Guide

What is Optimism Bias in a Business Case? How to spot and reduce it

What is Optimism Bias in a Business Case? How to spot and reduce it

Chris Goodwin

7

min read

Chris Goodwin

7

min read

By necessity, business cases tend to be built before anyone knows exactly how the proposed investment will play out. That means that costs are estimated, delivery plans are provisional, adoption is uncertain, and benefits may depend on several things going right at once.


That uncertainty is unavoidable. The problems begin when it quietly disappears from the case.


A six-month implementation estimate becomes “six months”; a benefit that depends on strong adoption is treated as though the adoption will simply happen; and integration work is assumed to be straightforward because nobody has yet discovered why it won’t be. Eventually, an optimistic set of assumptions starts to be treated as an objective forecast.


This is optimism bias, and it can make an investment appear safer, cheaper, faster, and more valuable than the evidence really supports.

What does optimism bias mean in a business case?

In a business case, optimism bias is the systematic tendency to underestimate costs, duration, implementation difficulty, and risk while overestimating benefits and the likelihood of success.


An important thing to realize is that it doesn’t require anyone to be dishonest. Proposal owners often know the opportunity in far more detail than they know the problems that may emerge during delivery, and they may naturally become attached to an idea they’ve spent weeks developing.


Organizational incentives can reinforce this, as teams are generally rewarded for securing budget and creating momentum. Early estimates may be produced before enough evidence exists, then repeated in presentations and plans until they become difficult to challenge. Novel projects make the problem harder because there may be little internal history to compare against.


Once an early number becomes familiar, it can also become an anchor. A project first described as costing $500,000 may continue to be framed around that figure even after new information suggests a wider or higher range would be more realistic.


The result isn’t necessarily a bad business case; it’s often one in which uncertainty has been compressed too early.

Where optimism bias usually appears

Optimism bias rarely announces itself. It tends to sit inside assumptions that appear perfectly ordinary when read one at a time. Common examples include:


📅 Implementation timelines: assuming limited delay despite uncertain dependencies, procurement, testing, or data migration.


👥 Adoption rates: forecasting rapid uptake without allowing for training, resistance, or competing priorities.


📈 Benefit ramping: treating the full annual benefit as though it begins soon after launch instead of building over time.


🧑‍💻 Resource requirements: underestimating internal effort because attention is focused on external spend.


🔌 Integration complexity: assuming systems, data, and processes will connect more cleanly than experience suggests.


🔄 Change management: budgeting for the technology while overlooking the work required to change habits, processes, and ownership.


💵 Revenue forecasts: using an attractive commercial outcome without showing how sensitive it is to demand, pricing, or conversion.


⚠️ Risk: recording risks while leaving the core financial forecast largely untouched by them.


It’s also worth noting that these effects can compound. A project that launches three months late, reaches only 60% of expected adoption, and takes longer to ramp benefits may miss its value forecast by far more than any single assumption suggests.


That’s why decision-makers need to understand not only individual assumptions, but how several optimistic assumptions interact.

Optimism bias isn’t the same as dishonesty

A proposal owner who knowingly inflates benefits or suppresses likely costs presents an integrity and governance problem. However, optimism bias is different because the distortion can happen without conscious intent.


That distinction matters because it means that stronger policing alone won’t solve it. If every optimistic forecast is treated as evidence of bad faith, proposal owners are likely to become defensive rather than more accurate. The more useful response is to improve the process around the decision.


A reviewer might ask what evidence supports a 75% adoption assumption, what adoption comparable projects achieved, and what the economics look like at 50% adoption. Those questions challenge the forecast without requiring anyone to prove that the original estimate was “wrong.”


This is also where precision theater becomes relevant; a forecast can contain detailed spreadsheets, decimal places, and carefully calculated ROI while still depending on weak assumptions. Try as we might, unfortunately, mathematical precision can’t compensate for uncertain inputs.

How to reduce optimism bias in a business case

As much as we’d like it to, no process can completely remove uncertainty from an investment decision, and trying to do so creates a different problem. A useful business case should make uncertainty visible enough for people to understand what they’re approving. Several practices help:


🔍 Use independent challenge: A fresh pair of eyes can spot assumptions that have hardened into ‘facts’ without anyone really noticing. This doesn’t always require a formal review board; for smaller decisions, an informed reviewer outside the proposal team may well be enough.


🧾 Require evidence for important assumptions: Focus effort on the assumptions that could materially change the decision. Evidence might include supplier quotes, pilot results, usage data, customer research, or historical delivery performance. Where evidence is weak, say so.


🕰️ Compare with what actually happened before: If similar implementations normally take nine to twelve months, a six-month forecast deserves closer examination. Reference-class thinking makes this more deliberate by asking, “What typically happens to projects like this?” rather than relying only on the team’s view of this project.


📊 Use ranges, scenarios, and sensitivity analysis: Single-point estimates are convenient, but they can create false certainty. Model a sensible base case and downside scenario, then test the assumptions that are doing the heavy lifting. If the investment only works when adoption exceeds 80%, decision-makers should know that before approval.


📌 Make assumptions explicit: An assumption buried in a spreadsheet is difficult to challenge. Once stated clearly, it can be supported with evidence, discussed, monitored, and revisited. In KangaROI, for example, kAI can help surface missing evidence or assumptions and flag areas that need further validation, while preserving uncertainty where confidence is weak. The purpose isn’t to have AI “correct” the forecast, but to make the reasoning easier to inspect.


🔄 Learn after approval: Compare actual delivery dates, costs, adoption, and realized benefits with the original forecast. If integrations repeatedly take longer than expected, or benefits ramp more slowly, that becomes useful evidence for future decisions instead of a lesson that disappears into project history.


Together, these practices shift the discussion from “Is this forecast correct?” to the more useful questions: “What does it depend on? How strong is the evidence? What happens if our assumptions are wrong?”

FROM FORECAST TO DECISION

01

Optimistic estimate

One version of the future

Early assumptions become a single forecast for cost, timing, adoption, and benefits.

02

Expose uncertainty

Challenge what the forecast depends on

Add evidence, historical comparisons, ranges, scenarios, and sensitivity analysis.

03

Better-informed decision

Understand what could change

Decision-makers can see what is known, what is assumed, and how the outcome could move.

The aim isn't to remove uncertainty. It is to stop uncertainty being hidden inside apparently precise forecasts.

Practical takeaway: make uncertainty visible before approval

The goal of tackling optimism bias isn’t to make every business case pessimistic, nor to delay decisions until uncertainty disappears; realistically, that point may never come.


A better approach is to expose the important uncertainties before approval. Challenge assumptions, ask for evidence where it matters, compare estimates with relevant history, model sensible ranges and downside cases, and track what actually happens afterward.


A forecast can still be optimistic; it simply needs to show why that optimism is justified and what happens if reality is less accommodating.


This creates a stronger basis for approval because the decision is no longer resting on one polished version of the future. Decision-makers can see what’s known, what’s assumed, and which variables could change the outcome.

Conclusion

Optimism bias is difficult to eliminate because it often comes from reasonable people making estimates under uncertainty, becoming attached to a proposal, working with limited evidence, and operating in organizations that naturally prefer momentum to hesitation.


The answer isn’t to stop making decisions until forecasts are perfect; it’s to use decision processes that stop optimistic assumptions from quietly becoming “the numbers.”


When assumptions are explicit, evidence is visible, scenarios can be compared, and actual outcomes feed back into future decisions, business cases become more useful. They stop pretending there’s one precise future and instead help people understand the range of futures they may be approving.

Chris Goodwin

Chris Goodwin

Guest Writer

Drawing on a background in Economics and more than 2 decades of experience of building pricing models and pricing teams across the world, Chris brings deep expertise across a diverse range of industries.

Chris Goodwin

Chris Goodwin

Guest Writer

Drawing on a background in Economics and more than 2 decades of experience of building pricing models and pricing teams across the world, Chris brings deep expertise across a diverse range of industries.

Chris Goodwin

Chris Goodwin

Guest Writer

Drawing on a background in Economics and more than 2 decades of experience of building pricing models and pricing teams across the world, Chris brings deep expertise across a diverse range of industries.

RELATED INSIGHTS
RELATED INSIGHTS

Continue exploring

Continue exploring