From the experimental evaluation of the item-item collaborative filtering scheme we make some important observations. First, the item-item scheme provides better quality of predictions than the use-user (k-nearest neighbor) scheme. The improvement in quality is consistent over different neighborhood size and train/test ratio. However, the improvement is not significantly large. The second observation is that the item neighborhood is fairly static, which can be potentially pre-computed, which results in very high online performance. Furthermore, due to the model-based approach, it is possible to retain only a small subset of items and produce reasonably good prediction quality. Our experimental results support that claim. Therefore, the item-item scheme is capable in addressing the two most important challenges of recommender systems for E-Commerce-quality of prediction and high performance.