Revival, Continuous!

The dawn of the Internet of Things (IoT) has transformed the way we live, leading to a profound integration of technology in our daily livesBe it smart home devices that automate our homes, secure access control systems for businesses, or payment systems that allow for quick transactions through scanned codes, IoT has become ubiquitousAt its core, the effectiveness of these systems hinges on one crucial component—communication modules that facilitate data transferThis category of devices is essential for the interconnectivity of IoT technologies.

In fact, the demand and supply dynamics of communication modules have undergone exciting changesNot long ago, in 2021, the growth trajectory of these modules soared, with a staggering 35% increase in shipmentsHowever, the landscape soon changed; by 2023, the global shipment of IoT modules witnessed its first decline, a telling indicator of market maturation and saturation in certain niches.

Fortunately, the marketplace has shown signs of resilienceBy the second quarter of 2024, IoT module shipments bounced back with an impressive 11% year-over-year growthThis revival can be attributed primarily to the notable expansion within the market for Point of Sale (POS) systems, particularly portable devices like compact speakers, as well as the automotive sector's increasing adoption of IoT solutions.

Moreover, the exciting intersection of Artificial Intelligence (AI) and IoT is creating new opportunities for innovationThis fusion stands to enhance the sophistication of IoT applications, paving the way for smarter, more efficient systems.

Despite the promising potential this confluence offers, it is important to recognize that the integration of AI into IoT is still in its nascent stagesAs of 2023, AI modules constituted only about 6% of the total IoT module shipmentsHowever, the trajectory is poised for an upward leap, with projections suggesting that this figure could rise to 25% by 2030.

Today, various enterprises within the AI module supply chain are making significant strides

Advertisements

The upstream components, which primarily include chips, discrete devices, and printed circuit boards (PCBs), form the backbone of AI modulesRemarkably, chips alone account for more than 80% of the value in this arena.

Currently, Qualcomm dominates the market for AI module chips, packing in a substantial share, while domestic alternatives remain limitedThankfully, this scenario is improving, with companies such as MediaTek and Huawei HiSilicon emerging as viable substitutes.

Among mid-stream manufacturers of AI modules, Meig Smart is rapidly advancing, having already launched several products in this domainIts SNM970 model boasts an impressive AI computing power of 48 TOPS, marking it as the industry's first high-power AI module capable of running advanced inference modelsThis technological breakthrough has opened the doors for collaboration with multiple domestic manufacturers, allowing them to procure high-performance modules for their innovative AI glasses.

Not stopping there, other players in the market like Quectel and Gtbit have also expedited their AI module rolloutsQuectel's SG368Z series of AI intelligent modules targets applications such as facial recognition and hazard area alerts, providing holistic scene applications and integrated algorithmsMeanwhile, Gtbit's SC171 is designed to handle 12 TOPS of AI computing power, targeting the imagery and visual fields.

For Quectel, positioning itself strongly within the AI domain is of paramount importanceIt currently stands as a market leader, manifesting a robust 36.5% global market share as of the second quarter of 2024—significantly outpacing the second-ranked Gtbit.

That said, competition among communication module manufacturers is ferocious, with only minor technical variances separating themThe primary distinction lies within their specific application areas, leading to frequent fluctuations in the rankings of leading firms.

Emerging players like Lierda are storming the field, leveraging their budget-friendly modules to compete for market share against established domestic brands

Advertisements

As a result, Quectel must continually innovate to maintain its competitive edge.

Moreover, while contending with intensifying competition, Quectel's profitability notably lagsIn the first three quarters of 2024, the company's gross profit margin stood at 17.9%, which falls short of Gtbit's 20.7%.

The distinction in gross margin between these manufacturers largely stems from their differing product structuresGtbit tends to focus on high-value consumer electronics, whereas Quectel primarily pursues the mobile payment, smart energy, and intelligent city sectors—areas characterized by lower profit margins, adversely affecting overall profitability.

AI modules are critical for boosting Quectel's gross marginThe company estimates that the cost of a high-rate 5G module hovers around $60 each, while AI computing modules can command prices as high as $200 per unit, leading to substantially higher profit margins.

In addition to efficiency strategies, Quectel has adopted competitive pricing policies to secure market share, ultimately resulting in a lower gross margin compared to its rivals.

However, relying on a low-price strategy in the long run is unsustainable for any businessBy fostering AI modules, Quectel can not only enhance its profitability but also shift away from its current low-priced approach.

So, is Quectel adequately prepared to ride the new wave driven by AI?

For starters, the industry revival has brightened the company’s financial outlook, providing ample financial reserves to explore business ventures related to AI modulesIn the first three quarters of 2024, the company's revenue reached ¥13.25 billion, marking a robust year-over-year increase of 32.9%. Moreover, the company's net profit attributable to shareholders skyrocketed to ¥360 million, with an astonishing 2078.7% growth.

That said, it should be noted that during the same period, the company reported an asset impairment loss of ¥47 million, affecting net profit by approximately ¥38 million

Advertisements

This impact, however, is considered reasonable and temporary.

The sources of this asset impairment stemmed from two key areas: a ¥22-million provision for bad debts preparation related to receivables, and a ¥24-million provision for inventory devaluationGiven that Quectel's accounts receivable soared by 69.6% to reach ¥2.56 billion, it is only prudent to provide for potential risks.

On the inventory side, driven by a recovering industry, the company reported a 2.7% decline in inventory year-over-year, while inventory turnover surged by 25.3%, thereby reducing the risk of future inventory impairment losses.

Additionally, the firm’s investment in research and development (R&D) is playing a fundamental role in the expansion of its AI module offeringsIn the first three quarters of 2024, R&D expenditures climbed by 2% to ¥1.23 billion, resulting in an R&D expense ratio of 9.3%—higher than competitors such as Gtbit and Zhongji Xuchuang.

In terms of R&D results, beyond the previously mentioned AI intelligent modules, Quectel has also developed an AI algorithm platform, "Craftsmanship," distinguishing itself as one of the industry’s few firms possessing both hardware and software capabilities for edge computing development, alongside self-developed AI algorithms to provide AI large model training services to end customers.

Furthermore, the models designed by the "Craftsmanship" platform are compatible with standard X86 CPU architectures and Nvidia graphics cards, as well as supporting embedded ARM architectures from Qualcomm and Spreadtrum, making it an exceptional offering.

Given that the hardware costs linked to ARM architecture can be reduced by over 50% in comparison to X86 architecture, Quectel's AI computing platform therefore stands to enhance the competitiveness of its AI module products, enabling access to larger market segments.

In summary, as AI and IoT grow more intertwined, the demand for AI modules is poised to rise

Advertisements

Advertisements