73% of installs of Meta’s AI agent Muse are active on Day 1, and just 23% remain active by Day 14, according to a chart that the X account @intelfabs posted on X on 20 October 2026 and credited to a Morgan Stanley note titled “How Much Capacity Will Muse Need and 5 Agentic Reads from the Answer.” By our arithmetic, that means about 68% of the people active on Day 1 were no longer active by Day 14. The figures land as Meta prepares to report third-quarter results after the US market close on 21 October 2026.
In plain terms, retention measures how many people who install an app are still using it some days later. A Day 14 figure counts the share of new installs that are still active two weeks after they started.
Business Pill · CHURN
A short explainer on churn: the customers who quietly stop using a product. It teaches the general idea only and says nothing about any company in this story.
The key insight: As we read it, the number that matters for Muse is no longer installs but how many of them stay: the chart credited to Morgan Stanley shows 73% active on Day 1 and 23% by Day 14.
The chart is titled “Muse wins day one. Then loses the user.” It plots Muse against Instagram, Google and Facebook between Day 1 and Day 14, and labels only the two Muse figures: 73% and 23%.
The lines for Instagram, Google and Facebook carry no value labels. On the chart, all three end above Muse at Day 14, but the chart does not state their figures, so we do not give them here.
In the post that carried the chart, @intelfabs wrote: “Day-one adoption isn’t the problem. Retention is.” The account added that “If the product improves, that gap could become the opportunity.”
Share of Muse installs active on Day 1 (73%) and Day 14 (23%), as charted by @intelfabs on 20 October 2026 and credited to a Morgan Stanley note. The 50-point gap and the 68% figure are our arithmetic. Source: @intelfabs on X, citing Morgan Stanley.
What Was Posted on X
The demand is already there. Now $META needs to make Muse sticky.
Day-one adoption isn’t the problem. Retention is. If the product improves, that gap could become the opportunity.
BTW, you can just buy a Muse account, very cheap too!!!
The @intelfabs post that carried the chart, on 20 October 2026. It writes that Metaneeds to make Muse sticky and that day-one adoption is not the problem but retention is. The chart it attaches credits a Morgan Stanley note and labels Muse at 73% active on Day 1 and 23% by Day 14; the opinions in the post are the account’s own.
What Morgan Stanley Said About Muse
Morgan Stanley reiterated its Overweight rating on Meta on 19 October and raised its price target to $820 from $775, Stocktwits reported, citing CNBC. The firm noted that Muse’s daily active users were not growing as quickly as downloads, according to the report.
Muse is still capturing around 300,000 new downloads in the US per day, but daily active users aren’t scaling as quickly, with 14-day retention at about 20%, Morgan Stanley said, as reported by Stocktwits. The firm said the trend was unsurprising given the early stage of the agentic AI market, but that Muse would need continued productinnovation to develop more use cases and daily user habits.
The two figures for Day 14 differ: about 20% in Stocktwits’ account of the note, and 23% on the chart that @intelfabs credits to Morgan Stanley. We could not read the Morgan Stanley note itself.
How It Compares With Other Retention Data
Sensor Tower data published earlier this month gives a similar picture from a different angle. About 73% of people who installed Muse returned the following day, according to September 2026 Sensor Tower data shared on X by Menlo Ventures partner Deedy Das, eWEEK reported on 14 October. By Day 14, 26.8% of new ChatGPT users were returning, compared with 22.5% for Muse, eWEEK wrote, and ChatGPT overtook Muse between Days 8 and 9.
Das himself wrote that Muse’s retention is “almost as good as ChatGPT,” and that it is “Not quite in the elite D14 tier” of WhatsApp at 67%, YouTube at 59%, Instagram at 54% and TikTok and Snap at 37%.
In the US alone, Muse had a 30% retention rate after 7 days and 23% after 14 days, according to Sensor Tower data reported by TechCrunch on 20 October, against an average of 19% and 16% for ChatGPT, Claude and Grok.
As we read it, which comparison you pick decides the story. Against other AI assistants in the US, Sensor Tower’s 23% looks strong; against Meta’s mature social apps, the same Day 14 number looks low. The 73% on the chart matches the next-day figure eWEEK reported from Sensor Tower data, and the 23% matches the US 14-day figure TechCrunch reported, but we could not confirm which data Morgan Stanley used.
Connecting the Coverage
Our Muse downloads piece on 20 October covered TechCrunch’s report, citing Sensor Tower, that Muse had 10.9 million downloads since its September debut, 10.2 million of them in the United States, and that its daily active users reached 2.5 million for the first time on 16 October.
Our Muse share-links piece covered Meta pausing the feature that lets people share Muse conversations after the security firm Claroc reported that more than 6,800 share links could be found through public DNS records. Our Meta trust week piece counted five dated trust events across Meta’s AI products in the eight days to 20 October.
Our Grok Bot piece covered SpaceX giving 800,000 paid weekly users for Grok Bot on its 20 October call, and an X post by Michael Sikand that attributed a 23% 14-day retention figure for Muse to Morgan Stanley. What is new today is the chart itself, with the Day 1 figure beside the Day 14 one.
The Structural Read
As we read it, the comparison set decides the verdict. Sensor Tower data reported by TechCrunch put Muse’s US 14-day retention at 23% against a 16% average for ChatGPT, Claude and Grok, while Deedy Das listed Instagram at 54% on Day 14.
As we read it, Morgan Stanley’s own framing points to product, not distribution: the firm said Muse would need continued productinnovation to develop more use cases and daily user habits.
As we read it, the gap between Muse’s 300,000 daily US downloads and its slower-growing daily active users, as Morgan Stanley described it, is the figure to compare with anything Meta says about Muse in its results.
@intelfabs on X, 20 October 2026
“Day-one adoption isn’t the problem. Retention is.”
Three Implications
FOR META INVESTORS Stocktwits reports that Meta reports third-quarter results after the close on Wednesday, and that investors will be watching for updates on Muse adoption, retention and monetization.
FOR AI APP BUILDERS eWEEK reports that ChatGPT started behind Muse on next-day retention but overtook it between Days 8 and 9, with 26.8% returning by Day 14 against 22.5% for Muse.
FOR BUSINESSES TRYING MUSE eWEEK notes that the retention figures alone cannot show task accuracy or time saved, only how often new users came back.
The framework says that in any product cycle there are two curves to follow: “The first is what your underlying technology can do. The second is what your users can actually do with it.”
As we read it, Muse’s retention figures measure the second curve: how many new users keep finding a reason to come back once the novelty of Day 1 has passed. The framework itself does not discuss Muse.
What Is Not Established
We could not read the Morgan Stanley note, so the 73% and 23% figures are as charted by @intelfabs, and the about 20% figure is as Stocktwits reported it from CNBC. The chart does not say which markets, dates or data provider its figures cover, and it gives no values for Instagram, Google or Facebook. Meta has not published its own retention figures for Muse in the sources we read.
Business Engineer Framework
The Product Overhang Doctrine
A Business Engineer analysis of the gap between what AI can do and what users can actually do with a product.
A chart credited to Morgan Stanley puts Muse at 73% active on Day 1 and 23% by Day 14, and Morgan Stanley said Muse needs more use cases and daily user habits. As we read it, retention, not downloads, is the Muse number to watch when Meta reports today.
A note on sourcing. This piece is built on a chart posted by @intelfabs on X on 20 October 2026, which credits a Morgan Stanley note; on Stocktwits’ report of that note, which cites CNBC; on eWEEK’s report of Sensor Tower data and Deedy Das’s own post; and on our earlier sourced coverage. We could not read the Morgan Stanley note itself, and the retention figures come from third parties, not from Meta. We did not contact Meta. Nothing here is financial or investment advice.
Gennaro Cuofano is a CRO and tech executive who has worked in AI since late 2015, starting with bringing NLP, natural-language generation, voice and chatbot products to the marketing industry. His background is in law and finance: he holds a Master's degree in Law and an International MBA with an emphasis on corporate finance (LUISS Business School and the University of San Diego, 2012), and worked as a financial analyst at a real-estate investment firm in San Diego and as an assistant controller. His work focuses on business model strategy, business engineering and, more broadly, structural analysis: how companies make money, read from their own filings. He created FourWeekMBA and leads research there and at The Business Engineer, his newsletter on AI and business strategy, with over 95,000 subscribers and more than 1,000 published analyses. His writing has also appeared on Entrepreneur (for example, How to Digitize a Family Business in Europe) and HackerNoon. Find him on LinkedIn and Substack. See how we source.
Scroll to Top
Discover more from FourWeekMBA
Subscribe now to keep reading and get access to the full archive.
