Next-generation sequencing (NGS) assays for HLA typing are rapidly being introduced into clinical practice. For common NGS assays, sequenced libraries consist of a mixture of labeled DNA fragments from multiple amplicons and samples. During our validation of a commercially-available NGS assay, we found sample and locus representation varied considerably, compromising data quality. In order to ensure even amplicon and sample loading, we optimized the locus and sample pooling steps.The NGSgo HLA typing kit (GenDx) was used to type the HLA-A, B, C, DRB1, DRB3/4/5, DQB1, DQA1, DPB1 and DPA1 loci, with modifications to the manufacturer’s protocol described below. After PCR amplification, each amplicon was quantified using Quantifluor One dsDNA dye (Promega). Equimolar amounts of each amplicon from a single sample were pooled, and fragmentation, adapter ligation, and indexing steps were performed per the manufacturer’s protocol. Before pooling to a single library, each patient sample was quantified by twoKAPA Library Quantification (Kapa Biosystems) for the number of viable DNA libraries, and TapeStation gel electrophoresis (Agilent) for fragment size measurement. These two metrics were used to estimate equimolar pooling of each patient sample into a final library pool before being loaded onto a MiSeq (Ilumina) for sequencing.After introduction of quantification of all amplicons and equimolar pooling, locus representation on a per-sample basis was much more even; loci ranged from 0.24 to 3.94 times ideal representation using the manufacture’s protocol, but only 0.70 to 1.5 times ideal representation after equimolar pooling. Likewise, sample-level variation in pooling was extreme using the manufacturer’s protocol (ranging from 0.26 to 3.39 times ideal representation), and was much more even after addition of KAPA and TapeStation quantification of individual samples (ranging from 0.76 to 1.19 times ideal representation).Using the modifications described here, we were able to obtain evenly-loaded samples, which allowed for the use of smaller, less expensive flow cells, or the use of a shared portion of a larger flow cell. Together, these modifications enhance the data quality for all patient samples while reducing the overall cost of the NGS assay.