Facilities Management · Case Study
99.9% Uptime. No Rewrite. The Cleaning Operations Platform That Came Back From the Edge.
When a cleaning operations business runs entirely through one platform, a crash isn’t a technical incident – it’s a commercial one. This is how Emvigo stopped it from happening again.
Project Overview
A Single Server. An Entire Business. And an Infrastructure That Had Been Quietly Failing for Months.
A Swedish facilities management company was running their entire operation of scheduling, assignments, key access, and workflows through a single web platform. When it worked, the business worked. When it didn’t, everything stopped.
By the time they approached Emvigo, it was stopping every afternoon. MongoDB was consuming 100% CPU. The single-server architecture was collapsing under multi-tenant load during peak scheduling hours. Deployments were manual FTP processes, adding further risk to an already fragile system.
The constraint was firm: no full rewrite. They had invested in the existing Meteor.js stack and needed targeted rescue and not a costly overhaul that would take months to deliver and require rebuilding institutional knowledge from scratch.
Emvigo diagnosed before prescribing, stabilised the architecture in three months, and delivered 99.9% uptime. After this, the client disbanded their internal development team and made Emvigo their exclusive engineering partner.
Project At a Glance
Client
Industry
Sweden
Emvigo’s Role
Technology Stack
- Meteor.js
- Vue.js
- MongoDB Atlas
- AWS ECS
- Elastic Beanstalk
- New Relic
- CloudWatch
- Sentry
The Challenge
The Worst Time for a System to Fail Is Peak Hours. This One Did It on Schedule.
The platform wasn’t experiencing occasional technical difficulties. It was failing predictably, repeatedly, and at the worst possible time with direct consequences for coordinators, cleaning staff, and customers on every occasion.
Customer Perspective
- Facilities managers booking and managing cleaning schedules faced unpredictable slowdowns and outages at the exact hours they depended on the system most
- Inaccurate or missing time logs caused by system failures led to billing disputes and eroded the trust that a subscription cleaning relationship depends on entirely
- Cleaning staff were unable to start or end mission logs during outages, resulting in complete data loss for jobs that had been physically completed
- Manual admin corrections to fix missing hours, delayed invoicing, disrupted payments, and consumed coordinator time that should have been directed at operations
Business & Operational Problems
- MongoDB running at 100% CPU was causing predictable afternoon degradation, with no monitoring infrastructure in place to diagnose or anticipate failure events
- A single-server architecture meant the entire operation shared one point of failure, with no isolation between tenants and no capacity to absorb load spikes without collapse
- Weak tenant isolation meant one high-usage customer could degrade the platform experience for every other customer simultaneously
- Manual FTP deployments were adding operational risk to every release, with no staging environments, no rollback capability, and no repeatable deployment process
- The client's investment in Meteor.js ruled out a full rewrite; the solution had to stabilise existing infrastructure rather than replace it
Running a mission-critical platform that's becoming less stable as your business grows, and a full rewrite isn't an option?
Emvigo diagnoses before prescribing, and stabilises what exists rather than replacing what works.
Product Strategy
Emvigo's Strategic Role: Stabilise First. Scale Second. Never Rebuild What Doesn't Need Replacing.
The first and most important strategic decision was to spend ten days diagnosing before proposing anything. Root causes had to be identified before any solution was scoped. That discipline alone separated this engagement from the pattern of quick patches and deferred problems the platform had already accumulated.
The architectural priority was eliminating single points of failure and removing resource contention without touching the Meteor.js investment the client had made. AWS containerisation, MongoDB Atlas migration, and real-time monitoring were introduced as infrastructure layers around the existing stack.
Stability isn’t a feature you add. It’s a structural condition you create by removing everything that prevents it. That requires diagnosis, not assumptions.
– Emvigo Engineering Team
01
Diagnose Before Prescribing
A 10-day deep-dive audit identifying MongoDB bottlenecks, architecture root causes, and performance failure points before any solution was proposed or scoped.
02
Eliminate the Single Point of Failure
Migration from a single failing EC2 instance to AWS ECS and Elastic Beanstalk containerisation with MongoDB Atlas. This removed the architectural condition that was causing the crash every afternoon.
03
Real-Time Visibility Before Any Further Change
New Relic, AWS CloudWatch, and Sentry were deployed to establish a full monitoring infrastructure, ensuring every performance event was visible, traceable, and actionable from that point forward.
04
Automated Deployments Replacing Manual Risk
CI/CD pipelines across Dev, QA, UAT, and Production replacing manual FTP, making every future release faster, safer, and repeatable without adding operational risk.
05
Preserve the Stack, Remove the Instability
Meteor.js and Vue.js were retained entirely with the client’s existing investment, protected while the infrastructure underneath it was made fit for purpose.
Our Solution
Four Structural Fixes. No Rewrite. One Platform That Finally Held Under Load.
Emvigo delivered a complete infrastructure stabilisation across four phases in three months. We did this without touching the application code the client had invested in, and without disrupting live operations throughout the process.
Root Cause Discovery Audit
A structured 10-day technical audit identifying every MongoDB bottleneck, architecture failure point, and performance degradation cause - producing a diagnosis that the entire subsequent engagement was built on rather than assumed from.
Monitoring and Diagnostics Infrastructure
New Relic, AWS CloudWatch, and Sentry were deployed across the entire platform. This gives the client real-time visibility of performance events for the first time, ensuring no future issue can go undetected or untraced.
Automated CI/CD Pipeline Deployment
Manual FTP deployment process replaced with automated pipelines across Dev, QA, UAT, and Production environments, eliminating deployment risk, enabling rollback capability, and making every future release a controlled, repeatable process.
AWS Containerisation and Database Migration
Single EC2 instance replaced with AWS ECS and Elastic Beanstalk containerisation, with MongoDB Atlas as the managed database layer. This handles multi-tenant concurrent load without degradation and removes the CPU bottleneck that had been causing daily failures.
Business Outcomes
99.9% Uptime. Zero Afternoon Crashes. And Then They Let Their Internal Team Go.
99.9% SLA
Uptime Achieved
3 Months
Timeline
From Daily Crashes to Enterprise-Grade Reliability
The afternoon degradation pattern that had been disrupting scheduling, billing, and staff operations every working day was eliminated entirely, with 99.9% uptime sustained from the point of production deployment.
Zero CPU Bottlenecks Post-Delivery
MongoDB Atlas and AWS containerisation resolved the 100% CPU condition at its architectural root through structural changes that removed its cause.
Automated Deployments Replaced Manual Risk
CI/CD pipelines across all environments replaced the manual FTP process, giving the development workflow the same reliability that the infrastructure had just achieved.
Internal Dev Team Disbanded - Emvigo Became Exclusive Partner
The client disbanded their internal development team entirely and made Emvigo their sole engineering partner for all ongoing development and roadmap leadership.
Achievement Unlocked: Results Like These
72% of successful projects start with one conversation
Ready to Build Something This Impactful?
Emvigo specialises in turning ambitious digital products into reality — at speed, at scale, without sacrificing quality. Your breakthrough project is one call away.
- ISO 9001:2015 Certified
- 13+ Years Experience
- 500+ Global Clients
- AWS Partner
Get In Touch
Have a question for our team or need help with your project?
Our team is ready to provide client references, estimate your project, or answer any other question related to your IT initiative. We typically respond within 2 business hours.
Start the conversation
- Free consultation
- No obligation
- We reply in < 2 hours
🔒 Verified & Secured
