The Spiral Model

Basketly wants live rider tracking: a moving dot on a map showing where your groceries are. The screen is easy. The scary part is whether riders' cheap phones send their location reliably while riding through traffic. Building the whole feature and discovering this at the end would waste months. The Spiral model is built for exactly this kind of doubt.

The rule

Each loop of the spiral tackles the biggest remaining risk first.

The Spiral model grows software in loops. Each loop passes through four quarters:

1. Plan objectives: decide what this loop must achieve and what the options are.
2. Analyse risk: find what could sink the project, and reduce it, often with a quick prototype or experiment.
3. Build and test: develop and test this loop's piece.
4. Review and plan the next loop: stakeholders evaluate the result and decide whether and how to continue.

Each loop is larger and more complete than the last, which is why the drawing looks like a spiral widening outwards. The model suits large, expensive or uncertain projects. It is heavier than it sounds: every loop needs real risk analysis and a review, so it rarely fits a small, well-understood feature.

For testers, Spiral means testing happens in every loop, and early loops often test prototypes rather than finished features. Measuring whether a prototype meets its target is testing too.

The shortcut

At the start of each loop ask: "What could kill this project right now?" Whatever you answer is what this loop must test or prove.

The trap is running the loops in a comfortable order, starting with the easy screen. Spiral only works if each loop faces the most dangerous unknown.

Drills

Answer fast — trust the rule.

Remember

• Each loop of the spiral tackles the biggest remaining risk first.
• Ask "What could kill this project right now?" at every loop.
• Testing happens in every loop, including testing prototypes.