Racing the Market: An Industry Support Analysis for Pricing-Driven DevOps in SaaS

Alejandro Garcia-Fernández,José Antonio Parejo,Francisco Javier Cavero,Antonio Ruiz-Cortés
2024-09-24
Abstract:The SaaS paradigm has popularized the usage of pricings, allowing providers to offer customers a wide range of subscription possibilities. This creates a vast configuration space for users, enabling them to choose the features and support guarantees that best suit their needs. Regardless of the reasons why changes in these pricings are made, the frequency of changes within the elements of pricings continues to increase. Therefore, for those responsible for the development and operation of SaaS, it would be ideal to minimize the time required to transfer changes in SaaS pricing to the software and underlying infrastructure, without compromising the quality and reliability.% of the service; %i.e., this development and operation should be Pricing-Driven. This work explores the support offered by the industry for this need. By modeling over 150 pricings from 30 different SaaS over six years, we reveal that the configuration space grows exponentially with the number of add-ons and linearly with the number of plans. We also evaluate 21 different feature toggling solutions, finding that feature toggling, particularly permission toggles, is a promising technique for enabling rapid adaptation to pricing changes. Our results suggest that developing automated solutions with minimal human intervention could effectively reduce the time-to-market for SaaS updates driven by pricing changes, especially with the adoption of a standard for serializing pricings.
Software Engineering
What problem does this paper attempt to address?
The problem that this paper attempts to solve is: How to minimize the implementation time of SaaS (Software as a Service) pricing changes in software and underlying infrastructure without affecting quality and reliability. Specifically, the paper explores the industry's support for pricing - driven DevOps (Development and Operations), aiming to provide more efficient solutions for SaaS developers and operators to deal with frequent pricing changes. ### Background and Problem Description of the Paper With the popularization of the SaaS model, suppliers provide customers with multiple subscription options through pricing structures (including different plans and add - ons). Although this flexibility enhances the user experience, it also brings management and implementation challenges, especially in the case of frequent pricing changes. In order to remain competitive, SaaS developers and operators need to quickly apply these pricing changes to software and infrastructure without affecting the quality and reliability of the system. ### Main Research Questions 1. **Evolution of Pricing Configuration Space and Evaluation Space**: - Research how the configuration space and evaluation space in the SaaS pricing model evolve over time. - Analyze the growth trends of these spaces and their impact on SaaS providers' management of pricing - driven development and operations. 2. **Are Feature Toggle Tools Suitable for Optimizing Pricing - Driven Feature Toggling?**: - Evaluate whether existing feature toggle tools can effectively support pricing - driven feature toggling. - Explore the gap between current tools and the requirements of pricing changes. 3. **Can Yaml4SaaS Adequately Represent the Actual Pricing Model?**: - Verify whether Yaml4SaaS can accurately model and represent SaaS pricing models in the real world. - Identify the limitations of Yaml4SaaS in expressing complex pricing structures. ### Solutions and Methods - **Analysis of Pricing Configuration Space and Evaluation Space**: - By modeling the pricing structures of more than 150 SaaS products from 30 different SaaS products and tracking their evolution over six years, it is revealed that the configuration space grows exponentially with the number of add - ons and linearly with the number of plans. - Define and analyze the configuration space and evaluation space of pricing to understand their dynamic changes. - **Evaluation of Feature Toggle Tools**: - Compare 21 different feature toggle solutions and evaluate their support capabilities in achieving pricing - driven feature toggling. - Propose three capability levels (L1, L2, L3), corresponding to configuration file toggling, dynamic context evaluation, and pricing - aware evaluation respectively. - **Verification of Yaml4SaaS**: - Use Yaml4SaaS to model the pricing models of 30 SaaS products to verify its expressive ability and limitations. ### Conclusions - **Evolution of Configuration Space and Evaluation Space**: - The configuration space grows exponentially with the increase in the number of add - ons, and the evaluation space expands with the increase in the number of functions, increasing management complexity. - **Applicability of Feature Toggle Tools**: - Currently, only a few tools partially support pricing - aware evaluation, and most tools only support basic feature toggling. - **Expressive Ability of Yaml4SaaS**: - Yaml4SaaS can model most complex pricing structures, but has limitations in expressing formula calculations, cross - add - on dependencies, and other complex constraints. ### Formula Presentation The configuration space \(C\) is defined as follows: \[ C(P, A, E, D)=\{(p, a) \in P \cup \emptyset\times P(A) \mid \text{isV alid}(p, a, E, D)\} \] where the `isV alid` function is defined as: \[ \text{isV alid}(p, a, E, D) \iff \begin{cases} p \in P \cup\{\emptyset\} \\ a \in P(A) \\ ((p \neq \emptyset) \lor (a \neq \emptyset)) \\ |P| \geq 1\Rightarrow p \neq \emptyset \\ \end{cases} \]