Published on the 10/06/2026 | Written by Heather Wright
Turning chat into checkouts…
Meta has entered the enterprise AI race, with the launch of an agent designed to help companies manage customer interactions, sales and operational workflows across Meta’s messaging platforms.
Unveiled at a company conference in London, the Meta Business Agent operates across WhatsApp, Messenger and Instagram, responding to customer enquiries, recommending products and booking appointments inside chat conversations. Meta says the agent can also qualify sales leads and complete transactions.
“The agent can also qualify sales leads and complete transactions, extending its messaging tools into revenue‑generating workflows.”
While Meta is dominant in online ads which account for the vast majority of its revenue, it has struggled when it comes to actually selling its products. The company’s messaging ecosystem, however, is already used as a communications layer for many businesses of all sizes.
Meta has famed the new offering as part of a broader enterprise push, with the company saying the goal is to provide businesses of any size with an always-available digital agent to handle customer engagement and operational tasks.
The company is also launching a ‘Business Agent Platform’ to allow organisations to build and customise their own AI agents, including integrations with third-party systems such as Shopify, Zendesk and Shopee.
“The platform provides larger businesses with enterprise-grade controls, guardrails and measurement built in so they can define rules and offer personalised experiences, starting with the messaging apps their customers already use,” Meta says.
The move also hints at Meta’s AI ambitions and desire to compete with the likes of OpenAI, Anthropic and Google with enterprise versions of their AI tools.
It also comes as competition intensifies in the market for enterprise AI agents. Companies including Google, OpenAI and AWS are developing tools designed to automate customer interactions and business workflows, targeting similar use cases in sales, support and operations.
The Business Agent places Meta directly into this category, with executives explicitly positioning it as an enterprise play, rather than an incremental update to existing messaging products. The launch builds on earlier chatbot capabilities, but extends them to enable automated actions, aligning with broader industry shifts towards ‘agentic’ AI systems that can take on tasks, rather than simply respond to queries.
Unlike some competitors, Meta’s strategy is to deploy the capabilities inside its existing messaging platforms, where businesses and consumers are already interacting at scale.
Meanwhile the Business Agent Platform places Meta in competition with vendors offering standalone AI agent frameworks and enterprise automation platforms. (As an aside, the Business Agent launch came just days after Meta floated the idea of becoming a hyperscaler and renting out its compute infrastructure.)
Chatting with customers, closing sales
A free test version of the Meta’s agent service was released in select markets, including Mexico and India, late last year under the ‘Business AI’ name.
At the launch of Business Agent, Meta said more than one million businesses were already using a Meta Business Agent – presumably the free test offering – on WhatsApp and Messenger.
The new Business Agent provides automated capabilities to allow companies to handle customer conversations at scale. It’s designed to answer questions specific to a business, make product recommendations from business catalogues, book appointments and qualify incoming leads, close sales and escalate more complex queries to human staff where required – with the business setting the parameters for when a team member will step in.
It’s also offering the ability to generate summaries of chats that occurred overnight and provide insights on threads.
Meta says the agent can be deployed ‘in minutes’ within its platforms or integrated directly into existing enterprise infrastructure and, while initially available for free, will be a paid subscription, believed to be a business-focused subscription tier for Meta One, the umbrella brand for Meta’s paid subscription offers, launched earlier this month. Larger enterprises are expected to be charged based on token usage, similar to existing pricing models for business messaging on WhatsApp.
While the initial rollout focuses on customer engagement and sales tasks, Meta is positioning the Business Agent as part of a broader toolkit for businesses.
It says it is developing additional features that would expand the agent’s role into areas such as conducting market research, surfacing product insights, connecting with tools to manage your calendar and providing competitive intelligence.
Challenges ahead
For enterprises, rolling out autonomous agents in customer-facing channels raises practical implementation questions, particularly around integration, governance and customer experience.
Meta says the agent can connect to third-party tools such as Shopify and Zendesk and can be embedded into existing business workflows.
For enterprises, deeper integration with systems such as CRM and ERP platforms would be required to align the agent with customer records, order management and service processes. Whether businesses are willing to connect those systems to Meta’s platform could be a factor for enterprise adoption, particularly given the sensitivity of enterprise data and existing system architectures.
While Meta brings scale through its messaging platforms, adoption at enterprise level will depend on how effectively those capabilities integrate with existing systems, operate within governance frameworks and deliver reliable customer experiences at volume.
