Programmatic advertising: connected TV, display and video

The agency doesn't set it; the placements and deals do. Premium streaming inventory and direct deals with publishers usually have their own minimums, and below them there is no campaign. We check this during planning, with the specific placements in your plan, and tell you before you commit to anything.

The inventory is largely the same; what changes is control and visibility. In DV360 you select placements one by one, apply allowlists and blocklists, control frequency across all of the campaign's channels and see the actual split of impressions by placement. For small accounts that difference doesn't justify the complexity; above a certain volume, it does.

Yes. The report includes the split of impressions by placement, so you see where each impression was served and what each environment cost. It's one of the reasons to buy programmatically rather than on an automated network, where that detail isn't disclosed.

It's measured by what happens next, not by the click. The three signals we watch are visits from the households reached, the trend in branded searches while the campaign is on air and view-through conversions within a window agreed before launch. When volume allows, we propose an incrementality test with a control group, which is the only thing that measures the real contribution.

It's the classic programmatic risk, and it's controlled before, not after. We work with reviewed placement allowlists, category exclusions and environment verification. On CTV and in direct deals the problem practically doesn't exist, because the inventory is closed in advance.

By reach and frequency over the area, not by direct attribution. Out of home there's no cookie or identifier, so nobody can tell you which sale came from which screen. What you can do is plan by impressions over a specific area and compare its behavior with an equivalent area without the campaign.