Ask almost anyone involved in investment decisions whether scenario modeling is important, and you’ll get a consistent answer. Of course it is. Everyone understands that forecasts are uncertain, assumptions change, and outcomes rarely play out exactly as planned.
But look at how most business cases are actually built, and a different picture emerges. Instead of multiple scenarios, sensitivity analysis, or clear ranges of outcomes, what you typically see is a single set of assumptions and a single projected return, presented as if it’s solid enough to base a decision on.
That gap between what we say matters and what we actually do is where scenario modeling quietly falls away. It’s not controversial, and it’s not misunderstood. It’s simply underused.
Scenario modeling isn’t rejected, it’s squeezed out
In most organisations, scenario modeling isn’t actively dismissed (there’s no meeting where someone dramatically declares it isn’t valuable), it’s just that it tends to get deprioritised as the business case takes shape.
When timelines are tight, the focus shifts toward getting a coherent case in front of decision-makers, so people are scrambling about to align assumptions, build a narrative, and produce a clear ROI figure that can be explained and defended. Scenario modeling, by definition, introduces more moving parts, as it requires additional assumptions, more calculation, and more explanation at precisely the point where teams are doing all they can to simplify to hit deadlines.
So (perhaps understandably?) it often gets pushed to “later” in the process. And because “later” always also coincides with deadlines, reviews, and approvals, it rarely comes back in any meaningful way.
So it’s actually less about capability and more about process; if scenario modeling isn’t absolutely demanded from the outset, it's perhaps inevitable that it doesn’t survive the pressure to deliver quickly.
It makes the conversation harder, not easier
There’s also a more uncomfortable reason why scenario modeling is avoided, namely that it changes the nature of the conversation.
Although hoping it will actually be accurate might be fanciful, a single projected ROI allows for a clean, confident narrative, giving stakeholders something straightforward to align around. Once multiple scenarios are introduced, that clarity starts to fragment, as outcomes vary, assumptions become more visible, and the conversation shifts from “what’s the return?” to “how dependent is this on things going right?”
Now that’s a much more honest discussion, but it’s also a more difficult one. It introduces doubt where there was previously confidence (even if it was misplaced), and it forces stakeholders to actually have to engage with uncertainty rather than move past it.
In some orgs that’s fine, but in environments where alignment and momentum are prioritised, that shift can feel like a step backwards. So instead, the uncertainty is compressed into a single number that everyone knows isn’t right, but is treated as if it’s precise enough.
Over time, that just becomes normal; not because it’s effective, but because it’s easier to manage.
Most teams don’t know what to model, so they either overdo it or avoid it
Even when there’s genuine intent to include scenario modeling, teams often struggle with where to start. The idea of “testing different scenarios” sounds straightforward, but can quickly become slightly ambiguous in practice.
Without a clear approach, one of two things usually happens. Some teams try to model too much, adjusting multiple variables across multiple scenarios until the model becomes difficult to maintain and even harder to explain. Others go in the opposite direction, simplifying to the point where the scenarios don’t meaningfully change the outcome, which makes the whole exercise largely pointless.
What both sides are missing is a focus on the variables that actually move the result. Us, gnarled modeling veterans, will be able to tell you that more often than not, only a small number of assumptions have an impact on ROI worth your time, but identifying those requires a level of discipline that isn’t always present in how business cases are constructed.
So scenario modeling either becomes overly complex or effectively meaningless, and in both cases, it unfortunately reinforces the idea that it’s just not worth the effort.
The way we build business cases doesn’t support it
There’s also a structural issue that’s harder to ignore; many business cases are still built in tools and formats that just weren’t designed for iteration, so trying to do so is unnecessarily painful.
When models live in static spreadsheets or slide decks, even small changes can have disproportionate effort attached to them. Adjusting an assumption often means tracing your way through multiple calculations (that will often seem logical to whoever wrote them, but will more often than not be accompanied by precisely no documentation or explanation, so are effectively unintelligible to anyone else), updating dependent figures, and revalidating outputs. Comparing scenarios becomes a manual process, and keeping everything aligned quickly turns into a version control problem.
When faced with that, scenario modeling isn’t just conceptually harder; it’s practically inefficient. It’s not difficult to understand why teams aren’t willing to explore multiple variations when each one feels like rebuilding part of the case from scratch.
This is where intent and execution diverge most clearly. Organisations may value scenario modeling in principle, but the tools and processes at their disposal make it difficult to apply consistently in reality.
What scenario modeling should look like in practice
For something that is often treated as complex, effective scenario modeling is actually usually quite focused and relatively simple. It doesn’t require dozens of variations or highly intricate models, as all it boils down to is whether it helps decision-makers understand how outcomes change under different conditions.
In practice, that typically means starting with a clear base case and then defining a small number of alternative scenarios that reflect meaningful shifts in key assumptions. Those scenarios should be grounded in reality, not extremes, and should be easy to explain without requiring deep technical context.
The key point that many people miss is that the goal isn’t to explore every possible outcome, but to make the uncertainty visible in a way that improves the quality of the decision. When done well, scenario modeling doesn’t overwhelm the discussion, it sharpens it.
The cost of treating it as optional
It’s easy to view scenario modeling as something that improves a business case, but isn’t strictly necessary. And for a while, you can probably even get away with that approach, but the consequences of that mindset tend to have a way of catching up with you later.
Decisions made on a single set of assumptions often carry more confidence than they should. When outcomes start to diverge, the underlying drivers aren’t always well understood, which makes it harder to either explain what changed or to respond effectively. Over time, this creates a disconnect between projected and realised value that can be difficult to trace back to its source.
More importantly, it limits the organisation’s ability to learn. Without a clear view of how different factors influence outcomes, each new business case is built with roughly the same level of uncertainty as the last.
Now, scenario modeling isn’t a panacea that entirely eliminates that uncertainty, but it does make it visible earlier, when it can still influence the decision.
Practical takeaways: how to make scenario modeling happen
ℹ️ - This is just a summary, but we also have a deeper dive into Scenario Modeling
If scenario modeling is consistently underused, the solution isn’t to make it more sophisticated; it’s to make it part of how business cases are built in the first place.
💡 Build it into the process, not as an add-on: Mandate at least one alternative scenario in every business case, so that assumptions are challenged early rather than revisited under pressure later
📊 Limit the scope: Focus on a small number of key drivers that materially affect outcomes, instead of trying to model every possible factor, but actually losing clarity in the process
🧮 Use tools that reduce friction: Make it easy to adjust variables and immediately see the impact, so testing scenarios feels like part of the workflow rather than rework
⚠️ Normalise uncertainty: Treat variability as something to explore and understand, rather than something to minimise or hide to present a cleaner narrative
💡 Assign clear ownership: Ensure someone is explicitly responsible for defining, maintaining, and explaining scenarios, so it doesn’t get diluted or skipped altogether
When these conditions are in place, scenario modeling becomes less of an additional task and more of a natural part of decision-making.
Conclusion
Scenario modeling isn’t underused because people don’t believe in it. It’s underused because, in practice, it introduces friction, complicates conversations, and challenges the way business cases are typically presented.
That doesn’t make it less valuable though, if anything, those are actually the reasons why it matters.
A single projected outcome may be easier to communicate, but it rarely reflects how decisions play out in reality. Scenario modeling shifts the focus away from defending one number and toward understanding how outcomes change when assumptions move, which is where most of the real risk sits.
That shift doesn’t simplify decisions, but it does make them more grounded. And over time, that tends to be the difference between business cases that look convincing at the point of approval and those that continue to hold up once they’re put into practice.
