Meta Muse for Small Business has been live since 29 September and promises you an intern who never sleeps. The AI agent reads your Instagram insights, ads data, and Shopify figures, writes drafts, and suggests campaigns. The catch for you: at launch, it is only available in the US and Canada. For marketing professionals in Austria, however, that is no reason to sit back and do nothing.

What can Meta Muse for Small Business actually do?

Muse is Meta's personal AI agent, introduced on 8 September. The business version evaluates sales, campaigns, and social media activities, drafts growth plans from them, identifies conspicuous expenses, and prepares email responses (Help Net Security). To do this, Muse connects to Instagram business accounts, Facebook pages, and Meta advertising accounts, along with tools such as Canva, Shopify, Klaviyo, Slack, Stripe, Notion, and QuickBooks (Meta). In the process, Meta learns what you sell, what you sound like, and what your customers constantly ask (TechCrunch).

In concrete terms for social media, this means Muse deduces from your own metrics which posts and ads have worked recently and builds new campaign drafts based on that. The routine work—from reporting to initial copy suggestions—thus shifts to the agent. What remains for you is what no agent delivers: values, timing, and an intuition for what suits your brand.

The safety leash holds firm. According to Meta, nothing is published, sent, or spent without your approval. In plain terms: the agent drafts, you decide. Most features are free; for extended functionality, subscription plans are available, costing up to 100 US dollars per month according to MediaPost.

A day earlier, on 28 September, Meta also launched the Meta Enterprise Platform featuring Muse Agent, Meta Business Agent, Muse API, and Muse Code. Meta is thus simultaneously addressing both small and large players. Anyone who views social media merely as a posting channel has missed the playing field.

Why is Muse (not yet) available in Austria?

At launch, Meta only mentions the US and Canada (Meta). As far as we know, there is no official European release date. schieb.de cites GDPR, the AI Act, and the Digital Markets Act as points of friction. A look back at history helps provide context. Meta AI launched in September 2023 in the US and, according to this overview, did not arrive in the EU until March 2025—around 18 months later. Plan in months, not weeks. However, the fact that the icon is already visible in Instagram suggests it might arrive sooner rather than later.

Austria will not fall behind because of this—on the contrary. 30 percent of domestic companies with 10 or more employees used AI in 2025, compared to an EU average of 20 percent (Statistics Austria, 24 June 2026). Since 2021, the figure has risen from 9 to 30 percent. That is lived reality here. Among companies considering AI, 15 percent cite a lack of know-how, and 11 percent each cite data privacy and legal uncertainty as roadblocks. That is precisely where your competitive edge begins.

"The agent comes later. You need your database now."

— Olivia Ulbing-Sommeregger, Founder oha! digital

What does this mean strategically for SMEs and large companies?

For SMEs, Muse is the answer to Meta's own diagnosis. According to Meta, small businesses don't lack ideas; they lack hours. The workflows behind it also exist without Muse: ChatGPT, Gemini, Claude, and co., plus Canva, Klaviyo, or Zapier create similar shortcuts—just without direct Meta data access to insights and ad accounts.

In addition, there is a social media effect that many underestimate. If every brand generates drafts using the same agents, feeds will become even more interchangeable. The accounts that succeed will be those that bring a distinct voice, genuine human faces, and proprietary data. The tools will become identical; your material will not.

Large companies should view the Enterprise Platform as a signal. According to Zuckerberg, the company that helps "hundreds of millions of businesses" connect with customers now wants to add its next business pillar. Those who do not establish clear rules for approvals, data protection, and AI disclosure now (see our article on the AI Act) will slow themselves down later.

The bold thesis: an agent is only as smart as your setup. Garbage in, garbage out. That is precisely what we at oha! are currently cleaning up with clients before agents are granted access.

How to get your marketing agent-ready in 5 steps:

1. Declutter accounts. Ensure your Instagram business account, Facebook page, ad account, and Business Manager are cleanly linked, roles are clarified, and legacy dead weight is removed.

2. Check tracking. Pixels, Conversions API, and UTM logic must be accurate; otherwise, the agent will analyse numbers that lie.

3. Document your brand voice. Put tone of voice, no-gos, top-performing posts, and the ten most common customer questions into one document. That is fuel for any agent.

4. Define approval rules. Who is permitted to do what without a human in the loop? Account for data protection and AI labelling right from the start.

5. Test a workflow. 30 days, one use case (for example, content drafts or reporting), one KPI such as hours saved or cost per lead. Large brands should start as a pilot with clear governance.

Those who have ticked these boxes can simply turn it on the day of the EU launch, rather than scrambling to clean up first. The side effect is almost more valuable: clean accounts and robust metrics improve your campaigns today, regardless of whether an agent is reading along. That is precisely why we at oha! always start projects on AI and performance marketing with this foundation.

Want to know where your setup stands? Send us a quick enquiry, and we will take a joint look at your accounts, tracking, and the right AI workflows for your team.