The Influence of Software Process Maturity and Customer Error Reporting on Software Release and Pricing

Abstract

Software producers are making greater use of customer error reporting to discover defects and improve the quality of their products. We study how software development differences among producers (e.g., varying levels of process maturity) and software class and functionality differences (e.g., operating system versus productivity software) affect how these producers coordinate software release timing and pricing to optimally harness error reporting contributions from users. In settings where prices are fixed, we characterize the optimal release time and demonstrate why in some cases it can actually be preferable to delay release when customer error reporting rates increase. The manner in which a firm's optimal release time responds to increases in software functionality critically hinges on whether the added functionality enhances or dilutes user error reporting; in both cases, the effect of added functionality on release timing can go in either direction, depending on both firm and product market characteristics. For example, when processing costs are relatively large compared to goodwill costs, firms with lower process maturity will release earlier when per module error reporting contributions become diluted and release later when these contributions become enhanced. We also examine how a firm adapts price with changes in error reporting levels and software functionality, and finally provide implications of how beta testing influences release timing.

Pricing Strategies under Behavioral Observational Learning in Social Networks

“…() studied investment incentives for network infrastructure owners under different pricing strategies. August and Niculescu () examined how a software firm optimally determines its pricing in a setting where it can harness customer error reporting. Sellers' other strategies, such as the introduction of product upgrades (Ji et al.…”

Renegotiation of Software Outsourcing Contracts

“…They considered testing as continuing during system operation, that is, post-release testing. August and Niculescu [1] further examined the impact of software demand on post-release testing. In addition, Jiang et al [16] considered both software reliability and marketrelated benefits and derived not only the optimal testing time but also the optimal number of testers.…”

“…To exclude the unrealistic case in which testing time is negligible, we assume to ensure the marginal benefit of testing is greater than its marginal cost. 1 The utility of the system to the client decreases with the total number of system bugs. Therefore, the total utility of the client can be expressed as:…”

Software Defect Prediction from Code Quality Measurements via Machine Learning

“…If more software defects are found during testing than expected, not only can this cost the company money, but it can also impair their ability to meet customer deadlines or mission milestones. Software defects can also complicate the warranty period of a software product, thus making it difficult for a software development company to move on from a program when it needs to continue to dedicate resources to supporting customers as they find new defects through normal use [25].…”