Retrofit of existing steel buildings often requires strengthening of the connection regions. One common connection, the bolted beam-column connection, is often strengthened in design using stiffened extended endplates, or with continuity plates welded between the column flanges. In a retrofit scenario, adding stiffeners to the endplate is difficult due to the concrete slab and metal deck, and excessive field welding of continuity plates may be uneconomical. Simplifying retrofit efforts, and for economy, connection strength may be improved by simply adding more bolts to the connection. Current code methods, broadly generalized to all connection configurations, are currently based on component experiments having only one bolt on either side of the column web. This study experimentally investigates strengthening of bolted beam-column connections, having no column web stiffeners, using more than one bolt on either side of the column web. Six full-scale bolted beam-column connections are tested, representing exterior beam column connections (beams attached to only one column flange). Connections with both extended and flush endplates are considered. Two column sections (HE300A and HE300B) are tested along with HE300B beams creating both equal-column-beam, and weak-column strong-beam scenarios. Analytical simulations provide insight into local connection demands, and experimental results are compared with current code methods. The experiments indicate that closer inner-bolt spacing relative to the column web increases connection moment capacity but decreases rotation capacity (connection ductility) due to increased bolt prying forces from column flange distortions. The outer bolt of multiple-bolt-per-row configurations contributes very little to the connection resistance when column web stiffeners are not considered. With the exception of specimen T-3B which failed through bolt-thread shear after 0.02 rad, all connections with multiple bolts per row still achieved rotations greater than 0.06 rad. The Eurocode 3 component method and adapted Eurocode 3 procedures conservatively predicted the connection strength of each test specimen, including weak-column strong-beam assemblies, and accurately identified the initial connection limit states.