Cell production yield rate, the share of usable defect-free cells a factory produces, is one of the strongest levers on the financial health of a battery plant. As electrification scales across mobility, grid storage and industry, manufacturers must grow output while holding quality and margin. This case study places the reader as the industrial director of a French startup preparing a 5 GWh cylindrical sodium-ion gigafactory in France by 2030, tasked with designing a line that delivers competitive yield from day one.
Why yield decides profitability
Small yield gains carry large financial weight. A 1 percent yield improvement at a 5 GWh factory can save tens of millions of dollars a year, and process optimisation with better yield can cut manufacturing cost by up to 20 percent. Because material costs often exceed 70 percent of total cell cost, every scrapped or reworked cell destroys disproportionate value; in figures cited during the study, each point of yield loss can mean roughly 30,000 euro per day, or around 10 million euro a year. The cautionary example is Northvolt, once seen as Europe's leading cell maker, which raised over 15 billion US dollars from backers including Goldman Sachs, Volkswagen and BMW yet struggled with low yields in critical processes, missed production targets, and filed for bankruptcy in early 2025. High yield also supports sustainability by cutting scrap, energy use and waste of constrained metals such as lithium, cobalt and nickel.
The critical manufacturing steps
Sodium-ion manufacturing mirrors lithium-ion closely: electrode preparation and slurry mixing, coating onto foil, drying, calendering, cutting, assembly with a separator, stacking or winding, electrolyte filling, sealing, and finally formation and aging. Most steps affect yield. A study of more than 220 experts identified coating, stacking and electrolyte filling as the most rejection-prone processes, and in a fifteen-step sequence only vacuum drying was judged not to cause rejection. If each of fourteen critical steps carried a 0.2 percent rejection rate, the rolled throughput yield would be about 97.2 percent, still below the 98.5 percent benchmark observed at Sunwoda's lithium-ion factory in China, well above the sub-95 percent typical of European and US lines. Slurry mixing suffers from uneven dispersion, agglomeration and air bubbles driven by impurities and poor mixing control. Coating defects such as uneven thickness, pinholes and delamination cause internal resistance variation, hotspots and short circuits. Poor calendering creates micro-cracks or over-compression, though automated control with thermal-expansion sensors and roller-pressure feedback can cut these errors by up to 80 percent versus manual operation.
Where the bottleneck moves as plants improve
An instructive finding is that the worst offender shifts as a line matures. In smaller-scale plants, coating is usually the top cause of rejection. In highly optimised facilities like Sunwoda, coating has been tamed and winding becomes the new bottleneck, responsible for an estimated 0.3 to 0.5 percent of rejection on its own through misalignment, particle contamination or separator wrinkling. Slitting and notching add their own risk, since burrs and edge tears can create microscopic short circuits and dust later in the line. Beyond the process steps, indirect factors matter. Consistent, impurity-free active material is a prerequisite, and Europe faces a shortage of skilled labour and limited awareness of battery manufacturing complexity relative to Asian competitors, both of which raise rejection rates. Two levers stand out for lifting yield: data-driven quality management across the line, and targeted optimisation of the formation step.
What it means for new gigafactories
For a startup building a line from scratch, the lesson is to design quality control in from the outset rather than chasing it later. Raw material quality and consistency, particularly with Asian sourcing, feed directly into cell yield, so supply-chain discipline is a yield strategy in itself. End-of-line testing improves outgoing quality but carries risk if used as a crutch, since it catches defects rather than preventing them and can mask upstream instability. The broader message is that yield is a system property. It depends on material consistency, process control, workforce skill and data infrastructure together, and the payoff, measured against Northvolt's collapse, is the difference between a viable plant and an unviable one.
Key Takeaways
A 1 percent yield gain at a 5 GWh factory can save tens of millions of dollars a year.
Material costs often exceed 70 percent of cell cost, so scrap and rework destroy disproportionate value.
Coating, stacking and electrolyte filling are the most rejection-prone steps; only vacuum drying was judged risk-free.
Sunwoda's 98.5 percent yield far exceeds the sub-95 percent typical of European and US lines.
As coating is optimised, winding becomes the new bottleneck, causing an estimated 0.3 to 0.5 percent of rejection.
Automated calendering control can cut defects by up to 80 percent compared with manual operation.
Data-driven quality management, formation optimisation and consistent raw materials are the main levers for higher yield.
Disclaimer: This case study was developed and presented by BatteryMBA participants as part of the Case Study Track. Views, analysis and recommendations are the authors' own. BatteryMBA does not take responsibility for the accuracy or completeness of the content and it should not be relied upon as investment, engineering or legal advice.
This is the public summary, the full case study lives inside the programme
Every BatteryMBA cohort runs the Case Study Track: small teams build the full recommendation, backed by a written document and a live presentation, supported by the BatteryMBA team. Full case study documents are not shared outside the programme. programme.