How do I use T-Shirt Sizing to estimate the effort required to deliver an Information Product?
T-Shirt Sizing is a quick, relative estimation method used to gauge the effort required to deliver an Information Product
How do I use T-Shirt Sizing to estimate the effort required to deliver an Information Product?
T-Shirt Sizing is a quick, relative estimation method used to gauge the effort required to deliver an Information Product. Instead of assigning precise time estimates, it uses sizes, Small, Medium, Large, Extra Large (or even Oversized), to represent a guesstimate of complexity, effort, and timeframe. It gives stakeholders a shared understanding of delivery expectations without committing the data team into fixed timelines.
To determine the T-Shirt Size, the data team assesses several key factors:
Data Availability
Data availability has a major impact on the Information Prodsuct T-Shirt Size, if the required data is already collected and modeled, the Information Product can be delivered quicker. Sourcing and integrating new data will markedly increase effort and timeframes.
Business Logic Complexity
Business logic complexity is another factor, simple calculations and filters are quick to implement. Complex business rules, transformations, aggregations, or machine learning models require more effort and lead to larger sizes.
Data Standardisation
The need for data standardisation also impacts complexity. If there is a need to combine or harmonise data from multiple disparate data sources, effort and timeframes increase.
Delivery Type Complexity
Delivery Type complexity is considered as well, a simple dashboard is quicker to build than a multi-layered report, a visualisation with drill-throughs, or an API that integrates with external systems.
Data Sync Requirements
Finally, Data Sync rate affects effort and timelines, batch updates are generally simpler to implement, while near real-time processing increases the complexity markedly.
Once these factors are assessed, the data team assigns a relative size based on how long they believe it will take to deliver.
For example:
Small, 1 iteration
Medium, roughly 3 iterations
Large, 5 to 6 iterations
Extra Large, 12 or more iterations
Keeping a consistent ratio between T-Shirt sizes builds predictability and helps compare Information Products more easily. The data team determines the relative sizing ratios. Over time, their sizing guesstimates will improve, just like a good wine, they get better with age.
T-Shirt Sizing is always assigned by the data team. They have the technical expertise and context to assess the work realistically, sizes should never be assigned by stakeholders alone.
And most importantly, T-Shirt Sizing is a guesstimate, not a promise. It’s designed to guide prioritisation, not set delivery deadlines in stone.