Explore the decision before you commit.
What-if scenarios
Change an assumption and compare the outcome with the original plan. See the effect on cash, stock or capacity without changing the underlying records.
What if the largest receipt arrives late?
Change one thing
Change one figure, run the same sums again and compare. The same change always gives the same answer.
Many possible outcomes
Run the scenario many times using realistic ranges from your forecasts, and see the spread of results as well as the middle.
Step by step
Play the process forward one step at a time, with queues and capacity, to see where work backs up.
What matters most
Find which figure really moves the result, and what happens in a bad case as well as an average one.
Because they answer different questions.
A forecast estimates what is likely. A scenario shows what would happen if something specific occurred, including unlikely things, which are often the most useful to test. Mix the two up and a worst case ends up quoted as the plan, so they are kept apart and labelled differently.
What if £75,000 arrives after the forecast period?
Keep the same planned payments and change only the receipt timing.
Cash simulationIllustrative records| Scenario | Receipts | Payments | Closing cash |
|---|---|---|---|
| On-time receipt | £90,000 | £210,000 | £60,000 |
| Late receipt | £15,000 | £210,000 | −£15,000 |
The late-payment scenario ends £15,000 below zero.
Both scenarios start with £180,000. Review collection timing and payment options without changing the original forecast or bank records.
- It takes realistic ranges from your forecasts, rather than a single figure that won’t hold.
- It uses your own rules and sums, so the arithmetic in the scenario matches your books.
- It uses real timings and capacity from process mining, so queues behave the way yours do.
- It sends plan comparisons back to planning, so you can pick the plan that holds up in a bad week, not only the one that looks best on average.
A scenario doesn’t promise or approve anything.
Scenario results are easy to misuse, so the labels are strict. A worst case is labelled as a worst case. A model of part of your business says which part it covers. No scenario can approve anything, and if one suggests an action, that action goes through the same approvals as any other.
Before you commit.
- “What happens to cash if this supplier is two weeks late?” — the effect on every order that depends on them.
- “What if we take this rush order?” — what it pushes back, and what that costs elsewhere.
- “How reliably can we really deliver on time?” — tested against your safety stock, not a target someone typed in.
- “Which of these two plans survives a bad week?” — the difference an average hides.
- “How much does this depend on the supplier’s delivery time?” — the figure that really moves the answer.
Frequently asked questions.
Is a simulation result a prediction?
No. A prediction estimates what is likely. A scenario shows what would follow from something you chose to test, often something unlikely on purpose. That’s why every result is labelled as a scenario.
Where do the ranges come from?
From Cobalt ERP’s forecasts, with their ranges kept. Using one average figure instead is how scenarios end up confidently wrong.
Can I run a scenario on my live figures?
Yes. It reads your live records but changes nothing, and nothing it produces can approve a change.
What is a bounded digital twin?
A model of one part of your business, with what it covers and what it leaves out stated clearly. Those limits are shown with every result, not left in the small print.
Try it on your own records.
Book a demo and ask it a real question about your business. Every answer shows how it was worked out and which records it used.
Book a demo