0.6. Python vs Excel
Can you be swayed from the most popular analytical tool in Pricing?
Despite Excel being the default tool for everything analytics in Pricing, and often praised for being incredibly easy to use and versatile, there are plenty of probems that arise with Excel workflows.
Transparency and reproducibility are fundamental principles that ensure your analysis can be trusted, understood, and improved over time; Excel's strengths are not in these areas.
Python is quickly replacing a lot of Excel workflows in Insurance pricing, albeit very late compared to other industries. In teams where Python has become a staple, Excel use is a bottleneck.
It is still very hard to convince a Pricing Analyst that Excel is not that great for analytics, could you be swayed?
Challenges with Excel in building transparent analytics
Excel strength is for quick calculations and simple prototyping, but when it comes to building complex pricing workflows, it presents several challenges:
- Formulas are often hidden inside cells with limited documentation, making it difficult to trace how a particular result was obtained.
- The layout of data and calculations is spread across multiple sheets, creating a tangled structure that is hard to follow.
- Manual steps such as copying data, refreshing links, or adjusting parameters are common, increasing the risk of errors.
- Version control is limited, so tracking changes or collaborating safely can be cumbersome.
- Complex processes tend to become tightly coupled to individual spreadsheets, creating knowledge silos.
These factors can make Excel-based analytics fragile, opaque, and hard to reproduce consistently.
How Python supports transparency and reproducibility
Python, as a programming language, offers tools and practices that naturally promote these principles:
- Code explicitly documents each step of data preparation, transformation, and modeling.
- Modular functions and scripts enable reuse and clear separation of concerns.
- Integration with version control systems allows detailed tracking of changes and collaborative workflows.
- Automated testing frameworks help ensure the correctness of complex processes.
- Workflow automation reduces manual intervention, minimizing human errors.
- Rich libraries and tools facilitate scalable handling of large datasets and advanced analytics.
Using Python can make analytical workflows more robust, easier to maintain, and adaptable to changing business needs.
Excel still has a place
Excel remains useful for exploratory analysis, quick ad-hoc calculations, and communicating simple results to business users. Its accessibility and familiarity are strengths in certain contexts.
However, when analytics workflows start to become slightly more complex, require frequent updates or revisiting in the future, Excel’s limitations become more apparent. Processes can become brittle, error-prone, and hard to evolve.
Choosing Tools to Uphold Analytical Integrity
The key is not about choosing a specific tool, but about adopting workflows that prioritise transparency and reproducibility. Whether that involves Python, R, or other platforms, the focus should be on:
- Writing clear, well-documented code or formulas.
- Automating repetitive tasks.
- Using version control or other change tracking methods.
- Designing modular, testable processes.
- Ensuring outputs can be regenerated consistently.
Investing in these practices will pay dividends in the quality and longevity of your pricing analytics.
Conclusion
Excel’s ubiquity and ease of use make it a natural starting point, but it can fall short for scalable, transparent, and reproducible pricing analytics.
Python offers a compelling environment to build workflows that embody these principles, though the broader focus should be on adopting engineering discipline regardless of the tool.
Prioritising transparency and reproducibility enables teams to deliver high-quality analytics that can evolve with the business and withstand scrutiny.
Choosing the right tools is important - but committing to sound analytical practices is what truly drives success.