Business cases are rarely built in ideal conditions, as in most organisations, teams are asked to justify significant investment decisions while key inputs are still shifting. Cost estimates are evolving, delivery timelines are not fully validated, and assumptions around adoption or benefit realisation are based on partial evidence at best.
Despite that, there’s still pressure to produce a clear, confident answer to the question: should we invest or not?
This is where things usually go wrong, as teams take two approaches, which are equally flawed, just in different ways. Some teams wait for that mythical state of “enough” data (that inevitably never actually fully arrives), while others push forward with overly precise forecasts that look convincing on paper but quietly rely on fragile assumptions.
The real challenge isn’t a lack of data; it’s how to build a decision-ready business case when the data is incomplete, inconsistent, or uncertain. In this blog, we’ll explore the fact that strong business cases don’t eliminate that problem; instead, they structure it.
Incomplete data is normal in serious investment decisions
Most meaningful business cases are built before the full picture is available, and this is especially true for transformation programmes, AI initiatives, operational redesigns, compliance investments, or new product development. These aren’t situations where historical data will cleanly predict future outcomes, as they involve change, and by definition, change introduces variability.
Trying to delay a business case until everything is known often creates a range of different problems; decisions get pushed back, opportunities are missed, and assumptions quietly harden without being properly tested. In practice, stronger organisations tend to work in layers:
📌 Early assumptions to frame direction
📊 Progressive refinement as more evidence emerges
🧭 Scenario-based thinking to understand a range of potential outcomes
The business case becomes less about finding a single correct answer and more about building a structured view of what could happen and why. This is a shift that matters because it changes the role of the business case from a static justification document into a decision tool.
The real issue is not assumptions, but hidden assumptions
It’s basically impossible for a business case not to contain assumptions, but the problems arise when those assumptions aren’t clearly expressed, challenged, or revisited after the fact.
In many cases, assumptions end up embedded inside spreadsheets as if they are facts, e.g., a productivity gain is entered as a single percentage, a cost saving is assumed as a fixed value, or a delivery timeline is treated as predictable rather than conditional.
Undeniably, it’s comforting to see that level of precision, but uncomfortably, that doesn’t actually reflect reality. A more credible approach is to make assumptions explicit and explain where they come from.
👉 e.g. Instead of “Productivity improves by 18%”, a far stronger framing would be “Productivity improvement is estimated between 10% and 18%, based on comparable automation initiatives, stakeholder input, and early pilot indicators”
Most business cases draw from a mix of sources:
📊 Historical internal performance
📚 Industry benchmarks
🧠 Expert judgement from delivery teams
🧪 Pilot or proof-of-concept results
🤝 Vendor input and estimates
However, none of these sources is perfect on its own, so the strength comes from combining them transparently rather than hiding them inside a single number.
Precision is often mistaken for credibility
It’s easy to assume that more precise forecasts create stronger business cases. So it’s counterintuitive to hear that in reality, the opposite is often true. This is because a highly precise ROI figure can create a false sense of certainty if the underlying assumptions aren’t equally robust. For example, a forecast that predicts exactly “147.3% ROI” looks confident, but it doesn’t necessarily help decision-makers understand risk.
What matters more is how the range of outcomes is structured and explained, so instead of a single-point forecast, stronger business cases often use ranges:
➖ Conservative outcome
🟰 Expected outcome
➕ Optimistic outcome
This changes the nature of the discussion, as the focus shifts from whether the number is correct to what drives variation between scenarios, and that’s where the meaningful analysis happens:
🧮 What assumptions drive the upside case?
⚠️ What conditions create downside risk?
📊 Which variables most influence ROI?
This is also where scenario modelling becomes genuinely useful, as it allows organisations to test how changes in adoption, cost, timing, or performance affect overall outcomes.
Confidence is more important than precision
One of the more common misunderstandings people have when developing a business case is the assumption that stakeholders expect certainty. In reality, most experienced decision-makers are comfortable with uncertainty, but what they need is clarity on where that uncertainty sits and how it affects the decision.
The difference between a weak and strong business case is often not the level of precision, but the transparency of confidence. A useful distinction is:
📏 Precision: how exact the numbers appear
🎯 Confidence: how reliable the underlying assumptions actually are
A business case can be highly precise yet still low confidence if it’s built on untested assumptions. Conversely, it can also be less precise but far more trustworthy if the assumptions are well understood and clearly communicated.
One practical way to reflect this is by assigning confidence levels to key inputs, e.g.
Area | Confidence |
Infrastructure costs | 🟢 High |
Delivery timeline | 🟡 Medium |
Adoption rate | 🟠 Medium-Low |
Productivity uplift | 🔴 Low |
It’s important to realise that this doesn’t reduce rigour, it improves it, because it makes uncertainty visible and therefore easier to manage.
Business cases should evolve, not freeze at approval
Another common issue is treating the business case as a one-time approval artefact. Once funding is secured, many organisations stop revisiting the assumptions that justified the decision in the first place, so the business case becomes a static document rather than a living model.
That creates a gap between projected value and realised value, which is often only discovered later during post-implementation reviews. Stronger organisations treat business cases as something that evolves over time:
🧭 Early stage: directional estimates
🗺️ Planning stage: refined cost and delivery assumptions
🧪 Pilot stage: validated behavioural inputs
🚚 Delivery stage: tracking actual vs expected performance
🔁 Post-implementation: learning and recalibration
This approach turns the business case into a feedback loop rather than a one-off justification. Over time, it improves forecasting quality because organisations learn which assumptions tend to hold and which consistently diverge from reality.
Practical ways to build stronger business cases with imperfect data
Even without perfect information, there are consistent ways to improve the quality of a business case:
✂️ Separate facts from assumptions: Distinguish clearly between what is known, what is estimated, and what is assumed. This immediately improves clarity in stakeholder discussions.
↔️ Use ranges where outcomes vary: If an input is uncertain, forcing a single number creates false precision. Ranges reflect reality more accurately and improve decision quality.
🔎 Focus on the variables that matter most: Most business cases are driven by a small number of key variables such as adoption, timing, and efficiency. These should get disproportionate attention.
📚 Make the source of assumptions visible: Benchmarks, pilots, expert input, and historical data all add context. Even imperfect estimates become more credible when their origin is clear.
🔁 Treat the business case as iterative: A business case should improve over time. As new information emerges, assumptions should be revisited rather than left static.
Conclusion
Perfect data is rarely available when investment decisions need to be made, and waiting for it usually delays progress without actually improving the quality of the decision.
Stronger business cases don't rely on certainty; instead, they make uncertainty explicit, structure it in a way that's easy to reason about, and focus attention on the assumptions that matter most.
When that happens, the conversation changes. Instead of debating whether a single forecast is right, stakeholders focus on understanding what drives value, where risk sits, and how outcomes might vary under different conditions.
At the end of the day, that’s what leads to better decisions in practice; not perfect data, but better structured thinking.
