Modern digital advertising systems rely on two primary targeting mechanisms: keyword targeting, where advertisers manually specify search terms, and item-based targeting, which matches ads to buyer queries through semantic similarity. While both approaches are effective in isolation, they leave a coverage gap, particularly for broad queries and new items, where item-based targeting often underperforms. This paper presents a novel hybrid targeting framework that bridges this gap by transferring successful query–item associations from keyword targeting into item-based campaigns via semantic embeddings. Our method leverages historical query–item pairs to create robust query representations and applies similarity-driven transfer to expand the candidate query set for item-targeted ads. To balance coverage and quality, we introduce adaptive thresholding and evaluate multiple query selection strategies. Experimental results, including large-scale A/B tests on eBay's advertising platform, demonstrate that our approach significantly expands impression coverage for broad queries while maintaining relevance. The best-performing configuration selected a small set of high-quality queries per item combined with an adaptive relevance threshold, achieving over 5% incremental coverage while preserving click-through rate (CTR) and conversions-over-impression (CVI) performance. These findings highlight the effectiveness of combining the breadth of keyword targeting with the precision of item-based targeting, offering a scalable solution to the coverage gap in ad marketplaces.