Meta Platforms META shares are trading at a premium, as suggested by the Value Score of C. In terms of the forward 12-month price/sales, META is trading at a premium of 6.01X, higher than the Zacks Internet Software industry’s 3.91X and Amazon’s AMZN 2.73X. However, Meta Platforms’ shares are trading at a discount compared with Alphabet GOOGL and Microsoft MSFT, shares of which are trading at 8.83X and 8.06X, respectively.
Image Source: Zacks Investment Research
So, is the Meta Platforms stock a buy or hold at this level? Let’s find out.
Meta Platforms shares have dropped 6.7% year-to-date (YTD), underperforming the broader Zacks Computer & Technology sector’s fall of 2.9% and Alphabet’s decline of 1.7%. However, META shares have outperformed Microsoft and Amazon, shares of which have dropped 9.1% and 19% over the same time frame, respectively.
Image Source: Zacks Investment Research
Meta Platforms is spending heavily on AI research, models and infrastructure. The company now expects 2026 capital spending between $115 billion and $135 billion. Per CNBC, Alphabet, Meta Platforms, Amazon and Microsoft on a combined basis are expected to spend roughly $700 billion on developing AI infrastructure in 2026. Although these investments are expected to boost META’s prospects over the long term, a challenging macroeconomic environment, regulatory issues (in the European Union and the United States) and stiff competition in the ad market from the likes of Alphabet, Amazon, Snap and TikTok, among others, is expected to remain an overhang on the shares. Higher capital spending is also expected to squeeze free cash flow.
META is expected to suffer from higher operating expenses. For 2026, META anticipates total expenses between $162 billion and $169 billion, driven by higher infrastructure costs, which include third-party cloud spend, higher depreciation, and higher infrastructure operating expenses. Higher employee compensation driven by hires in the AI space is expected to drive expenses.
META’s focus on integrating AI into its platforms — Facebook, WhatsApp, Instagram, Messenger and Threads — is driving user as well as advertising engagements. AI is heavily dependent on data, of which META has a trove, driven by its more than 3.58 billion daily users, including 2 billion daily actives each on Facebook and WhatsApp. Time spent across platforms is expected to benefit from Meta Platforms’ continuous ranking optimizations.
AI recommendations that deliver higher quality and more relevant content are expected to drive engagement. The company is using Meta AI to boost user experience. In the fourth quarter of 2025, within Meta AI, the number of daily actives generating media tripled year over year. The company expects to advance the capabilities of META’s underlying media generation models and ship new features to further enhance the product experience in 2026. Focus on expanding personalization on Meta AI is expected to help the company understand user interests and preferences, as well as identify the most relevant content across the META platform. The company has started testing Meta AI business assistant with advertisers, which helps with tasks like campaign optimization and account support.
Beyond improvements to its recommendation systems, Meta Platforms expects to use the models developed by Meta Superintelligence Labs to deliver compelling and differentiated AI products. The ad business is benefiting from an improved AI ranking system. The company has a strong pipeline of ad supply opportunities on both Threads and WhatsApp Status over the long term. Ads are now running globally in Feed on Threads, and META plans to optimize the ad formats and performance before increasing supply. The company has extended its Andromeda ads retrieval engine so it can now run on NVIDIA, AMD and MTIA.
The latest multi-year deal with NVIDIA will support META’s build-out of data centers optimized for AI training and inference, as well as its core business. The NVDA deal supports META’s AI-related endeavors. The company is also adopting NVIDIA Confidential Computing for WhatsApp private messaging and the NVIDIA Spectrum-X Ethernet networking platform across its infrastructure footprint.
