Five Pitfalls to Avoid in Analytical Projects

Many projects underdeliver not because the analysis was flawed, but because it wasn’t applied in the right way or aimed at the right goal. Whether you’re modelling a business process or exploring new ways to present insight, being aware of common stumbling blocks can help ensure your work is not only technically sound, but practically useful. Here are five pitfalls to watch for, and how OR thinking can help you avoid them.

1. Solving the Wrong Problem

It’s easy to rush into modelling or coding before taking a step back to understand what’s really needed. Stakeholder demands, tight timelines, or the appeal of a particular method can all push you towards premature solutions. But even the most polished model won’t help if it’s answering the wrong question.

Start by exploring the problem space properly. Talk to the people involved. Use OR tools like Soft Systems Methodology or causal mapping to uncover what's actually going on. Often, the real issue is quite different from the one you were first presented with, and it’s worth the effort to find that out early.

2. Overcomplicating the Model

When you're trying to demonstrate value, it can be tempting to build a highly complex model. But sophistication doesn’t always add clarity. A model that's difficult to explain or understand can slow down decision-making or be ignored entirely.

Instead, focus on creating something clear and functional. Begin with a straightforward version that captures the main dynamics. Only add layers of complexity if they lead to better insight or more reliable conclusions. A model that’s simple and used beats one that’s clever and overlooked.

3. Neglecting Stakeholder Engagement

Without regular input from stakeholders, your work risks drifting out of alignment with real-world needs. You might miss crucial context or make assumptions that don’t hold up. Don’t treat stakeholders as a formality. Involve them meaningfully, whether that’s through informal conversations, collaborative workshops, or shared testing. You’ll uncover richer insights and build stronger support for your recommendations.

4. Ignoring Data Limitations

You won’t always get the data you hoped for. It may be inconsistent, incomplete, or not collected with your analysis in mind. Waiting for perfect data can stall progress, but charging ahead without recognising the gaps can be just as risky.

Good analysis is about working smart with what you have. That might mean combining data with expert opinion, making assumptions transparent, or testing how your results respond to uncertainty. These are familiar challenges in OR, and the discipline offers well-tested ways to deal with them.

5. Failing to Plan for Implementation

It’s not enough for your findings to be accurate, they also need to be usable. If your results don’t fit how decisions are made or can’t be easily applied, they may never leave the slide deck.

Think ahead. Ask who will be using your work and in what setting. Adapt your outputs accordingly. Whether that means simplifying a dashboard, creating a quick-reference tool, or identifying where follow-up support is needed, planning for action increases your chances of seeing your analysis make a real-world difference.

Successful analysis isn't just about numbers or algorithms. It's about clarity, relevance, and delivery. By drawing on OR’s systems thinking and paying attention to these common pitfalls, you’ll improve not just your models, but the change they help bring about.

If you would like to increase your skill set to avoid these pitfalls the courses below may be of interest:

Introduction to Soft Systems Methodology

Problem Structuring Methods for Strategic and Complex Problems