When Siloed Marketing Teams Raise Acquisition Costs: The Price of Anarchy Inside a Single Advertiser

Firms that divide customer acquisition among product, regional or channel teams can see the total cost per lead rise even when each team meets its own target. The subject of the study is the customer acquisition system of a single advertiser whose teams draw leads from a shared pool of prospects, and the gap between the equilibrium reached by independent teams and the cost-minimizing allocation. The purpose of the paper is to specify when this situation forms a congestion game, to quantify its coordination cost and to translate the known remedies of such games into advertising management. Methodology. The study applies the theory of atomic splittable congestion games. The model assumes homogeneous, deduplicated leads and fixed quotas, with the volume of each team raising the cost of comparable leads for the others; because Google Ads keeps campaigns within one account from bidding each other up, the cost increase comes from depleting the shared pool rather than from auction prices. A closed-form benchmark is derived analytically and checked by numerical simulation, and two conditional extensions draw on Braess's paradox and on the bullwhip effect. Results. The paper introduces the Funnel Coordination Gap, the share of acquisition cost that coordination can avoid. A single team produces no gap and each additional team widens it: in the benchmark, ten teams spend 22% more than a coordinated firm, the avoidable share approaches 25% as teams multiply, and independent teams take up to twice the cost-minimizing volume from the shared pool. Known bounds for this class of games cap the avoidable share at one quarter for identical teams and at one third for teams with different quotas. A personal internal charge per lead, an established toll mechanism, makes the cost-minimizing allocation an equilibrium. Under stated conditions, a costless link between funnel stages can raise total cost by one third, and lower bounds on lead variance multiply across stages unless end-customer conversion data are shared. Conclusions. The internal team structure of an advertiser is a measurable source of acquisition cost, and its size depends on the number of teams and on how sharply the cost of the shared pool rises with volume. Practical implications. Managers obtain a diagnostic of team structure, an internal price for shared sources, a holdout test for channel removal and a controlled experiment that could falsify the central assumption. Value / originality. The paper contributes the conditions under which acquisition inside a firm is a congestion game, a closed-form advertising benchmark placed against exact bounds, and a translation of known pricing mechanisms into management instruments.