Silicon Labs’ Self-Service Journey in the Semiconductor Industry

The semiconductor industry has always operated at the intersection of complexity and precision. Products move through deeply interconnected manufacturing networks, production capacity shifts constantly, and a single disruption can ripple across multiple suppliers, geographies, and customers.

For Silicon Labs, that complexity is magnified by its business model.

“We focus 100% in IoT,” explained Operations Analytics Manager, Arun Kumaran. “And ever since then we have been a world leader in wireless connectivity.”

As a fabless semiconductor company, Silicon Labs does not own manufacturing facilities. Instead, it focuses entirely on IC design, R&D, and product innovation, while outsourcing production to specialized manufacturing partners across the semiconductor ecosystem.

That model offers flexibility and scalability, but it also creates a planning challenge unlike almost any other industry.

The company must coordinate highly dynamic supply networks involving multiple suppliers, manufacturing stages, capacities, and constraints, all while maintaining resilience and speed in a volatile market.

To manage that complexity, Silicon Labs built an increasingly sophisticated planning environment with o9. More recently, the company evolved that model further through a self-service approach that shifted ownership, integration management, and operational flexibility closer to the business itself.

Managing complexity in a fabless semiconductor model

Kumaran began by outlining the unique structure of the semiconductor industry.

Traditional integrated device manufacturers, or IDMs, design and manufacture their own products. Fabless companies like Silicon Labs take a different approach, outsourcing production to specialized “pure play” manufacturers while concentrating internally on product design and innovation.

That flexibility comes with trade-offs.

“When you don't own your own manufacturing facilities, how do you know how much supply you have to meet the demand?” Kumaran asked.

Unlike simpler manufacturing environments, semiconductor products move through multiple stages such as wafer fabrication, probe, assembly, and testing. Different suppliers often specialize in different stages, meaning a single product can pass through several companies and locations before completion.

On top of that, resilience requires multiple qualified suppliers across stages.

“We have alternate suppliers for each of the processes,” Kumaran explained. “This will ensure that if a problem does happen at one particular supplier, we still have alternate suppliers who can produce the product for us.”

The result is a highly interconnected network with thousands of SKUs, dynamic capacities, varying cycle times, and constantly changing supplier allocations. “And yes,” Kumaran said, “that is our challenge number one, a complicated supply network.”

Why spreadsheets stopped scaling

Like many organizations, Silicon Labs initially relied on Excel-based supply planning models.

The problem was not simply the volume of data. It was the variability.

Different planners built different assumptions. Capacity and cycle time information changed constantly. Data formats varied. Comparing versions became difficult. Aligning teams became even harder.

“You can already imagine that everybody will have their own formats and assumptions,” Kumaran said.

At scale, that approach became unsustainable.

The company needed a system that could automatically establish supply networks, synchronize assumptions, and generate plans from a common foundation.

Silicon Labs first implemented o9 demand planning in 2018, followed by supply planning in 2019, just ahead of the COVID-era semiconductor supply crisis. Over time, the company expanded into allocation planning, supply plan generation, and eventually a self-service operating model introduced in 2023.

Today, integrations between o9 and Silicon Labs’ existing systems automatically load supply network data, capacities, cycle times, inventory, and demand information directly into the platform.

“This means that our users no longer need to set this up themselves,” Kumaran explained. The company now operates from what he repeatedly described as “a single source of truth.”

Turning supply planning into a strategic capability

One of the biggest advantages of the new approach is consistency.

The same logic governs supply planning across all SKUs. Capacity assumptions are aligned centrally. Data is integrated automatically. Scenario planning can run continuously against updated demand and supply conditions.

“We take a monthly locked supply plan version, and we use it for medium- to long-term supply chain decision-making,” Kumaran said.

That includes supplier forecasting, sourcing strategies, and operational planning.

At the same time, Silicon Labs can generate daily planning versions to reflect the latest changes in supply and demand.

The company also uses o9 extensively for scenario planning.

“We have stressed our supply chain by increasing the demand quantity to be more than 20% up and identified the potential sources of bottlenecks,” Kumaran explained.

The ability to model disruptions, sourcing alternatives, and supply constraints has become critical for maintaining preparedness in an industry where volatility is constant.

“In conclusion, with o9, we have been able to respond quickly to dynamic changes in our supply chain through an integrated and automated supply plan,” he said.

Why Silicon Labs moved toward self-service

As Silicon Labs’ products and supply chains became more complex, the company recognized another challenge: speed of enhancement and operational ownership.

“One of the main reasons why we actually moved into a self-service model was… the semiconductor industry always tends to change,” explained Business Process Analyst Daniel Hendri.

The company wanted greater control over integrations, data quality, and enhancement cycles.

Previously, integrations existed across multiple layers and systems. Debugging became difficult because ownership was unclear. Teams often spent significant time trying to determine whether an issue originated within o9 or on the client side.

“There were always these kind of discussions which actually delayed the debugging time,” Hendri said.

The self-service approach centralized transformation logic on the client side, established clearer ownership, and ensured cleaner upstream data before information reached o9. That shift produced several operational benefits.

Faster runtimes, cleaner data, and fewer delays

The impact of the self-service model was measurable almost immediately.

“Our demand planning batch and supply planning batch runtime has actually improved by more than 15%,” Hendri said.

Because transformations were reduced and integrations simplified, planners received outputs more quickly and could make decisions earlier.

Debugging time also improved significantly.

“Previously, there were a lot of issues which actually even took us like 2 to 3 days,” Hendri explained.

By embedding data quality checks upstream and introducing alerts before bad data entered o9, the organization reduced many of the manual troubleshooting steps that had previously slowed planning cycles.

The self-service approach also reduced dependency on external enhancement cycles. Smaller changes could be implemented faster, with business users and IT teams collaborating directly through clearly defined governance processes.

The structure itself became a major part of the success. “It’s actually a shared ownership model,” Hendri said.

Business teams define requirements, priorities, and success criteria. IT and solution teams manage integration reliability, architecture, governance, and data quality. “That’s precisely the reason why our self-service model works wonders,” he said.

Building a more scalable operating model

For Silicon Labs, the long-term goal is not simply faster planning. It is a more scalable and resilient operating model.

“We are actually building a system towards a faster, cleaner, more stable, more scalable and a more secured way,” Hendri said.

That means building reusable logic, enforcing clear ownership, maintaining transparent integrations, and ensuring that clean, reliable data drives decision-making throughout the organization.

Kumaran’s advice to other companies reflects the lessons Silicon Labs learned through the process.

“Start implementation by actually picking a good operating reference model,” he said. “Design it on a scalable basis and enable self-service.”

For Silicon Labs, the shift to self-service planning was not only a technology decision. It was an organizational decision about ownership, governance, and agility in one of the world’s most complex supply chain environments.

And in an industry where volatility is constant, that flexibility has become a competitive advantage.

Watch the full aim10x Digital 2026 video below.

Watch

Live Demo with an o9 Expert

Get a personalized walkthrough with an o9 solution specialist and see how to drive faster, smarter planning decisions across your enterprise.

request demo