Microsoft Q1 Preview: Copilot Struggles, Customers Not Satisfied (Rating Downgrade)
Contributor since 2023 / 3099 Followers
Summary:
- Microsoft shares have declined 2.62% amid AI competition and Copilot’s inconsistent performance, leading to my downgrade to a strong sell.
- Customer dissatisfaction with Copilot’s reliability and value is threatening Microsoft’s AI-driven growth, casting doubt on future earnings and market position.
- Microsoft’s high valuation metrics, coupled with slowing EPS growth and substantial Capex needs, suggest shares are overvalued and face a potential 26.19% downside.
- Without significant improvements in Copilot’s data architecture and performance, Microsoft risks alienating its customer base and undermining its AI investments.
hapabapa
Investment Thesis
Microsoft (NASDAQ:MSFT) (NEOE:MSFT:CA) shares have sold off since the last time I wrote on the tech giant, with shares falling 2.62% since my last analysis over the summer.
From my standpoint, I think a big driver of this selloff (even though the overall market has risen 5.49%) has been the broader market realization that the AI revolution is not a winner-takes-all market, especially when it comes to competition among tech giants.
While Microsoft has invested in one of the leading AI companies (OpenAI), their current position does not guarantee their dominance.
Now, corporate customers of Microsoft’s Copilot software are complaining that their flagship product is not adding enough value to justify the price they are paying.
Reports indicate that many corporate users are dissatisfied. The criticisms suggest that Copilot often produces inconsistent results based on minor prompt adjustments, leading to frustrations over its reliability (and questions on how far it can scale in the enterprise environment). Even Salesforce (CRM) CEO Marc Benioff posted his opinion on why customers are not getting much value from Copilot.
Recent analysis (found by Seeking Alpha analyst, Damir Tokic) revealed that over 80% of AI projects ultimately fail. For an enterprise, if an AI project fails, they can simply cancel their AI subscriptions. For Microsoft, they are highly vulnerable given their substantial capital expenditures in this area—over $44 billion in the past year alone (Damir’s analysis). I agree with Damir’s point: Microsoft’s valuation is clearly at risk if these AI initiatives do not translate into widespread adoption. Right now, customer feedback is threatening to push the company in the wrong direction.
With Microsoft facing a big reset on Copilot expectations, their valuation appears increasingly stretched, as the street’s anticipated growth from AI may not materialize as expected. I think the stock’s current performance suggests that investor confidence is wavering.
With this, I am now downgrading my stance on the company to a strong sell given the heavy reliance on AI success given the enterprise environment is getting more intense and the high failure rate of AI projects.
Why I’m Doing Follow-Up Coverage
Microsoft’s stock performance has lagged behind the overall market since my last analysis in July, with the sell-off largely influenced by the intense competition in AI that has become increasingly apparent since my last coverage. Initially, I thought the implications of this intensified competition were not a big deal for Microsoft, but it is now clear that the company’s position is far from secure. Previously, I thought in my analysis that Microsoft had competition from Google (GOOG) but that the market was big enough for both of them. Now, I am less optimistic about Microsoft.
In light of this competitive pressure, I think the upcoming earnings report is crucial.
I think analysts’ Q1 2025 earnings projections are overly optimistic because they do not account for the operational challenges facing Copilot (and a smaller order book as a result).
The expectation that Microsoft can simply continue to capture market share through Copilot may be misguided. Many users do not appear to see sufficient value.
The purpose of this follow-up coverage is to show what the customer problems are with Copilot and how this will affect earnings. Microsoft needs to do more in order to support future Copilot growth.
Why Copilot Is Struggling
The struggles faced by Microsoft’s Copilot were pointed out by Salesforce CEO Marc Benioff in an X (formerly Twitter) post. He characterized Copilot as “a flop,” attributing its failures to several factors.
Microsoft rebranding Copilot as ‘agents’? That’s panic mode. Let’s be real—Copilot’s a flop because Microsoft lacks the data, metadata, and enterprise security models to create real corporate intelligence. That is why Copilot is inaccurate, spills corporate data, and forces customers to build their own LLMs. Clippy 2.0, anyone? Meanwhile, Agentforce is transforming businesses now. Agentforce doesn’t just handle tasks—it autonomously drives sales, service, marketing, analytics, and commerce. With data, LLMs, workflows, and security all integrated into a single Customer 360 platform: This is what AI was meant to be.
The root of the problem appears to be in Copilot’s struggle to deliver high-quality answers consistently, leading to widespread dissatisfaction among users. This has been raised in some user forums already, citing its poor responses and even weak use of enterprise internal data to develop sufficient answers.
In February, it was reported that the company launched an investigation into “disturbing” responses made by Copilot chatbot’s feature. A month later, more customers complained about the performance of the tool, which they compared to ChatGPT.
While Salesforce is working on a competing tool, I think Benioff is right. Businesses expect AI to enhance productivity and streamline operations, yet if Copilot falls short in providing reliable and accurate responses, it undermines the very value proposition that justifies its cost.
Q1 Preview
Microsoft is set to report their FY Q1 2025 earnings on October 30 after the bell, with consensus estimates suggesting an EPS of approximately $3.10, reflecting a year-over-year growth of just 3.80%. Revenue for this quarter is projected to reach $64.56 billion, up 14.23% from the previous year.
Diving deeper, earnings revisions over the past three months reflect a more cautious position.
According to data from Seeking Alpha, there have been 8 upward revisions and an eye-opening 31 downward revisions for EPS. On the revenue side, 16 upward revisions are dwarfed by 32 downward revisions over the same 90-day trailing period. At this point, I think expectations are beginning to adjust to reported customer dissatisfaction with Copilot and what a slowdown means to the long trajectory of the firm.
Personally, I am pretty bearish at this point because I think Copilot is not shaping up to be the long run growth driver Microsoft wanted. On the upcoming conference call, I’ll be looking for any discussion from management on how they can address user complaints against Copilot.
AI can be highly useful in a lot of cases, but it has to be deployed correctly, with the combination of the right data and the right inferencing from the foundation models that are being used. I am just not seeing that from Copilot.
Valuation
We can see that EPS growth is slowing for Microsoft (only 3.8% YoY), yet the tech giant trades at an extended P/E that is not justified given where EPS growth is right now. The company’s forward P/E ratio comes in at 32.52, well above the sector median of 24.16.
From a price to sales ratio, the company trades at 11.40x times next year’s revenue. This is also well above the sector median’s 2.97 by 283.52%.
Unfortunately, the issue now for tools like Copilot is that market reception has moved from ‘I will try any AI tool that could make my business more efficient’ to ‘I will be careful in which tools I provide my employees.’ This market shift is not a winning one for Microsoft, and I think the outlook for earnings is more inclined towards further downward revisions for EPS estimates. Analysts are beginning to price in the reality of slower growth for Copilot and, by extension, Microsoft’s overall performance. The current earnings estimates feel overinflated to me.
Given these valuation metrics and the overall downward trend in earnings revisions, Microsoft’s shares appear overvalued. I think given EPS growth is slowing, shares should be repriced to the sector median forward P/E. If Microsoft’s forward P/E were to align more closely with their sector’s median P/E ratio, this would represent roughly 26.19% downside.
Bull Thesis
Microsoft has indirectly bet the future of their Office 365 suite (and by extension the future image of the company) on Copilot. For me, my bullish thesis revolves around the use of advanced models, specifically OpenAI’s latest model o1, to enhance Copilot, so users feel like they are getting the value they want out of it.
For Microsoft, however, much of the $44 billion in infrastructure investments they made over the last year will not help them scale up new models like o1 to power Copilot which means they face an incredibly heavy Capex cycle ahead to build the GPU infrastructure to power a new model like o1 that may (not guaranteed) make Copilot powerful enough for many users to continue to pay for it.
NVIDIA’s (NVDA) new Blackwell architecture and Advanced Micro Devices’ (AMD) chips help speed up inference for the new o1 model. Both are not cheap, and both just hit the market. This means many of the GPUs Microsoft has purchased cannot inference fast enough for them to be useful for these new models. As of the last quarter, Microsoft was spending $19 billion on Capex ($76 billion on an annualized basis). This is up 72.7% YoY on an annualized basis from $44 billion. The company has explicitly said Capex will be higher in FY 2025 than it was in FY 2024.
Even then, (with more powerful GPUs) Copilot, like other AI tools, is dependent on high-quality data inputs. Any AI model is directly tied to the quality of the data used to train it. Microsoft is planning to offset this by building a more effective feedback system into their Copilot tool to help understand where users are using it (and keep coming back, implying they value it) and where they are not using it. Data-driven insights are key. But this is not enough on its own to build a high-quality product.
Unfortunately, this is where Copilot has shortcomings (we saw this already with users complaining it does not ingest and process core data adequately to produce quality insights).
Specifically, Copilot lacks a knowledge graph, which is a relational database structure that facilitates implicit connections between disparate data points that allow AI tools to produce more nuanced and contextual analysis. So, even if Copilot identifies where the product is lacking (from a user feedback standpoint) the core tool lacks the database it needs in order to be successful. This means there’s a lot more investment (and potentially Capex) that needs to happen to make the data Copilot pulls from more reliable. Both of these cost money.
Because of this, Copilot is currently vulnerable to the pitfalls of ‘bad data in, bad data out.’ Poor data quality can diminish user trust and adoption rates, as organizations may seek alternatives that offer better data management and insight generation.
Takeaway
Microsoft’s Copilot, while designed to improve workplace productivity, is still heavily contingent on the quality of the data ingested. Without the right data architecture, its results are compromised, given the high price Microsoft seeks to charge.
Copilot does not adequately support the ingestion of high-quality data. Users expect reliable, consistent output, yet reports indicate that many find Copilot’s performance lacking when compared to competitors.
In light of this, I think the outlook for Microsoft shares has gotten worse heading into earnings and my personal belief has shifted to a strong sell.
If the company cannot improve Copilot, the tech giant really risks alienating its customer base.
Analyst’s Disclosure: I/we have a beneficial long position in the shares of AMD either through stock ownership, options, or other derivatives. I wrote this article myself, and it expresses my own opinions. I am not receiving compensation for it (other than from Seeking Alpha). I have no business relationship with any company whose stock is mentioned in this article.
Noah Cox (main account author) is the managing partner of Noah’s Arc Capital Management. His views in this article are not necessarily reflective of the firms. Nothing contained in this note is intended as investment advice. It is solely for informational purposes. Invest at your own risk.
Seeking Alpha’s Disclosure: Past performance is no guarantee of future results. No recommendation or advice is being given as to whether any investment is suitable for a particular investor. Any views or opinions expressed above may not reflect those of Seeking Alpha as a whole. Seeking Alpha is not a licensed securities dealer, broker or US investment adviser or investment bank. Our analysts are third party authors that include both professional investors and individual investors who may not be licensed or certified by any institute or regulatory body.