On Tuesday, Pivotal Research raised Meta Platforms' price target sharply from $790 to $930, while maintaining a "buy" rating, implying about 25% upside.
Pivotal Research on Tuesday will$Meta Platforms (META.US)$The target price has been sharply raised from $790 to $930, while the "Buy" rating is maintained, implying roughly 25% upside. Analyst Jeffrey Wlodarczak's core thesis is that Meta's returns on its AI investments are "coming into view"—early adoption of the Muse personal AI assistant has been robust, and following the launch of the Muse Wonder model, it has already positioned itself among the industry's frontrunners.
This view is far from isolated on Wall Street. S&P Global's consensus of 63 analysts puts the target price at $799, with a "strong buy" rating; however, the average target price implies only about 7.5% upside from Meta's Monday closing price of $742.96. Pivotal's $930 target is roughly 16% above the consensus, placing it at the high end of the range.
The two pillars of Pivotal's bullish outlook
Pivotal's bullish thesis rests on two verifiable data points.
First, the installation speed of Apple. The research report notes that in the first 12 days after its launch, Muse accumulated 2.8 million installs, surpassing ChatGPT's early mobile performance. This figure is backed by third-party data—Sensor Tower estimates ChatGPT's global downloads during the same period at 3.7 million, while Pivotal's cited "surpassing" is based on a comparison using North American Apple device metrics.
Second, a substantial catch-up in model capabilities. Pivotal believes that Muse Wonder is$Meta Platforms (META.US)$This cutting-edge model, unveiled for the first time since Alexandr Wang restructured the AI business in August 2025, delivers performance on par with or close to that of leading lab‑developed models. Independent benchmarking by Artificial Analysis corroborates this assessment: Muse Wonder Full scored 56 on the Intelligence Index, matching Anthropic's Claude Sonnet 5.5 and outpacing OpenAI's GPT‑6 Astra and Google's Gemini 4 Argon, trailing only Claude Opus 5.5's 58. The standard version of Muse Wonder earned a score of 52.
From a product perspective, Muse Wonder introduces two core capabilities tailored for agentic scenarios: Active Memory, which preserves user information across conversations, and Active Focus, which enables the model to persist in advancing the task even when encountering obstacles, rather than abandoning its goal. The model supports a context window of approximately one million tokens, and its training was conducted at Meta's Prometheus data center in Ohio.
Leading in downloads doesn't necessarily mean strong retention.
However, behind the glossy facade of download figures, the retention curve is far from encouraging.
According to Sensor Tower data, Muse achieved a 73.3% day-1 retention rate, significantly higher than ChatGPT's 41.3%. However, this advantage rapidly narrowed after the seventh day: by day 14, Muse's retention had dropped to 22.5%, below ChatGPT's 26.8% for the same period and also lower than that of WhatsApp, YouTube, Instagram, and other apps.$Meta Platforms (META.US)$The level of our own mature products.
BNP Paribas analyst Nick Jones previously noted that while Muse's app‑open rate has stabilized, it remains lower than that of Google, ChatGPT, and Meta's established apps, suggesting the product has yet to fully integrate into users' daily routines. The firm maintained its "Outperform" rating on Meta and an $885 price target, but lowered its long-term monetization outlook for Muse. Jones's key view is that Muse's initial download momentum largely stems from Meta's robust distribution capabilities within its social ecosystem, and that "easy to download, hard to retain" is a common challenge for consumer‑grade AI products.
The historical baggage of monetization
Needham analyst Laura Martin maintains a "Hold" rating and has not set a price target, citing the following reasons:$Meta Platforms (META.US)$"It is known for its slow monetization of new products."
This assessment is backed by historical precedent. In 2012, Meta acquired Instagram for $1 billion, yet its first ads did not appear until late 2013; in 2014, it bought WhatsApp for $19 billion, but advertising only began in June 2025—more than a decade later. Martin notes that Muse is currently free, and Meta has explicitly stated it will not share users' conversations or data with its advertising business, "suggesting limited monetization potential."
More realistic constraints stem from the capital expenditure side.$Meta Platforms (META.US)$The 2026 capital expenditure guidance stands at $130–$145 billion, with Q2 capital spending already reaching $31.1 billion, while free cash flow plummeted from $8.55 billion in the same period last year to just $784 million. Pivotal's claim that "AI‑driven returns are coming into view" must be assessed against this rapid pace of cash consumption. Whether Muse can shift from a "user‑growth story" to an "revenue‑growth story" will determine whether its $930 price target is forward‑looking or overly optimistic.
$Meta Platforms (META.US)$The company will release its Q3 earnings after the market closes on October 21, Eastern Time, with market expectations projecting quarterly revenue of approximately $63.15 billion. At that time, investors will have their first opportunity to assess the actual revenue contribution from the Muse launch, as well as management's specific guidance on capital expenditures through 2027 and the roadmap for monetizing AI. Pivotal's aggressive pricing strategy will face its first meaningful test in two weeks.
Edited by melody