Manu's Vault

Small-Batch-Production

4 min readStatus:Seedling

What It Is

Small-batch production is the practice of processing work in small, complete increments rather than accumulating large quantities before moving to the next step. It is one of the most counterintuitive — and empirically powerful — ideas in lean manufacturing, applied directly to startup product development by Eric Ries.

The Envelope Example

Ries illustrates the principle through envelope stuffing. Imagine you must stuff 100 envelopes: fold a letter, insert it, seal, stamp. The “efficient” large-batch approach says: fold all 100, then insert all 100, then seal all 100, then stamp all 100. This feels faster because each repetitive motion is batched.

But it’s slower — and far riskier. The small-batch approach (fold → insert → seal → stamp one envelope, then repeat) produces the first completed envelope almost immediately. More importantly, if the envelope is the wrong size and the letter won’t fit, you discover this on the first unit, not after 100 letters have been folded the wrong way.

Small batches surface defects immediately. Large batches bury defects until the end, after maximum investment.

Manufacturing Origins

Toyota discovered this in the postwar period. American mass-production factories had long changeover times between production runs, making large batches economical. Toyota, constrained by smaller markets and tighter capital, invested in reducing changeover time rather than accepting it as fixed. Shigeo Shingo formalized this as SMED (Single-Minute Exchange of Die) — the practice of reducing machine changeover times from hours to minutes, making small batches economically viable and ultimately superior.

Womack and Jones documented this in Lean Thinking (1996): reducing batch sizes and the WIP (work-in-process) inventory between steps is a primary source of lean’s speed and quality advantages. Less in-process inventory means fewer hidden defects, shorter cycle times, and faster feedback when something goes wrong.

Application to Startups

In product development, large batches mean: build many features, test everything at once, deploy in big quarterly releases. Small batches mean: release one feature, test one hypothesis per cycle, deploy continuously.

The startup connection to Build-Measure-Learn-Loop is direct — small batches shorten the feedback loop. You learn what customers actually want sooner, and you fail smaller. Large batches extend the time between building and learning; small batches compress it.

Future Connections

Will connect to Large-Batch-Death-Spiral, Continuous-Deployment, Hypothesis-Pull, Adaptive-Organization, Five-Whys when created.

Sources

  • Ries, Eric (2011). The Lean Startup: How Today’s Entrepreneurs Use Continuous Innovation to Create Radically Successful Businesses. Crown Publishing. ISBN: 978-0-307-88791-7.

    • Chapter 9 (Batch) — primary treatment; envelope example, Toyota origins, IMVU application.

  • Shingo, Shigeo (1985). A Revolution in Manufacturing: The SMED System. Productivity Press. ISBN: 978-0-915299-03-8.

    • Foundational text on Single-Minute Exchange of Die; the original technical argument for why reducing changeover time makes small batches viable and superior to large batches.

  • Womack, James P. and Daniel T. Jones (1996). Lean Thinking: Banish Waste and Create Wealth in Your Corporation. Simon & Schuster. ISBN: 978-0-684-81035-8.

    • Documents Toyota’s small-batch principles across industries; explains how WIP reduction drives cycle time compression and defect visibility.

  • Humble, Jez and David Farley (2010). Continuous Delivery: Reliable Software Releases through Build, Test, and Deployment Automation. Addison-Wesley. ISBN: 978-0-321-60191-9.

    • Software engineering application of small-batch thinking: the argument for continuous delivery over large releases; batch size as a primary driver of software delivery performance.

  • Forsgren, Nicole, Jez Humble, and Gene Kim (2018). Accelerate: The Science of Lean Software and DevOps. IT Revolution Press. ISBN: 978-1-942788-33-1.

    • DORA research empirically validates that high-performing software teams use smaller batch sizes, shorter integration windows, and more frequent deployments — confirming the manufacturing insight at scale across thousands of organizations.

Note

This content was drafted with assistance from AI tools for research, organization, and initial content generation. All final content has been reviewed, fact-checked, and edited by the author to ensure accuracy and alignment with the author’s intentions and perspective.

Graph View

Backlinks