French startup Concord has raised $3 million from Drysdale, Motier Ventures, Better Angle and A16Z Scout. Several adtech executives also took part, including Philippe Corrot of Mirakl, Rémi Lemonnier of Scibids, Arthur Querou of Vibe.co, Thomas Zaepffel of Adot and Christophe Pottier.
After transforming content creation, targeting and data analysis, artificial intelligence is now taking on one of the last bastions of manual work in digital advertising: campaign execution. That is the opportunity Concord is targeting.
Concord is developing an agentic infrastructure dedicated to media buying. Its founders are Nicolas Cosson, Antoine Chwalek and Nathan Venezia, who previously founded adtech Manadge and sold it to iHeartMedia.
The ambition sounds simple: enable AI agents to handle a large share of the operational tasks currently performed by traders and advertising-platform specialists. Priced as a small percentage of media spend, the service is arriving at a pivotal moment for the industry.
For the past 15 years, advertising platforms have steadily automated campaign optimization. Google, Meta, Amazon and TikTok now handle an increasing share of decisions related to bidding, targeting and delivery.
Yet despite this growing automation, operational teams still spend a significant amount of time on repetitive tasks: building campaigns, configuring audiences, managing naming conventions, checking budgets, producing reports and performing daily optimizations.
The paradox is striking. While algorithms already make many of the buying decisions, orchestration across platforms still relies heavily on human intervention. This execution layer is precisely what Concord aims to automate.
The company connects to the market's leading platforms, including DV360, Meta, Amazon Ads, TikTok and Google Ads, with Snapchat coming soon. Users can move from a brief to an activated campaign through a conversational interface.
At first glance, the sales pitch sounds like a familiar productivity promise. Concord says its users can reduce the time spent on campaign execution by as much as 70%. It already works with advertisers such as department store group Galeries Lafayette, holdcos and independent agencies including Haiku and Goodbuy Media. But its conversations with those agency groups suggest the issue goes far beyond productivity.
For several years, major media agencies have been trying to absorb the growth in digital media investment without increasing their headcounts at the same pace. To achieve this, they have relied heavily on offshoring, using service centers located in countries such as India, the Philippines, Morocco and Portugal.
This model reduces costs, but it is gradually reaching its limits. Campaign launch times are getting longer, processes are becoming increasingly standardized, and the growing complexity of advertising platforms requires ever more coordination.
"You save money, but you do not gain flexibility," says Nathan Venezia.
In this context, some agencies see Concord as a technology-driven alternative to offshoring. The goal is no longer simply to reduce execution costs, but to increase teams' operational capacity.
Venezia says the tool would let a trader currently managing €3 million in media spend oversee €5 million to €6 million. That frees up time for strategy and client relationships.
Another benefit of these agentic approaches is their ability to apply the same level of granularity to every campaign. In practice, agencies often focus their resources on their largest budgets. Local campaigns and lower-spend initiatives inevitably receive less attention because teams lack the time to manage them with the same degree of care.
Automation changes that economic equation.
A €1,000 campaign and a €1 million campaign can benefit from the same level of setup, oversight and optimization without creating any additional operational workload.
That is probably one of the most compelling opportunities in this new generation of tools. It is not just about working faster. It makes a level of sophistication once reserved for the largest media buyers economically viable across every campaign.
That proposition is likely to appeal to media agencies. They can use Concord "either to do more with the same number of people, or to do the same amount of work with fewer people," Venezia says.
That does not mean allowing an LLM to manage advertising campaigns on its own, however.
"We quickly identified the limitations of generative models when it comes to analyzing large volumes of data or making critical decisions," Venezia explains.
Concord's answer is to combine conversational agents, proprietary algorithms and deterministic rules. AI models primarily act as orchestrators and user interfaces, while calculations, analysis and control mechanisms remain managed by more traditional software layers.
The funding round comes as advertising platforms themselves begin deploying their own agentic infrastructure, particularly through MCP servers and conversational interfaces. To provide more comprehensive coverage, Concord plans to connect its platform to The Trade Desk, LinkedIn and Pinterest by the end of the year.
The company also intends to expand internationally very quickly.
"We are considering the best way to enter the U.S. and U.K. markets," Venezia says.
