An Azure platform for vCIO service management

The client sells software and training to managed service providers (MSPs) in the United States. One of its services is a virtual Chief Information Officer (vCIO) offering, where its people act as the IT strategy lead for MSP clients. As the client base grew, the system behind that service could not keep up. The answer was a custom vCIO management platform built on Microsoft Azure, designed, built, tested and rolled out over six months.

The client

The company was founded in 2005. It makes software tools that help MSPs run their operations, and it runs training programs for them. At the time of this project it had more than 500 MSP clients, who between them served more than 10,000 end users across a range of industries. Its main product is an all-in-one MSP management platform.

The vCIO service is where growth hurt most. Each new MSP client added data, service requests and reporting work, and the existing setup was not built for that volume.

The problem

Seven problems had built up as the business grew:

1. The infrastructure slowed down at peak times because it could not handle the volume of client data and service requests.
2. Assigning vCIO staff across a varied client base was inefficient and often matched the wrong person to the client.
3. Client information sat on several separate platforms, so nobody had a full view of a client's needs or of how the service was performing.
4. Detailed reports took a long time to produce, so clients and internal teams got their insights late.
5. The system was hard to tailor to an individual client's needs.
6. The service was reactive. Problems were handled after they appeared instead of being caught early.
7. Keeping data protected and compliant across many different client environments was getting harder.

Together they stood in the way of the company's growth and its position in the market.

What we built

We built a cloud-based vCIO service management system on Azure. It uses these Azure services:

• Azure SQL Database stores and manages client data in one place.
• Azure Functions runs serverless, event-driven jobs that scale on their own.
• Azure Logic Apps automates multi-step workflows and business processes.
• Power BI provides live dashboards and reports.
• Azure Active Directory handles sign-in and access control.
• Azure Cognitive Services adds natural language processing and computer vision features.
• Azure Machine Learning powers predictive models and personalized recommendations.

On top of those services the platform has a central client data repository, AI-assisted assignment of vCIO staff, customizable service templates, real-time monitoring and alerts, a predictive analytics engine, automated reports, a compliance management module, and a cloud-native architecture that scales with demand.

How the work was done

The project ran for six months in four phases.

Discovery and planning (1 month). The team reviewed the existing systems and processes, agreed on objectives and success criteria, and wrote the project plan and the data migration strategy.

Development and integration (3 months). Work ran in two-week Agile sprints. Core features came first and the more advanced ones were added after. Stakeholders reviewed progress regularly, and testing ran throughout.

Testing and refinement (1 month). This phase covered full testing, including stress tests and security audits. A pilot ran with 20 MSP clients of different kinds, and the platform was adjusted based on their feedback and on performance data.

Training and rollout (1 month). Staff got training materials and sessions. The platform went out to all MSP clients in phases, and a dedicated support team was set up with a feedback loop.

Two parts were harder than the rest. Moving data out of the legacy systems was complex, and we handled it with a custom ETL process built on Azure Data Factory. Getting the AI features accurate took extensive model training and refinement. Throughout the project, the company worked with Microsoft's Azure support team on technical problems and on cloud architecture and security practices.

Results

These are the results the company reported after the platform was in full use.

Capacity and efficiency:

• The system handled a 300% increase in client load without performance degradation.
• vCIO resource utilization rose 40%, so the team could take on more clients without adding staff at the same rate.
• Report generation time fell 80%, giving clients and stakeholders near real-time insights.
• Infrastructure costs fell 30%, which the company put down to Azure's pay-as-you-go pricing.

Service quality:

• With data in one place and AI-driven analytics, proactive service recommendations rose 25%.
• The platform offered 50% more service customization options.
• Net Promoter Score rose 15 points within three months of full implementation.
• Security-related incidents fell 60%, and the company reported a 100% improvement in audit pass rates for regulated clients.

Business and team:

• Revenue from the vCIO service line rose 35%.
• New client acquisitions rose 20% within the first year.
• Employee satisfaction scores in the vCIO team rose 25%.

What came next

When this case study was first published in July 2024, the company was looking at further work on the platform. Its list included more advanced machine learning for predicting IT issues, an AI assistant for support, multi-language support and region-specific compliance modules for MSPs outside the US, and a mobile app for vCIO functions and real-time alerts. It was also exploring IoT and edge computing support, blockchain-based audit trails, modules for more sustainable IT practices, and augmented and virtual reality for remote infrastructure management and training. These were plans at the time, not finished features.

If you run a service business on software that is slowing down as you grow, our web development team builds custom platforms like this one. Tell us what you have now and where it breaks.