SVPG · Free post · Discovery & customer research · Execution, roadmaps & process

Continuous Discovery

Marty CaganOct 24, 20123 min
SourceSVPG
KindFree post
PublishedOct 24, 2012
Originalsvpg.com ↗
N:

Marty Cagan describes a growing trend where product teams move from batch-style phases to continuous practices in both delivery and discovery. He traces how continuous integration, testing, and deployment have spread, shrinking the time between change and feedback and making problems easier to locate. He then argues the same logic should apply to discovery, where teams continually ideate, validate, and describe new backlog items rather than running a separate multi-week discovery phase. Many teams find Scrum's fixed sprints limiting and drift toward Kanban-style flow, sometimes keeping time-boxes only as a retrospective rhythm. The piece matters because it reframes discovery as an ongoing habit feeding a parallel delivery track, while noting trade-offs left for later articles.

01Key takeaways

  • Integrate and test changes continuously rather than in a late phase to find problems faster and narrow their sources.
  • Ship small, incremental changes frequently so that production issues are easier to diagnose and roll back.
  • Treat product discovery as an ongoing activity of ideating, validating, and describing backlog items, not a one-off phase.
  • Consider whether fixed sprints fit your flow; many teams shift toward Kanban or keep time-boxes mainly for retrospectives.
  • Expect trade-offs when moving to continuous models and plan for their effects on release planning and go-to-market work.

02Key sections

Continuous delivery in practice
Teams have progressively moved from end-of-cycle testing to continuous builds, regression testing, and micro-releases. Small, frequent changes make issues faster to detect and fix.
Extending the principle to discovery
Cagan argues product backlog items should be generated continuously through ideation, validation, and description rather than in a discrete discovery phase.
The dual-track model
A discovery track constantly feeds validated items to a delivery track that builds, tests, and deploys them, so the two run in parallel.
Implications for process
Scrum's time-boxed sprints can constrain continuous flow, pushing many teams toward Kanban or Kanban-like adaptations of Scrum.
Trade-offs ahead
Continuous practices are generally beneficial but carry costs, which the author promises to explore in future articles on release planning and marketing.

03From the post

“A partnership dedicated to teaching best practices to product teams and product leaders”

“Rather than a “Product Discovery Phase” where we come up with several weeks of validated product backlog items”Marty Cagan · SVPG
“In the dual-track analogy, the discovery track is continuously generating product backlog items, and the delivery track is continuously building, testing and deploying these items.”Marty Cagan · SVPG
“detecting and correcting issues in production is much easier when you are just changing one thing, or a small number of things, at a time.”Marty Cagan · SVPG

04Frameworks mentioned

Summary and takeaways written by PM Atlas; quotes are short excerpts. © the original author.