Meta Platforms (META): Deutsche Bank sees Muse opening a potentially meaningful new product and monetization cycle for Meta

Overview

Deutsche Bank assesses Meta’s Muse launch and wearable-device strategy as the start of a rich product cycle. It views fast downloads and a consumer-oriented agentic-AI experience as evidence of product-market fit, with commerce, subscriptions, payments and advertising creating several potential monetization channels over time.

Core views

Deutsche Bank’s central argument is that Muse has achieved unusually strong initial consumer traction for an AI application. It describes the service as combining an easy consumer interface, mobile-app ecosystem, autonomous transaction capability, largely free access and an always-on virtual-machine experience. Muse reached No. 1 in iOS Productivity, held a 4.9-star rating from 38,000 ratings, and generated more than 3 million US installs in its first 16 days according to Sensor Tower. The institution interprets these results as evidence of substantial product-market fit and a first-mover advantage in consumer agentic AI. The report believes Meta is deliberately prioritizing scale over near-term subscription monetization. The free tier permits up to 100 million tokens per week, while the Power tier offers 500 million weekly tokens for $20 per month, or $16 on muse.ai, and the Maximum tier offers 3 billion weekly tokens for $100 per month, or $80 on the web. Deutsche Bank expects Meta may later tighten free-tier usage allocations to encourage upgrades after it has built a large user base. Partnerships with Walmart, Best Buy, Gap, Sephora, Instacart, Wayfair, PayPal, Stripe, Shopify and Expedia are intended to expand use cases; the report thinks early integrations may provide partners with traffic without revenue sharing initially. It expects Meta eventually to seek a modest cost-per-completed-action advertising take rate as Muse becomes a larger source of traffic and sales. Amazon’s current block on Muse access makes Walmart’s partnership more strategically important, although Deutsche Bank believes Amazon could ultimately lift its restriction. Hardware is presented as a distribution and engagement lever for Muse. Meta introduced lighter wearable devices with additional Muse interactions, including third-generation Glasses with private-compute live translation and FDA-approved hearing-aid capabilities, audio-only Glasses with day-long battery life, and VR Glasses that are five times lighter than Quest 3. The VR Glasses are expected in spring 2027 at $1,299. Meta also showed the screen- and voice-enabled Muse Charm handheld device, which it said should ship for the holidays. Deutsche Bank sees these devices as an early demonstration that hardware can serve as an effective delivery channel for Meta AI and broaden both distribution and engagement. The report builds a 2030 US commerce-and-services opportunity from retail sales and personal-consumption-expenditure data, rolling spending forward at 3.5% annual inflation. It estimates agentic AI could address $2.5-$5.5 trillion of spending by 2030, with e-commerce the largest opportunity because agents could potentially influence more than 50% of online spend. Its Muse adoption scenarios assume 15%, 25% and 35% adoption among 216 million Meta US users, yielding 32.4 million, 54.0 million and 75.6 million Muse users. The model assumes that the paying-user mix rises as adoption and usage broaden. Deutsche Bank identifies subscriptions, transaction fees and interchange spread as the principal revenue sources. It estimates 2030 subscription revenue of $1.3-$6.4 billion. For transaction revenue, it assumes Muse captures 8%-18% of agentic-AI spending, has active merchant partnerships covering 25%-60% of its gross merchandise volume, and receives a 2%-5% take rate on that partnered volume. This produces estimated transaction-fee revenue of roughly $1.0-$29.0 billion; an assumed 5-10 basis-point share of interchange spread adds about $0.1-$1.0 billion. Total modeled Muse revenue is $2.4 billion, $10.7 billion and $36.3 billion in the three scenarios, equal to 0.5%, 2.3% and 7.8% of Meta’s FY30E revenue, respectively. The estimates exclude potential interest income from Muse Wallet balances. The report also flags future advertising opportunities in Muse search results and its Feed tab, as well as better advertising targeting if users permit data sharing with Meta’s broader organization. The institution explicitly considers the cost burden of operating a consumer agent. It estimates $25-$50 of virtual-machine-related capital expenditure per Muse user and $4.50-$9.50 of annual depreciation. Its monthly base-cost estimate is about $3.50-$6.50 per user, based on virtual-machine memory, storage, compute, depreciation and data-center operating costs. Including inference, estimated monthly service costs are $5.43 for Free users, $35.05 for Power users and $165.90 for Maximum users. Deutsche Bank notes that Meta could adjust token limits over time, following an industry pattern it characterizes as “shrinkflation,” which could contain costs and create more upgrade opportunities. Meta’s infrastructure investments and frontier-approaching Muse 1.3 models are viewed as a meaningful moat, although rising Muse adoption could reduce Meta’s flexibility to lease data-center capacity to third parties in the short term. Beyond the consumer offering, Deutsche Bank expects personal agents eventually to be supplemented by business agents connected to Meta’s broader business ecosystem. With more than 200 million businesses on Meta’s platform, it sees this as a potential future catalyst; the upcoming Watermelon model is identified as a nearer-term possible catalyst. The report argues that these catalysts, combined with what it calls Meta’s low valuation multiple, support a positive rerating. It raises the target price to $820 from $750, using 25x FY27 EPS rather than the prior 23x multiple. For Meta’s broader coverage universe, Deutsche Bank sees agents becoming a demand-aggregation layer between consumers and digital platforms. Outcomes depend on whether a company controls differentiated inventory or differentiated discovery, its dependence on paid acquisition, and whether it blocks agents, permits access or forms a partnership. Platforms with differentiated inventory, strong brands and direct consumer relationships appear best placed; businesses reliant on consumer friction, low transparency or subscription inertia face greater disruption. Online travel agencies may regard agents as another acquisition channel, while differentiated platforms such as Airbnb may have less urgency to participate. Etsy and eBay could benefit if agents surface long-tail inventory, whereas marketplaces reliant on sponsored-listing advertising could face pressure if agent-led shopping reduces impressions and seller advertising spend. Deutsche Bank considers Chewy relatively defended by Autoship, scaled fulfillment and tailored delivery, while Amazon’s refusal to permit broad Muse access may become harder to maintain if agentic commerce becomes a meaningful channel. Subscription-oriented businesses may face greater churn risk when agents identify and cancel underused services, though music-streaming services are seen as more insulated because of low cost, high engagement and limited content overlap.

Analysis framework

Deutsche Bank begins with observed download data and product demonstrations, then assesses Muse’s user proposition, partner ecosystem and hardware distribution. It models 2030 adoption, addressable spending, monetization and per-user infrastructure and inference costs under three scenarios, before extending the analysis to implications for travel, marketplaces and subscription businesses.

Methodology notes

• Other

  Scenario-based Muse revenue and cost model

  The report uses three 2030 scenarios for user adoption, paid-tier mix, addressable spending, market share, partner participation and take rates to estimate subscriptions, transaction fees, interchange income and per-user service costs.

• Industry AnalysisUpstream-Midstream-Downstream Transmission

  Agentic AI as a demand-aggregation layer

  The report evaluates how an AI-agent interface could redistribute discovery, traffic, transaction economics and bargaining power between consumers, platforms, merchants and inventory owners.

Asset mapping & comparison

Structured mapping from thesis to named assets (strengths, weaknesses, peers, risks).

• Meta Platforms (META.US)

  Primary subject and potential beneficiary of Muse adoption, commerce monetization, advertising surfaces and wearable distribution.

  Strengths

  Rapid early Muse downloads, large distribution base, infrastructure scale, partnerships and potential first-mover advantage.

  Weaknesses

  Near-term monetization may be modest while Muse remains a loss leader; the company remains resource and compute constrained.

  Comparison

  Deutsche Bank views Meta as holding a head start versus Apple, Alphabet and OpenAI in scaled consumer agentic AI.

  Risks

  Higher adoption could reduce flexibility to lease data-center capacity; revenue assumptions depend on future user adoption and merchant participation.

• Chewy

  A broader-coverage company assessed for potential agentic-commerce disintermediation.

  Strengths

  Autoship, scaled fulfillment and tailored delivery are cited as meaningful defenses.

  Weaknesses

  Investor concern has focused on possible disintermediation.

Key data

• Muse US installs>3 millionEstimated by Sensor Tower during Muse’s first 16 days in the US.
• Muse iOS app positionNo. 1 in Productivity; 4.9 stars from 38,000 ratingsEarly adoption indicators cited by Deutsche Bank.
• 2030 agentic-AI addressable opportunity$2.5-$5.5 trillionUS retail and services opportunity in Deutsche Bank’s scenarios.
• 2030 Muse users32.4 million / 54.0 million / 75.6 millionDerived from 15% / 25% / 35% adoption among 216 million Meta US users.
• 2030 total Muse revenue$2.386 billion / $10.703 billion / $36.311 billionThree scenarios, equivalent to 0.5% / 2.3% / 7.8% of Meta FY30E revenue.
• Monthly Muse service cost per user$5.43 Free / $35.05 Power / $165.90 MaximumIncludes virtual-machine base costs and inference costs.
• Target price$820Raised from $750; based on 25x Deutsche Bank’s FY27 EPS estimate versus 23x previously.

Impact & implications

The report says Muse could become a meaningful new revenue and engagement engine for Meta if it establishes durable consumer use cases and partner participation. For other digital platforms, it expects the main issue to become how value is divided between the agent interface owner and the inventory owner, with differentiated inventory and direct customer relationships offering relative protection.

Risks

• Muse may remain a near-to-medium-term loss leader because payment-interchange economics are modest before usage reaches critical mass.
• Meta’s resource and compute constraints could limit its ability to lease data-center capacity to third parties as Muse adoption expands.
• Merchant participation and consumer adoption may not become broad enough to establish a durable agentic demand-aggregation layer.
• Businesses that depend on customer friction, low transparency or subscription inertia may face greater long-term disruption from agents.

What to watch

• Muse’s ability to add partners, expand consumer utility and sustain adoption.
• Whether Meta narrows free-tier token allocations and converts heavier users to paid tiers.
• The launch of Watermelon, described as Meta’s most advanced model, as a potential earlier catalyst.
• Whether Amazon maintains its Muse block or negotiates an integration if agentic commerce becomes meaningful.
• The emergence of business agents across Meta’s ecosystem of more than 200 million businesses.