Use Case: Scale-Up - Progressive Rollout to Safeguard Infrastructure
Customer: E-commerce B2C Scale Up
Team Size: 50
Industry: B2C SaaS
Features Used: Progressive Rollouts
The problem
A fast-growing SaaS company (or scale-up) is adding new, potentially risky features or major architectural changes (e.g., a new data-intensive API endpoint, new microservice, or heavy-compute workflow). Because their user base is large and traffic is high, any bug, performance regression, or resource spike could severely impact infrastructure (latency, capacity, costs)..
Issues they are facing:
- Rolling out a new feature to everyone could overload servers, cause performance degradation, or crash production.
- If something breaks after deploy, rolling back via code is slow and risky (may require a fresh deploy).
- They don’t want to slow down innovation - engineers need to deploy code quickly, but they also need safeguards.
- Non-dev stakeholders (ops, SRE) want control: they want to monitor and limit exposure to risky features without blocking engineering velocity.
The solution
The team turned to FlagBox to simplify and speed-up their rollout process. After a quick integration,
they started using FlagBox to:
- Deploy the new feature behind a flag: the code is live in production, but only a subset of traffic sees it. This reduces risk.
- Start with a tiny cohort (e.g. internal users / canary group) to test system behavior under production load.
- Gradually increase the rollout (e.g. 5 % → 25 % → 50%) only if metrics remain stable.
- If something goes wrong (e.g. bug, performance spike), flag can be turned off immediately - no need for a full redeploy. Flagbox enables that via dashboard.