Segall Bryant Lifts Credo Stake 12.7% in Q1

Credo Technology Powers the AI Backbone

spiderwebs of wiring inside a data center. Data centers rely on them for connections within the same rack or row, and the demand for these cables is skyrocketing as hyperscalers like Google, Amazon, and Microsoft build new facilities constantly. Each new facility requires hundreds of thousands of these connections, creating a repeatable, high-volume revenue stream for Credo.

Credo has positioned itself as a primary supplier by iterating faster than legacy competitors. Their technology also extends to optical modules, which are used for longer distances across the data center campus, connecting different rows or buildings. The rise of Generative AI has fundamentally changed the game for connectivity. Training a model like GPT-4 or its successors requires thousands of GPUs working in perfect unison. They must communicate constantly, exchanging terabytes of data per second. If the connection lags or introduces errors, the training slows down, costing companies millions in compute time. Credo's solutions reduce latency and increase the bandwidth available to these massive clusters, making the entire system more efficient. Industry experts point out that the AI boom is not just about compute power; it is about the entire data pipeline. Credo owns a crucial piece of that pipeline. Their Digital Signal Processors (DSPs) are industry-leading, consuming very little power relative to their performance. This efficiency is a major selling point in an era where every watt of power is scrutinized.

The Energy Crisis in Modern Data Centers

Power consumption has emerged as the single largest challenge facing data centers today. Operators are desperate to reduce energy usage not only to save costs but to adhere to increasingly stringent environmental regulations. Electricity costs are rising globally, and in many regions, the availability of power is becoming a hard constraint on expansion. Credo Technology addresses this pain point directly through architectural innovation. Their chips are designed for low power consumption, a stark contrast to traditional solutions that can be energy hogs. Legacy interconnects generate immense amounts of heat, which in turn requires expensive, power-hungry cooling systems to maintain operational temperatures. Credo's approach minimizes both power draw and heat generation at the source, creating a cascading effect of savings.

A data center can save millions on electricity bills annually simply by switching to more efficient connectivity components. Analysts estimate that connectivity can account for a significant portion of a data center's power budget, often overlooked compared to the power draw of CPUs and GPUs. Reducing this overhead is now a top priority for Chief Technology Officers. The company's AECs are a prime example of this efficiency in action. They replace heavy, power-hungry copper cables (DACs) that struggle with heat over long runs, and they replace expensive optical transceivers for short reaches. Optical transceivers consume a lot of power per port due to the lasers and digital signal processing required. AECs offer a middle ground, providing the speed and reach of fiber with the cost and efficiency of copper.

This "sweet spot" is driving rapid adoption as the industry transitions toward 800 Gigabit Ethernet, with the next step being 1.6 Terabit Ethernet. These speeds require incredible precision and signal integrity; they also require power efficiency to be commercially viable. Legacy technology struggles to maintain signal integrity at these speeds without burning excessive power. Credo's newer IP was built specifically for this era of high-speed, low-power data transfer. Officials in the semiconductor industry have noted a distinct pivot in the market where performance is no longer the only metric; performance per watt is the new standard. Credo scores highly on this metric, making their technology indispensable for green computing initiatives.

Furthermore, Credo's IP licensing business contributes significantly to this narrative. They license their high-speed SerDes (Serializer/Deserializer) technology to other chipmakers who cannot afford to develop it in-house. This spreads their low-power philosophy across the industry, creating a recurring revenue stream with high margins. It also validates their engineering prowess; if other firms are willing to pay royalties to use Credo's designs, those designs must be best-in-class. This licensing model provides stability and hedges against the volatility of cyclical hardware sales. Segall Bryant likely factored this business model diversity into their decision. A company that relies solely on hardware sales is riskier, subject to inventory gluts and supply chain shocks. Credo has a dual engine of growth: they sell chips, and they sell the know-how to make chips. This resilience is attractive to long-term investors seeking weatherproof assets in the tech sector.

Decoding the 12.7% Strategy Shift

A 12.7% increase in holdings during a single quarter is a specific and deliberate action that speaks volumes about an investor's thesis. It is not a rounding error or a passive index adjustment. Segall Bryant & Hamill operates with strict investment theses, often employing quantitative models to identify undervalued companies with strong fundamentals and upward momentum. They do not typically buy on hype or short-term trends. They look for mispriced assets where the market has not yet fully priced in future growth. Credo fits this profile in the eyes of many value-oriented investors who see the current valuation as a discount to the company's potential in the AI infrastructure market.

The stock has seen fluctuations over the past year, mirroring the volatility of the broader semiconductor cycle, which is notoriously boom and bust. However, the structural demand for connectivity is different from consumer electronics cycles. It is driven by the build-out of cloud infrastructure and AI, which is a multi-year, secular trend unlikely to stop soon. By increasing their stake now, Segall Bryant is effectively buying into this trend, likely believing that the market has undervalued Credo's growth potential relative to its peers. The firm likely sees a gap between the current stock price and the company's intrinsic value based on future earnings power. This gap often closes as the company executes its business plan and revenues accelerate.

Regulatory filings provide the only window into these moves for the public. The report filed offers a snapshot of the past, showing what the firm held at the end of the first quarter. Market experts analyze these filings to predict future trends, looking for patterns across different firms. If one "smart money" investor buys, it is interesting; if several buy, it is a trend. In this case, Segall Bryant is standing out by increasing exposure while others might be trimming tech risk due to macroeconomic fears. This shows conviction. It suggests they have done deep due diligence, analyzing the order books, speaking with management, and vetting the technology against competitors. The increase effectively lowers their average cost basis or concentrates their capital in a winner, signaling to the market that a professional steward of capital sees a bright future. Retail investors often follow these cues, and the disclosure of such a move can create positive feedback loops. However, the core logic remains the business fundamentals: Segall Bryant is betting that Credo's technology and market position will yield outsized returns.

The Competitive Landscape and Technological Moat

To understand the significance of Segall Bryant's investment, one must look at the competitive landscape within which Credo operates. The data center interconnect market is crowded with giants, yet Credo has carved out a defensible niche. While large conglomerates like Broadcom and Marvell offer broad portfolio solutions, Credo's focused specialization in high-speed, low-power SerDes and DSP technology gives it a technological edge. This expertise is difficult to replicate; it requires years of R&D and specialized analog engineering talent. As data rates push past 200 Gigabits per second per lane toward 224G and beyond, the physics of signal transmission becomes brutally difficult. Credo's "moat" is its intellectual property portfolio, which allows it to solve signal integrity problems that generic solutions cannot.

Moreover, the industry is undergoing a transition in how data centers are architected. The shift towards "disaggregated" computing—where memory, storage, and compute are separated but connected by ultra-fast networks—favors high-speed connectivity specialists. Credo's AECs are becoming the standard for rack-to-rack connections in these modern architectures because they balance performance, cost, and power better than any alternative. The company is also actively involved in standard-setting bodies like the OIF (Optical Internetworking Forum) and IEEE, helping to define the future of Ethernet standards. By being at the table when these standards are written, Credo ensures its technology is aligned with the roadmap of the world's largest hyperscalers.

This positioning reduces the risk of technological obsolescence. While competitors try to catch up, Credo is already developing solutions for the next generation of speeds. The licensing revenue further strengthens this moat, as it integrates Credo's technology into the supply chains of other companies, effectively locking them into an ecosystem that relies on Credo's continued innovation. For an investor like Segall Bryant, this creates a scenario where the company has a clear path to market share gains without engaging in a race to the bottom on pricing. The value proposition is too high, and the switching costs for customers are too significant, for Credo to be easily displaced.

Risks, Challenges, and Future Outlook

Despite the strong bullish case presented by the technology and institutional backing, a thorough analysis must also account for the risks facing Credo Technology. The semiconductor industry is subject to cyclical downturns, and any slowdown in capital expenditure by hyperscalers could impact revenue growth. If the AI boom fails to meet the lofty expectations currently priced into the market, orders for high-speed connectivity could be delayed or canceled. Furthermore, the supply chain remains a potential vulnerability; reliance on third-party foundries like TSMC for manufacturing means Credo is exposed to global semiconductor capacity constraints and geopolitical tensions that could disrupt production.

Another significant risk is customer concentration. A large portion of Credo's revenue often comes from a small number of major customers. If one of these hyperscalers decides to develop its own proprietary connectivity solutions in-house—a trend known as vertical integration—Credo could lose a major account. Tech giants have the resources to design their own chips, and while Credo's IP is currently superior, the competitive landscape is always shifting. Additionally, valuation risk remains a concern; if the stock has run up significantly prior to Segall Bryant's purchase, any disappointment in earnings could lead to a sharp correction.

Looking ahead, the future for Credo hinges on the successful ramp of 800G and 1.6T product lines. If they can execute on production and capture the anticipated market share, the financial results should be robust. The next 12-24 months will be critical. Investors will be watching gross margins closely to see if the company can maintain profitability as they scale. Segall Bryant's increased stake suggests they believe the execution risk is manageable. The trajectory points toward a world where data movement is as valuable as data processing. As long as the digital economy expands, the roads built by Credo will see toll-like traffic, providing a steady stream of revenue. For the market, the question is no longer if high-speed connectivity is needed, but who will dominate the standard. Segall Bryant has placed its bet on Credo.