Microsoft: Cheap For All The Wrong Reasons

Microsoft is now the worst-performing Mag 7 stock, down 18% since August, despite no stated deterioration in fundamentals. The stock has re-rated to 9.4x forward EV/Sales and 24x GAAP P/E, making it the cheapest among its peers, but AI monetization is lagging expectations with Copilot penetration at just 3.3% of Microsoft 365 seats. The note points to valuation compression and slower-than-expected AI adoption rather than a fundamental crisis.

Analysis

MSFT’s underperformance looks less like a business problem and more like a crowded-duration de-rating: when a mega-cap with durable cash flows misses the market’s AI monetization hockey stick, multiples compress fast because the incremental buyer is paying for optionality, not current earnings. The key second-order effect is that capital is likely rotating within AI infrastructure toward names with clearer near-term monetization or faster revenue conversion, while software peers with similar copilots can be re-underwritten lower if enterprise adoption remains measured.

The slow penetration rate matters because it delays the feedback loop that would otherwise validate aggressive AI spending across the ecosystem. If Copilot remains a sub-5% seat penetration story over the next 1-2 quarters, investors may start questioning whether AI attach rates are constrained by workflow friction, budget fatigue, or weak ROI proof points; that would pressure not just MSFT, but also adjacent enterprise software vendors and channel partners that were implicitly underwriting AI-led seat expansion.

The contrarian setup is that the current move may be partially self-correcting: MSFT is now priced closer to a ‘great software company’ than a ‘perfect AI winner,’ which lowers the bar materially. Any evidence of improved seat expansion, bundle conversion, or stronger Azure-led AI consumption over the next 1-2 earnings cycles could trigger a sharp multiple re-rating because positioning has likely shifted from complacent long to more balanced. In other words, the stock is vulnerable to further derating if adoption stays sluggish, but it is also set up for a fast reversal on even modest proof of monetization.

Key Decisions for Investors

• Short-term: fade strength in MSFT via a 1-2 month call spread overwrite or outright short on rallies into earnings, targeting continued multiple compression if Copilot adoption fails to accelerate; risk/reward is attractive because downside is driven by sentiment and positioning rather than fundamentals alone.
• Pair trade: long the most visibly monetizing AI infrastructure beneficiary and short MSFT for 1-3 months, expressing the view that capex wins are better than software optionality until adoption data improves; this reduces market beta while isolating relative monetization execution.
• For longer-dated positioning, buy MSFT 6-9 month put spreads only if the stock re-rallies on general Mag 7 momentum; the thesis is a controlled downside hedge against another quarter of weak attach-rate data and further multiple compression.
• If MSFT sells off another 5-8% without a corresponding earnings downgrade, begin scaling into a tactical long: the stock is now cheap enough that even modest Copilot/Azure AI upside could drive a sharp 10-15% rebound on multiple expansion.