Ensono Uses AI/ML to Predict IT System Failure and Slash MTTR up to 70%

Modernizing infrastructure can grow complex — fast. Many organizations turn to a reliable managed services provider (MSP) to maintain exceptional operations, increasingly seeking AI-enabled partners that deliver smarter automation, earlier issue detection and a more resilient environment.

As an expert technology adviser, innovation partner and MSP, Ensono helps organizations modernize and manage their entire technology estate, including legacy infrastructure and mission-critical applications, with offerings that span consulting, mainframe and cloud managed services. Ensono’s outcome-driven solutions deliver cloud migrations and data center consolidations, supporting over 60 billion retail transactions and providing more than 24 million constituents with access to government platforms.

Delivering reliable results at this scale requires a flexible, powerful data foundation — which is why Ensono chose to partner with Snowflake. With the Snowflake AI Data Cloud, Ensono centralizes its data and powers software that accelerates service delivery, anticipates issues before they impact clients, and strengthens the overall customer experience.

“We’ve reached new heights in terms of customer satisfaction. Eighty percent of our clients recommend us to other customers. That’s a tangible measure of the quality of the delivery we provide to every one of our clients.” —Jim Piazza, Chief AI Officer, Ensono

Ensono works with large enterprise clients. The IT environments of these companies involve hundreds of servers, thousands of SaaS user accounts and terabytes of data. Multiply this complexity across a growing client base, and it’s clear why Ensono's software-first mentality prioritizes proactive detection. Ensono’s goal is shifting the MSP mindset from monitor-alert-respond to prevent-predict-optimize. “We’re trying to predict failures across our clients’ IT environments before they happen, which would allow us to step in, take action and prevent a business- or service-impacting event.”

Establishing a solid data foundation is an essential step in the predictive analytics journey. “Luckily, when I came on board, Snowflake was already here,” Piazza says. Having already integrated massive amounts of operational, system monitoring and log data into Snowflake, Ensono began to develop a customer-facing portal. “We bring all that data together in Snowflake and surface it through the Ensono Envision® client portal, giving our clients real-time visibility into their environments. They love it because it helps them move faster and make smarter decisions with confidence,” Piazza says. Combining Ensono’s unified data lake in the AI Data Cloud with the capabilities of Snowflake ML puts Piazza’s team in a position to innovate. 

“We wanted to deploy models as quickly as possible. And with Snowflake ML, we don’t have to worry about creating or finding another model hosting platform because we can use the Model Registry to manage and deploy models for inference with our existing pipelines,” says John Stamford, Vice President, Data Science and Machine Learning at Ensono. “With Streamlit in Snowflake, we don’t have to worry about maintenance. We can configure the models in production instead of changing the underlying code.”

Ensono leveraged Snowflake, alongside Evolution Analytics, a Snowflake partner specializing in data and analytics, to build and deploy Envision Predictive Engine (EPE), an ML-powered solution that delivers proactive failure insights across clients’ hardware, operating systems, applications and networks. Since launching, EPE has helped Ensono analyze more than 75 million events and 9 million alerts.

Ensono’s EPE models, managed with Snowflake Model Registry, estimate each support ticket’s probability of becoming a service-impacting event and surface likely precursors to significant issues. Creating a near real-time ingestion of data via the Snowflake Connector for ServiceNow accelerates time to insight and lessens in-house API development. Pushing confidence ratings back into ServiceNow provides frontline staff with vital context for prioritization. “The Snowflake Connector for ServiceNow is a significant part of the solution for us,” says Piazza. “A high-confidence ticket gets spotlighted as a ServiceNow popup window that says to treat the case urgently.”

Having clean data labels is critical for EPE models to be effective. By using OpenAI’s GPT models in Snowflake’s Cortex AI functions, Ensono data labeling is now a simple task, saving employees hundreds of hours. “For such a simple but important use case, we previously had teams manually review data in dashboards, which was tedious and difficult to scale,” says John Stamford.

All this enables Ensono to respond faster, automatically flag high-risk situations with EPE and avoid downtime. “When EPE proposes major incidents, the MTTR is 54% lower,” Piazza says. EPE also earned the Predictive Analytics Solution of the Year award in the eighth annual AI Breakthrough Awards, a program that received over 5,000 nominations.

Supplementing Envision Predictive Engine is DiagnoseNow, an AI-powered incident resolution tool created by Piazza’s team that automates root cause analysis for Ensono’s users. According to Piazza, “Its sole purpose is to get the right information into the right person’s hands so they can solve incidents faster.”

Built using Snowflake Cortex AI and Streamlit, DiagnoseNow brings together a variety of situational data and provides engineers with AI-guided resolution plans. Case-specific details, event timelines, error code summaries, similarity scores and recommended actions become available to technicians in less than a minute after initiating a request. “With just a quick click of a button, users can self-diagnose what may be going on,” Piazza says. 

Used in tandem with EPE, DiagnoseNow has helped Ensono spot more than 1,700 issues, reducing major incidents by 22%. In addition, a pilot study found that using DiagnoseNow reduces incident MTTR by as much as 70% in some cases. As a result, Ensono’s clients enjoy elevated levels of service, less downtime and lower support costs.

Continually refining EPE and DiagnoseNow for even better client outcomes is a priority for Ensono. For example, the team will be adopting Snowpark Container Services to accelerate performance for DiagnoseNow users. This enables users to have a faster experience and decrease the load time of the application.

As Ensono continues building its decision engine, Snowflake-managed Model Context Protocol (MCP) servers will play an important role. “One of the things I’m most excited about is the inclusion of MCP in the platform,” Piazza says. “Having more specialized models working together, with different systems, greatly improves the quality of outcomes. It’s like having a team of experts on demand.”

Looking ahead, Ensono’s partnership with Snowflake and its expanding AI portfolio will fuel greater innovation and client impact. By leveraging Snowflake’s AI Data Cloud, enhancing predictive capabilities, and exploring emerging tools like Snowpark Container Services and Model Context Protocol, Ensono is pushing the limits of proactive IT management. Ensono’s continued investment in AI and Snowflake enables faster resolutions, stronger resilience and smarter automation for the long term.