Sales Diversity Research Articles

Economic corollaries of personalized recommendations

The impact of recommendation systems (RSs) on the diversity of consumption is not transparent or well understood. Available studies, whether experimental or theoretical, show inconsistent and even opposite results, which manifests as debate in the literature. In this paper, we investigate the impact of two main recommender systems, neural collaborative filtering and deep content filtering, on sales diversity via a randomized field experiment. Our results confirm the capability of recommender engines in increasing or decreasing aggregate sales diversity. Nonetheless, they amplify homogenization and reduce individual-level consumption diversity. In conclusion, our research reconciles seemingly contradict previous findings and illustrates that the design of the RS is the decisive factor in homogenizing or diversifying product sales.

Recommendation technologies and consumption diversity: An experimental study on product recommendations, consumer search, and sales diversity

Research has found two contrasting effects of recommendations on sales diversity: the long-tail effect (enabling niche products to be discovered and purchased) and the rich-get-richer effect (making popular products more popular). Given the mixed empirical findings, however, the literature lacks scholarly efforts to systematically examine when and why recommendations result in certain consequences. To fill this gap, we design a laboratory experiment where participants, divided into two groups with and without product recommendations, engage in online shopping for two types of products, search and experience goods. We find that with recommendations, more diverse products are attended, but less diverse ones are purchased. The sales diversity-reducing (rich-get-richer) effect of recommendations is more pronounced when shopping for the search goods, whereas the search scope-broadening (long-tail) effect of recommendations is more pronounced for the experience goods. We theorize that recommendations not only generate a popularity bias, but also serve a preference-matching role, and the former is more pronounced when shopping for the search goods while the later for the experience goods. Consequently, recommendations result in shorter search for the search goods, but longer search for the experience goods. Our analysis suggests recommendation scope and intensity as useful measures to understand the underlying mechanism. We conclude with discussions on the implications of our results, including potential social costs of recommendation technologies such as distortions or biases in market demand.

A factual and perceptional framework for assessing diversity effects of online recommender systems

Purpose The purpose of this paper is to explore the effects of online recommender systems (RS) on three types of diversity: algorithmic recommendation diversity, perceived recommendation diversity and sales diversity. The analysis distinguishes different recommendation algorithms and shows whether user perceptions match the actual effects of RS on sales. Design/methodology/approach An online experiment was conducted using a realistic shop design, various recommendation algorithms and a representative consumer sample to ensure the generalizability of the findings. Findings Recommendation algorithms show a differential impact on sales diversity, but only collaborative filtering can lead to higher sales diversity. However, some of these effects are subject to how much information firms have about users’ preferences. The level of recommendation diversity perceived by users does not always reflect the factual diversity effects. Research limitations/implications Recommendation and consumption patterns might differ for other types of products; future studies should replicate the study with search or credence goods. The authors also recommend that future research should move from taking a unidimensional measure for the assessment of diversity and employ multidimensional measures instead. Practical implications Online shops need to conduct a more comprehensive assessment of their RS’ effect on diversity, taking into account not only the effects on their sales distribution, but also on users’ perceptions and faith in the recommendation algorithm. Originality/value This study offers a framework for assessing different forms of diversity in online RS. It employs various recommendation algorithms and compares their impact using not just one but three different types of diversity measures. This helps explaining some of the contradictious findings from the previous literature.

A New System-Wide Diversity Measure for Recommendations with Efficient Algorithms

Recommender systems often operate on item catalogs clustered by genres, and user bases that have natural clusterings into user types by demographic or psychographic attributes. Prior work on system-wide diversity has mainly focused on defining intent-aware metrics among such categories and maximizing relevance of the resulting recommendations, but has not combined the notions of diversity from the two point of views of items and users. In this work, (1) we introduce two new system-wide diversity metrics to simultaneously address the problems of diversifying the categories of items that each user sees, diversifying the types of users that each item is shown, and maintaining high recommendation quality. We model this as a subgraph selection problem on the bipartite graph of candidate recommendations between users and items. (2) In the case of disjoint item categories and user types, we show that the resulting problems can be solved exactly in polynomial time, by a reduction to a minimum cost flow problem. (3) In the case of non-disjoint categories and user types, we prove NP-completeness of the objective and present efficient approximation algorithms using the submodularity of the objective. (4) Finally, we validate the effectiveness of our algorithms on the MovieLens-1m and Netflix datasets, and show that algorithms designed for our objective also perform well on sales diversity metrics, and even some intent-aware diversity metrics. Our experimental results justify the validity of our new composite diversity metrics.

How Do Recommender Systems Affect Sales Diversity? A Cross-Category Investigation via Randomized Field Experiment

Recommender systems appear all across the internet. For e-retailers, this represents an opportunity to get more and niche products before customers’ eyes. However, we find that while implementing recommender systems does increase overall sales figures, it does not generally improve the relative sales for niche items, leading to a rich-get-richer situation. We find, across a wide range of product categories, that the use of traditional collaborative filters (CFs) is associated with a decrease in sales diversity relative to a world without product recommendations. The decrease in aggregate sales diversity may not always be accompanied by a corresponding decrease in individual-level consumption diversity. In fact, it is even possible for individual consumption diversity to increase as aggregate sales diversity decreases. CFs help individuals explore new products, but similar users still end up exploring the same kinds of products, resulting in concentration bias at the aggregate level. There is one insight for management: Traditional collaborative filters carry the unintended consequence of increasing concentration bias. A firm interested in exposing consumers to a broader assortment of products may prefer a different design from another simply interested in maximizing sales.

How Do Recommender Systems Affect Sales Diversity? A Cross-Category Investigation via Randomized Field Experiment

We investigate the impact of collaborative filtering recommender algorithms (e.g., Amazon’s “Customers who bought this item also bought”) commonly used in e-commerce on sales diversity. We use data from a randomized field experiment run on the website of a top retailer in North America across 82,290 products and 1,138,238 users. We report four main findings. First, we demonstrate and quantify across a wide range of product categories that the use of traditional collaborative filters (or CFs) is associated with a decrease in sales diversity relative to a world without product recommendations. Furthermore, the design of the CF matters. CFs based on purchase data are associated with a greater effect size than those based on product views. Second, the decrease in aggregate sales diversity may not always be accompanied by a corresponding decrease in individual-level consumption diversity. In fact, it is even possible for individual consumption diversity to increase while aggregate sales diversity decreases. Third, co-purchase network analysis shows that while recommenders can help individuals explore new products, similar users still end up exploring the same kinds of products, resulting in concentration bias at the aggregate level. Fourth and finally, there is a difference between absolute and relative impact on niche items. Specifically, absolute sales and views for niche items in fact increase, but their gains are smaller compared to the gains in views and sales for popular items. Thus, while niche items gain in absolute terms, they lose out in terms of market share. We discuss economic impact and managerial implications.

Targeted Promotions and Cross-Category Spillover Effects

Targeted promotions based on individual purchase histories are known to increase promotional response while reducing search effort, yet the opportunity costs of such targeted promotions remain poorly understood. We use field experiments on a mobile e-book reading app and an online ticket exchange to test how targeted promotions can decrease cross-category search and purchasing. Findings from three randomized field experiments consistently showed that targeted promotions reduce cross-category purchase activity. These negative spillovers can be large; while direct response was higher for targeted promotions, total sales for one of the experiments was significantly higher for untargeted promotions. We find evidence that the negative category spillover effects are due to a decrease in search activity. Precisely targeted offers lead consumers to take a less active role in information search, limiting cross-category exploration. Firms using highly targeted promotions should monitor for reductions in search, as an indicator of missed cross-selling opportunities and lower sales diversity at the customer level.

Blockbuster Culture's Next Rise or Fall: The Impact of Recommender Systems on Sales Diversity

This paper examines the effect of recommender systems on the diversity of sales. Two anecdotal views exist about such effects. Some believe recommenders help consumers discover new products and thus increase sales diversity. Others believe recommenders only reinforce the popularity of already-popular products. This paper seeks to reconcile these seemingly incompatible views. We explore the question in two ways. First, modeling recommender systems analytically allows us to explore their path-dependent effects. Second, turning to simulation, we increase the realism of our results by combining choice models with actual implementations of recommender systems. We arrive at three main results. First, some well-known recommenders can lead to a reduction in sales diversity. Because common recommenders (e.g., collaborative filters) recommend products based on sales and ratings, they cannot recommend products with limited historical data, even if they would be rated favorably. In turn, these recommenders can create a rich-get-richer effect for popular products and vice versa for unpopular ones. This bias toward popularity can prevent what may otherwise be better consumer-product matches. That diversity can decrease is surprising to consumers who express that recommendations have helped them discover new products. In line with this, result two shows that it is possible for individual-level diversity to increase but aggregate diversity to decrease. Recommenders can push each person to new products, but they often push users toward the same products. Third, we show how basic design choices affect the outcome, and thus managers can choose recommender designs that are more consistent with their sales goals and consumers' preferences.

Blockbuster Culture's Next Rise or Fall: The Impact of Recommender Systems on Sales Diversity

This paper examines the effect of recommender systems on the diversity of sales. Two anecdotal views exist about such effects. Some believe recommenders help consumers discover new products and thus increase sales diversity. Others believe recommenders only reinforce the popularity of already popular products. This paper seeks to reconcile these seemingly incompatible views. We explore the question in two ways. First, modeling recommender systems analytically allows us to explore their path dependent effects. Second, turning to simulation, we increase the realism of our results by combining choice models with actual implementations of recommender systems. We arrive at three main results. First, some well known recommenders can lead to a reduction in sales diversity. Because common recommenders (e.g., collaborative filters) recommend products based on sales and ratings, they cannot recommend products with limited historical data, even if they would be rated favorably. In turn, these recommenders can create a rich-get-richer effect for popular products and vice-versa for unpopular ones. This bias toward popularity can prevent what may otherwise be better consumer-product matches. That diversity can decrease is surprising to consumers who express that recommendations have helped them discover new products. In line with this, result two shows that it is possible for individual-level diversity to increase but aggregate diversity to decrease. Recommenders can push each person to new products, but they often push users toward the same products.. Third, we show how basic design choices affect the outcome, and thus managers can choose recommender designs that are more consistent with their sales goals and consumers’ preferences.