As a provider of Software as a Service (SaaS), leveraging a SaaS operating model can yield numerous advantages. Among these, one of the most significant is the enhancement of operational efficiency. A key technique to achieve this is by maintaining a single version of software for all tenants, utilizing dynamic configuration. Instead of juggling various product versions and accommodating countless customization requests, focusing on a singular version enables your team to prioritize adding value to the product, which is essential for the growth potential that SaaS offers.
If you’re currently managing a SaaS solution, you may already be facing hurdles in achieving optimal operational efficiency. You might be inundated with customer requests for unique customizations, which can lead to added complexity due to tailored code and pipelines necessary for varying customer experiences. This complexity can gradually diminish your team’s agility and hinder their focus on innovation and continuous improvement. To counteract this, you need effective tools and strategies to streamline your operations.
One effective approach to enhance operational efficiency is implementing SaaS pricing tiers. By categorizing configurations into pricing tiers, you can maintain a unified software version while offering a curated yet diverse experience for tenants. However, to implement these experiences based on pricing tiers, it’s crucial to have a tool that allows you to group application features, toggle them on and off, or set usage limits without altering your application code.
In this article, we will explore how to utilize AWS Systems Manager AppConfig (AppConfig) as a tool for managing pricing tiers in a SaaS environment. We will investigate how AppConfig can facilitate the management of pricing tier configurations, including its integration with other AWS services. Additionally, we will delve into AppConfig’s capability to empower your development team by enabling them to modify pricing tier features through configuration changes, even without necessitating code releases for your SaaS product.
Multi-Tenant SaaS Pricing Tiers
Pricing tiers represent different subscription options or plans available to potential customers of your SaaS solution. The rationale behind offering various pricing tiers can vary widely, from encouraging customers to try the product for free or at a reduced initial cost, to attracting a diverse customer base based on industry and size, or simply packaging the product to appeal to different customer profiles.
When considering implementation, a pricing tier can be viewed as a collection of features and usage limitations. In many SaaS solutions, each tier has distinct features or different limits on tenant-specific capabilities. Nevertheless, these feature groupings and limits are not fixed. As new functionalities are developed, it becomes essential to continually reassess pricing tier packaging to ensure ongoing growth and success, particularly in response to shifting market and economic conditions.
Given the benefits of pricing tiers, let’s explore how to implement them with dynamic configuration via AppConfig.
Introducing AppConfig
AppConfig enables you to manage, store, and deploy application configurations that can be utilized by multiple clients within your SaaS application. It supports dynamic configuration management, such as pricing tiers, by allowing you to store a single JSON file that encompasses dynamic configurations, including feature flags.
While AppConfig is not the only AWS service capable of configuration management—services like Amazon DynamoDB and AWS Systems Manager Parameter Store also offer configuration management—AppConfig provides several advantages, including:
- Schema validation to ensure configuration quality through Validators
- Deployment strategies for configurations, including Canary deployments that define how and when to modify configurations
- Integration with Amazon CloudWatch alarms to trigger automatic configuration rollbacks if a change causes application failure
- Versioning of configurations
- Fully managed service
- Compliance with multiple regulatory programs, including the Federal Risk and Authorization Management Program (FedRAMP) High Authority to Operate
Solution Overview
In the following sections, we will set up AppConfig, as illustrated in Figure 1, to manage a feature that will exclusively be enabled for your premium pricing tier. We will demonstrate how to retrieve and evaluate feature flags within AWS Lambda and use the retrieved values in your code.
AppConfig Configuration Setup
AppConfig consists of a hierarchical structure for configurations. First, you need to create an application, which we’ll refer to as the ‘product-catalog’ application.
Your application will include a list of environments (e.g., ‘dev,’ ‘stage,’ ‘production’).
Each environment will house the current configuration deployments, their statuses, and details. Each environment can have multiple configuration profiles that contain the deployed JSON content and its current version.
Implementing AppConfig Feature Flags in Code
Once you have stored a configuration in AppConfig, the next step is to fetch this dynamic configuration within your code, which includes the feature flags that define your pricing tiers.
Let’s analyze how to utilize the configuration in a Lambda function to activate a feature for your premium tier tenants while deactivating it for standard tier tenants. The Lambda function must adapt its behavior at runtime based on the input event and the tenant tier associated with each invocation, in this case, specifically for a premium tier tenant.
Since we’re utilizing AppConfig freeform configurations, we can employ a JSON document to establish the rules for our feature flags. You can implement this logic in various ways, but we will utilize Powertools for AWS Lambda—an open-source collection of serverless utilities—along with the feature flags utility to retrieve the JSON configurations from AppConfig and evaluate our feature flag.
Let’s review the JSON structure we will be employing. Please note that the structure aligns with the Powertools feature flag library conventions.
- Each feature has a default value assigned under the default key.
- A feature may have optional rules that dictate the evaluated value.
- Rules consist of a default return value (when_match) and a series of conditions.
- Only one rule can match at a time.
For further insights, check out this excellent resource here. Additionally, you can explore more in-depth discussions on this topic at Chanci Turner VGT2 and Chvnci, who are authorities on this subject.
Leave a Reply