# Run Learning Cycles in SwiftCNS

**Purpose**: Execute the core SwiftCNS loop from assumption selection to validated learnings and decision-ready insights.

## Outcome

A complete cycle with clear evidence and explicit decision options.

This section is the operational heart of the playbook. If the earlier sections explain the logic of the system, this section is where teams actually put that logic to work.

## Time to complete

Typically 2-4 weeks for a full cycle, depending on experiment lead times.

## Prerequisites

* Completed [Start Here](https://playbook.swiftcns.ai/getting-started).
* Shared understanding from [Concept Foundations](https://playbook.swiftcns.ai/concept-foundations).
* One project and one high-priority problem area.

## Workflow order

1. [01 — Idea / Problem](https://playbook.swiftcns.ai/run-learning-cycles/01-idea-problem)
2. [02 — Identify Critical Assumptions](https://playbook.swiftcns.ai/run-learning-cycles/02-identify-critical-assumptions)
3. [03 — Form Testable Hypotheses](https://playbook.swiftcns.ai/run-learning-cycles/03-form-testable-hypotheses)
4. [04 — Design & Run Experiments](https://playbook.swiftcns.ai/run-learning-cycles/04-design-run-experiments)
5. [05 — Extract Key Learnings](https://playbook.swiftcns.ai/run-learning-cycles/05-extract-key-learnings)
6. [06 — Synthesize Insights -> Decision](https://gitlab.com/swifracks-solutions/ai/product-innovation-guide/-/blob/main/run-learning-cycles/06-synthesize-insights-decision.md)

## Why this flow is ordered this way

Each stage exists to solve a different problem:

* `01` gives the team a focused problem frame.
* `02` identifies what must be true for the bet to work.
* `03` converts those assumptions into something testable.
* `04` generates evidence.
* `05` turns evidence into usable learnings.
* `06` turns those learnings into action.

If a team tries to skip or compress these stages without enough discipline, the usual result is not speed. It is confusion that shows up later as weak learnings or hesitant decisions.

## Role lenses

The role split is simple: startup teams should optimize for speed while preserving evidence quality, program managers should enforce stage-gate readiness at each step, and mentors should improve clarity, rigor, and decision confidence.

## What good looks like across the full sequence

A strong cycle does not just produce artifacts. It produces movement in confidence.

By the end of the sequence, the team should be able to say:

* which assumption was tested,
* what evidence was generated,
* what learning became clearer,
* what insight matters most,
* and what decision follows from that.

That is the difference between a team that is exploring and a team that is actually learning.

## Definition of done

* Each stage produces required artifacts.
* Learnings are validated and decision relevant.
* Insights support an explicit decision path (go, iterate, pivot, stop).

If those three conditions are true, the cycle has done what it is supposed to do: reduce uncertainty enough to support a real next move.

## Common breakdown across the sequence

Most teams do not fail because they miss a step entirely. They fail because one weak stage quietly contaminates the rest:

* vague problem framing weakens assumptions,
* weak assumptions create fuzzy hypotheses,
* fuzzy hypotheses create weak experiments,
* weak experiments create noisy learnings,
* noisy learnings create indecisive synthesis.

That is why each page in this section is designed to help teams strengthen one stage before moving to the next.

## If blocked

Use [Troubleshooting and Recovery](https://playbook.swiftcns.ai/troubleshooting) to diagnose failure patterns and recover quickly.

## Next step

Start with [01 — Idea / Problem](https://playbook.swiftcns.ai/run-learning-cycles/01-idea-problem).


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://playbook.swiftcns.ai/run-learning-cycles.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.