engagement, process simulation is above all an instrument of proof. You model the current process live in the workshop while the departments sit beside you and object — which produces agreement about the model before anyone argues about the result. The stress test then shows which step limits throughput, where the constraint moves after a measure, and what that measure is worth in time and money, each as a P10–P90 band rather than a single value. The result leaves as a branded PDF. Everything runs locally on your machine, with no account and no cloud, which removes the approval discussion about client data.
30min
From empty window to the first defensible number.
P10–P90
A range instead of false precision — the form a CFO signs.
2PDF
Proposal for the decision maker, documentation for traceability.
none
No account, no transfer, no data processing agreement.
The problem is not the analysis, it is the proof
Experienced consultants often see the constraint after half a day. The problem starts afterwards.
The finding stands, the measure is obvious, and then comes the one sentence in the steering committee that experience cannot defend against:
"That is your assessment. How do we know the number is right?"
Three usual answers, and why they do not hold:
- Benchmark. "In comparable organisations this takes 40 % less time." The client is not comparable, and they know it.
- Spreadsheet. An Excel calculation with processing times added up. It knows no queue, no peak days and no coupling — and anyone who knows about one of those three gaps can dismiss the number in one sentence.
- Experience. Carries where trust already exists, but not against an internal controller who is defending their own project.
What holds is a model the department helped build, that runs under variation, and whose assumptions are printed in the report. Not because it is more accurate, but because it is checkable.
Three moments in an engagement
Pitch — before the contract exists. A five-step model built in twenty minutes from the figures mentioned in the first conversation shows a prospect, on their own process, where the backlog forms and roughly what a measure would be worth. That is a different conversation from a slide about somebody else's reference. Label it as a coarse model — the honesty is part of the effect.
Workshop — agreement about the model. Model live while the departments watch. Objections in the workshop are a win: every correction raised there is one that will not be raised in the final presentation. Once the picture runs and everyone nods, the basis is settled — from then on the discussion is about measures, not about reality.
Close — the proof that outlives you. Save the baseline, change exactly one lever, compare, export. The report carries your letterhead, your logo and your accent colour. It contains the before/after case in money, the range and the assumptions. It gets passed around internally long after you have left — and carries your name when someone asks a year later who ran those numbers.
Client data never leaves your machine
The underrated advantage in an engagement: what is never uploaded needs no approval.
FlowVisual is a desktop application. No account, no sign-up, no transfer of model data to a server. Three practical consequences:
- No data processing discussion about yet another tool in your kit. With clients in public administration, healthcare, insurance and government, this is regularly the difference between "next week" and "next quarter".
- No data flowing back that you would have to explain in your own confidentiality agreement.
- Works offline — on the plant floor, in a meeting room without guest Wi-Fi, on a train.
What you model is volumes, durations and capacities, not personal data. But in case of doubt somebody has to verify that — and the verification disappears when nothing leaves the device.
The tool is one half of it
A simulator does not replace a method. It makes one provable.
FlowVisual was built inside consulting engagements at Balane GmbH and is the tool for the process consulting method Flowrefy. Anyone looking for the method behind it — approach, toolkit, worked example analyses — will find it there. Anyone who only needs the instrument does not have to adopt the method.
What has proven itself as a sequence in engagements:
- Scope it in one sentence. "From receipt of the invoice to release for payment." Without that sentence two departments discuss different processes.
- Model coarsely, five to ten steps. The constraint almost never sits in the detail. Fine detail costs time and does not move the result.
- Ask for ranges, not point values. "Between five and fifteen minutes" is a better answer than "ten on average" — and departments give it more readily, because it is more honest.
- Keep assumptions visible. Every estimated figure appears as an assumption in the report. That takes the sting out of attacks before they come.
- One lever per run. Three simultaneous changes produce a number nobody can attribute to a lever — and therefore no business case.
When you need a different tool
- Factory, warehouse or terminal design → a simulation lab (Arena, Simul8, AnyLogic, FlexSim, Plant Simulation). Transport logic, setup matrices and experiment designs are deliberately absent here.
- Standards-compliant BPMN documentation → a BPMN tool. There are six building blocks here and no notation validation; see BPMN simulation.
- Measuring the past from event logs → process mining. FlowVisual computes an assumption, not a log.
- A process archive across hundreds of flows with approval workflows → a BPM suite.
The full categorisation is in the article Process simulation software: four categories.
- Where it is used
- Live in the workshop, in the pitch, in the final report
- Report
- PDF with your own letterhead, logo and accent colour — proposal and documentation kept separate
- Form of statement
- P10–P50–P90 for time and money, utilisation on the peak day, constraint migration after a measure
- Operation
- Desktop, macOS 13+ and Windows 10/11, works offline, no account
- Client data
- Stays on the device — no transfer, no data processing agreement required
- Cost
- Modelling and stress testing are free; saving, comparing and export via the respective store
Frequently asked
Can I brand the reports with my own logo?
Yes. Letterhead, logo and accent colour are set once in preferences and then apply to every export. There are two outputs: the proposal PDF for the decision maker — before/after, impact in money, range — and the documentation PDF with model, parameters and assumptions for traceability.
Is this enough for a business case a CFO will sign?
For administrative and service processes, as a rule yes — provided the statement comes as a range and with visible assumptions. What sinks a business case is rarely insufficient computational depth; it is a number with two decimal places, no provenance, and no statement of what it does not cover.
Does the client have to buy the software too?
No. You model and compute on your device and hand over the result as a PDF. If the client wants to continue computing themselves, they can install it — modelling and stress testing cost nothing. For handovers inside running programmes that is often better than another licence contract.
How do I explain to a client that the numbers are estimates?
By writing it down. Every estimated input appears as an assumption in the documentation PDF, and every result comes as a P10–P90 band. That openness is not an admission of weakness; it is the reason the number holds. A value with a stated range and provenance is harder to attack than a point estimate with neither.
What is the difference to Flowrefy?
Flowrefy is the consulting method of Balane GmbH; FlowVisual is the tool for it. The method covers approach, interviewing and evaluation; the tool computes. You can use the tool inside your own methodology without adopting Flowrefy.
Build your first engagement model
Open a template, enter the figures from your last pitch conversation, run the stress test. Twenty minutes later you will know whether you want to take it into the next meeting.
Guide: seven steps to the number- Comparison
Process simulation software: four categories, four purposes
“Which tool should I buy?” is the wrong question. Software for business processes falls into four categories with four different purposes — and most bad purchases happen because someone bought from the wrong category.
Read - Method
Process costs in Excel: four mistakes that sink any business case
Nearly every business case for process improvement is built in Excel. And nearly every one contains the same four mistakes — not through carelessness, but because a spreadsheet structurally cannot represent certain things.
Read - Method
Reading P10, P50 and P90 — and where the average misleads you
Percentiles are not statistician's vanity; they are the only honest way to report a result that varies. Three numbers, three purposes — and three misreadings that routinely end up in decision papers.
Read