by Oz Etzioni, CEO and Co-Founder of Clinch
Media planning and buying are becoming more automated, more intelligent, and more capable of acting on signals in real time. The market is rightly excited about the opportunities that come with transparent bidding and in-fight optimization.
All this intelligence and automation lets large brands especially, run dynamic, personalized campaigns. National retailers, auto manufacturers, CPGs are deploying campaigns tailored to specific markets and audiences; often producing hundreds of creative variations in support.
But that scale also puts pressure on how those campaigns are executed. Managing hundreds of assets across different platforms quickly exposes the limits of a fragmented system. What works at a smaller scale starts to break down under the weight of complexity.
The real challenge in advertising is everything that happens before the bid. Autonomous media buying will only live up to its promise if the industry finally addresses the messy infrastructure of the campaign lifecycle.
The industry is mistaking the final step for the whole system
There is a growing assumption in adtech that once media buying becomes autonomous, the rest will naturally fall into place. Media buying attracts attention because it’s the clearest expression of autonomous advertising; it is the part companies can point to and say, “Look, the machine is working.” But buying an impression is not the same thing as running an effective campaign. Media buying is the final step, and treating it as the core of autonomous advertising overlooks the many steps that need to happen before it can deliver on that promise.
Long before a bid is placed, brands must generate creative at scale, align messaging to specific KPIs, adapt assets for different formats and placements, personalize content across audiences and contexts, and ensure the creative that ultimately runs reflects both the brand’s goals and the realities of the media environment.
That is where the real operational complexity lives. It is not glamorous, but when it isn’t working, you really notice. When those inputs are misaligned, automation at the buying layer has limited impact. Autonomous media buying can optimize what to buy downstream, but it cannot compensate for a disconnected system upstream.
The infrastructure before the bid is still fragmented
This is the contradiction at the heart of modern advertising. Media systems are becoming more intelligent, but the execution layer surrounding them remains highly manual. Creative, media, and data still live in separate environments, and information moves between them through a mix of spreadsheets, approvals, and human workarounds. Even sophisticated advertisers are still dealing with fragmented workflows at exactly the moment when complexity is increasing.
And complexity is increasing quickly. AI is accelerating creative production and omnichannel execution is now table stakes. In short, global brands are managing campaigns across more markets, audiences, formats, and variables than ever before. So while the industry celebrates smarter bidding, the operational layer connecting all of that activity is under more strain. Execution is now the bottleneck.
Autonomous advertising needs an operating system
If advertising is becoming a continuous, adaptive system, the underlying system has to support that. It needs an environment where creative, data, campaign logic, and activation can function as part of one coordinated workflow rather than as a collection of disconnected tasks. That is the missing layer in much of today’s conversation around autonomous advertising.
What’s truly needed is comprehensive infrastructure that handles everything from point zero through the moment the impression is bought. That includes creative at scale, messaging logic, workflow coordination, omnichannel activation, and the feedback loops that allow campaigns to improve over time.
Media buying is becoming autonomous. That is real progress. But the industry should not confuse one important layer with the full system. The next phase of advertising will not be defined simply by who automates the bid best. It will be defined by who builds the infrastructure that makes autonomous advertising function seamlessly.
