Unilever uses AI in social media by bringing content production in-house and making it digital-first. Its Sketch Pro studios, built with IPG Studios for the Home Care division, produce social-first assets, and its digital product twins, built with NVIDIA Omniverse, cut cost and turnaround time for brands such as Dove and TRESemmé. This article covers what Unilever has reported and what brands can learn from it. For the wider discipline, see EPR's guide to AI communications.
Social media marketing has reached a saturation point. Every brand is posting, boosting, reacting, and optimizing, yet much of it blends into an indistinguishable stream of content. The differentiator is no longer presence. It is precision.
From Content Calendars to Content Intelligence in AI Social Media
Most brands still operate on static content calendars: pre-planned posts distributed across platforms with limited adaptability. That model is misaligned with how social platforms work, and Unilever has been moving away from it. Marketing Dive reported that Sketch Pro is meant to help Unilever shift from a linear, TV-first production model to social-first storytelling, and that Unilever plans to put half of its marketing spend into social while broadening its work with influencers.
That shift changes the planning question. Instead of asking "What should we post this week?" the question becomes:
- What content is performing right now?
- What formats are gaining traction in specific segments?
- How should creative adapt in real time?
Brands that do this well use performance data to see which visual styles, messaging tones, and formats (short-form video, static, interactive) are working, then iterate continuously rather than wait for post-campaign analysis.
Creative at Scale, Not Generic at Scale
One of the biggest misconceptions about AI in marketing is that it leads to generic output. Sketch Pro shows the intended alternative. It is an in-house design network that, according to TechInformed, was live in London, São Paulo, Mumbai, and Jakarta, with plans to reach 21 markets by 2026. Marketing Dive adds that it is not tied to a single AI platform and that its toolset includes Google's Veo 3 video generator.
The point of a network like this is localization. A single campaign concept can be adapted for:
- Geography
- Language nuances
- Audience behavior
- Platform-specific dynamics
The key is not automation alone. It is structured experimentation. AI shortens testing cycles, so marketers can find what resonates and scale it.
Social Listening as a Strategic Input
Public reporting on Unilever's setup focuses on production and spend rather than listening, so this section describes the practice brands can pair with it. AI-driven social listening has moved beyond basic sentiment analysis. It can surface emerging behaviors, preferences, and cultural shifts, which means:
- Product messaging can adapt to real-time conversations
- Campaigns can align with emerging trends before they peak
- Risks can be identified early
If a product feature begins trending organically, a fast production system can amplify it with aligned content immediately. That is proactive alignment with audience momentum, not reactive marketing.







