Ad Verification Proxy

An ad that renders correctly from the office tells you very little. What matters is the creative served to a real visitor in the market you bought.

What Goes Wrong After Booking

Creatives get swapped, placements get substituted, geo-targeting misfires and redirect chains pick up intermediaries nobody agreed to. None of this is visible in a reporting dashboard, which describes what was billed rather than what was displayed. It is also invisible from a corporate connection, because ad systems commonly treat datacentre ranges as non-human and serve them differently or not at all.

Checking Properly

Load the placement from a residential address inside the target market. Capture the creative, the surrounding placement and the full redirect chain — the chain is where most substitution actually happens and it is the part usually skipped. Sample repeatedly rather than once, because rotation means a single check can easily land on the correct creative while the problem sits in the other slots.

Comparing Against The Booking

Check against what was ordered, not against last week's capture. Drift compared with yesterday looks like normal rotation; drift compared with the insertion order is the finding. Keep the market, timestamp and address with each capture so a discrepancy can be evidenced later.

Handling A Discrepancy

When a capture does not match the booking, collect two or three more before raising it. Rotation means a single mismatch can be an unlucky sample rather than a systematic problem, and a report built on one observation is easy to dismiss. Once the pattern is established, present the captures with markets and timestamps attached — that is the form the conversation with the publisher will need anyway.