IG Group Holdings plc is a global fintech company, providing over 1.3 million customers with access to more than 21,000 financial markets. Founded in 1974 and headquartered in London, the Group offers leveraged trading, stock trading and investments, and cryptocurrency trading through its proprietary platforms.
Operating in a highly regulated industry where customer trust is paramount, IG Group runs a global network that must be secure, resilient, and available around the clock. Balancing constant change with the predictability customers expect is one of the defining challenges for the IG network engineering team.
Without pre-change verification, the change window itself becomes the test. Validating in production means the risk has already been taken. Forward Predict moves verification upstream, evolving the change window from a moment of exposure into a controlled deployment. IG Group is actively working toward an autonomous networking future.
After more than a year working with Forward, the team had come to rely on the platform for trustworthy network insight, with pre and post-change configuration checks already validating that updates kept the network compliant and connectivity intact. Even so, every significant change involved weeks of meetings and approvals that consumed several hours of engineering and required late night maintenance windows with no certainty of the outcome until the change was live.
Forward Predict enables them to make and verify changes in the mathematically accurate network digital twin, which virtually replicates the entire production environment. This step proves the outcome before deployment to the production network and gives the IG Group team confidence that every proposed change will maintain compliance and deliver the intended result.
“Once you see the platform, you understand how it works, and trusting it is easy,” said Steve Bamford, Senior Network Engineer at IG Group. “I have deliberately tested scenarios that, if pushed live, would have isolated parts of the network and taken the network down. Forward Predict caught them. Being able to test changes across our entire production network before they go live is a game-changer, and it accelerates our autonomous networking journey.”
The IG team also uses Forward Predict to test failover scenarios. In one exercise, they modeled an error that would have isolated one of their primary U.K. data centers, then ran their rollback procedure entirely within the platform. The result was a fully documented scenario, complete with evidence that the rollback would have returned the network to full connectivity, all without ever withdrawing security from production infrastructure.
“Attempting that on the live network would have meant at least ten minutes of disruption, which simply isn’t practical in our industry,” Bamford said. “With Forward Predict, we can prove our rollback works without taking anything offline. It is a one-to-one copy of our environment, and we can use that as evidence with auditors and the business leadership.”
Evidence-based documentation of change impact before go-live has transformed IG’s review process. Using Forward Predict to validate outcomes in advance, the team can move from sequential approvals to rapid sign-off, materially accelerating the pace of change.
“If it works in Forward Predict, we can comfortably say we can deploy it without risk,” Bamford said. “It also reduces the number of late Friday night change windows. A lot of changes that used to require a 10 PM slot we can now run with confidence at end of day.”
As IG’s entire leaf and spine fabric runs BGP, small mistakes can have outsized consequences. A common example is the difference between “no neighbor IP address shutdown” and “no neighbor IP address.” The text is nearly identical, but the second command tears down the entire neighbor relationship and the connectivity that depends on it. In a long, manually reviewed change scripts, that kind of mistake is easy to miss.
“Our whole infrastructure runs on BGP. The leaf and spine, our connectivity to ISPs, CDNs, all of it. Changes to the routing between data centers, or even within a data center, are complex and critical. Being able to predict the outcome before executing a change means we deliver it right the first time and in a shorter window,” according to Bamford.
Onboarding fix clients is one example where pre-change verification translates directly into business outcomes. The process demands a 100 percent success rate and previously required input and approvals from multiple teams. With Forward integrated into the automation flow, IG has compressed this from a weeklong process to a same-day delivery following security approvals, improving customer satisfaction and allowing the business to recognize revenue sooner.
IG Group’s vision extends well beyond faster changes. Specific future use cases include validating configurations during major migrations, modeling the impact of large infrastructure shifts before pushing to production, and using Predict as the verification step inside automated fault-resolution workflows.
“Forward Predict is key to our drive toward autonomous networking and will be central to any automated fault resolution we build.”
To learn more about how Forward Predict can help your team move fast and break nothing, visit forwardnetworks.com/predict
