Geo-Conquesting DOOH Strategy and Screen-Level Buy List

The deliverable is a screen-level buy list: panel and venue IDs, OpenOOH taxonomy codes, the deal or direct contract each sits behind, daypart, and the catchment each panel is there to cover, plus a set of catchments deliberately left unbought so the campaign has a control. It arrives as a spreadsheet and a map, and it is written to be executed by whoever you choose, including a buyer who is not us. No media is bought inside this engagement. How the catchments are built: Isochrones on the real road and pedestrian network rather than radius rings. An eight-minute drive or a six-minute walk traced with OSRM or Valhalla over OpenStreetMap data, around competitor sites verified by hand against Overture Maps and the operator's own location list. Radius rings put screens across a river or a rail cut and count them as reachable. On a Prague trade area that difference has been about a third of the panels. Overlap and cannibalisation: Every conquesting catchment is intersected with the catchments of your own locations. Panels inside both are flagged, because spend there mostly reaches people you already have, and in a gym or clinic network that overlap is the single largest source of wasted screen budget we find. The list separates conquest panels, defence panels and the overlap, and prices each of the three. The holdout: Matched catchments, matched on population density, competitor presence and your own pre-period performance, are left out of the plan. The readout runs as a geo experiment using GeoLift or a Bayesian structural time series, and we state the minimum detectable effect before the buy, so an inconclusive result is a known risk rather than an argument in month three. What we refuse: We do not buy device identifiers observed inside a competitor's premises, and we do not build audience segments from location data collected at gyms, clinics, pharmacies or places of worship. Under GDPR and the ePrivacy regime that inference is special category data in all but name, and the fact that a vendor will sell it to you does not make it lawful for you to use. We will say this in front of your legal team, and we would rather lose the brief than argue the point. Mobility panel data has one use here: confirming a catchment carries the footfall we modelled, before anything is booked. Not included: Media buying and execution, which is a separate engagement and one you are free to place elsewhere. Creative. Brand positioning. Competitor pricing or product analysis. Legal opinion of any kind. We flag the risk, your counsel rules on it. Who this is not for: Single-location businesses, where the analysis is a walk around the block and a conversation with the landlord. Advertisers with fewer than about twelve locations, where the holdout cannot be constructed at all. And anyone who wants a share-of-voice figure against a named competitor, a metric that flatters whoever commissioned it and that we do not produce. How the work runs: Two working sessions, the second spent on the contested catchments panel by panel with the reasoning written down beside each decision. The list is versioned, so when a network pulls inventory mid-flight you can see what replaced it and on what grounds. Everything is handed over in open formats, GeoJSON and CSV, because a buy list you cannot open without us is not a deliverable.