Social Rocket
More AI output does not automatically mean better marketing quality. At a recent industry event hosted by Digiday, Glossy and Modern Retail, marketers described a growing “good enough” problem: automated workflows speed up production, but brand standards, clean controls and strategic priorities risk being pushed aside.
Some of these statements come from a discussion held under Chatham House rules and should therefore not be read as a representative study. They do, however, point to a concrete operational problem that many social, creative and performance teams now have to solve: how can AI be scaled without diluting the brand?
Human Review Becomes the Quality Filter
One participant from the premium beauty sector explicitly described AI-generated creatives as rough drafts that must be reworked by people before publication. Other participants reported campaign material where even basic brand elements such as logos had not been implemented correctly.
For marketing teams the consequence is simple but important: automation needs a binding approval stage. This applies especially to publicly visible assets, paid campaigns and content containing brand claims or sensitive product information.
A sensible workflow therefore separates at least three steps: clear guidelines before generation, automated production or editing, and a final human quality check. Our practical guide to the AI Creative Director with Claude focuses precisely on such brand rules and approvals.
More Output Can Make Brands More Interchangeable
The quality problem is not only a craft issue. When many companies use the same models with similar prompts, data sources and optimisation goals, communication can become more uniform.
This concern also came up at DMEXCO 2026. There, marketing leaders discussed how efficiency gains from AI can be combined with brand identity, trust and human leadership. The debate is shifting from “Can AI do this?” to “Which parts of the process should we actually hand over to it?”.
This is also relevant for creator and social teams. Platforms do not automatically reward maximum production volume. Recognisability, format understanding, community closeness and a consistent voice remain distinct quality factors.
Measurement Needs to Change Too
Alongside creative quality, participants criticised an overload of metrics. According to Digiday, Isabel Perry of DEPT argued for measuring the gaps in the customer journey more closely, rather than continuing to collect ever more generic metrics.
AI makes the background even more complicated: alongside humans, bots and agents are increasingly accessing content. This changes attribution, retargeting and the interpretation of traffic. Social Rocket has already analysed why bot traffic becomes a retargeting and measurement problem.
AI search works similarly. If you only look at an aggregated visibility score, it is easy to miss whether a brand actually appears as a source, receives referral traffic or drives conversions. That is why AI visibility and AEO measurement should capture several levels separately.
A practical quality standard for AI workflows
For marketing teams, the discussion boils down to a simple principle: AI should increase production capacity, not lower the approval threshold.
Before scaling, brands should therefore define which elements are non-negotiable – such as logo, tone of voice, claims, visual language, source requirements and legal approvals. Only then can you check where automation actually saves time.
This becomes especially important when AI agents also take over media buying and campaign management. According to a recent IAB Europe survey, many industry representatives expect agentic media buying to be in operational use in the short term. The more autonomously systems act, the more important it becomes to define boundaries and traceable controls in advance.
Sources: Digiday: Marketers bemoan the culture of “good enough”, DMEXCO: Scaling Intelligence 2026