네이버 검색광고 성과형 입찰 자동 최적화 시스템 구축

초록

This study designs and implements a performance- based automatic bidding optimization system using the Naver Search Ads Flatform to address the issues of rank stagnation and excessive bidding competition prevalent in traditional manual, rank-based bidding systems. We conducted an experiment with "Children's Observatory" under severe constraints, including keyword limitations, the inability to track conversions, and a limited daily budget. A dual- strategy approach was employed, simultaneously operating an impression-focused strategy and a click-efficiency-focused strategy. By setting distinct Key Performance Indicators such as Cost Per Mille and Click-Through Rate for each team, we applied automation to manage both budget and efficiency. The results empirically demonstrated that even in such a constrained environment, the proposed automated system can achieve significant performance improvements. Specifically, Team A increased impressions by over 205%, while Team B improved click rate by approximately 70% compared to the previous manual bidding method. This study presents a practical, data-driven automated advertising operation model applicable to the domestic search advertising market, especially for small to medium-sized advertisers facing similar limitations.