Most IT leaders are trying to do three things at once: modernise their analytics, sensibly move to the cloud and keep a lid on costs. This often results in a tangle of overlapping tools and partial migrations that are expensive to operate and complex to govern.
As artificial intelligence and cloud adoption accelerate, that tangle will not survive. Organisations are entering a “great IT rationalisation” in which legacy systems, siloed data and redundant processes come under sharp pressure. The question is not whether that rationalisation will happen but whether it will be planned and data-driven or forced by budget cuts and risk events.
Why cloud and AI are forcing the issue
Cloud was sold as a way to escape capital expenditure and move faster. In reality, many companies now run duplicate analytics stacks on-premises and in multiple clouds. At the same time, teams are developing new AI services that operate outside existing governance frameworks. That combination pushes costs up and makes it harder to answer basic questions such as: which platform should own this workload? And who is accountable for the data behind this decision?
A recent SAS white paper describes a familiar pattern. Analytics estates grow through mergers, projects and once-off cloud experiments. Over time, licensing and infrastructure spend increase, while utilisation remains uneven and skills are stretched across too many tools.
At the same time, the demand side is exploding. Business teams want more machine learning, real-time scoring and experimentation. If every new AI use case means another point solution, rationalisation is going to be painful.
A single architecture, not another silo
The alternative is to treat this moment as a design opportunity. Instead of adding yet another tool, organisations can consolidate analytics and AI workloads on a single, cloud-native architecture that scales up when needed and scales down when it does not.
This is the role SAS Viya is designed to play. SAS Viya is a cloud-native analytics and AI platform that can run in the public cloud, in a private environment or as a SAS-managed service. It gives IT and data leaders a single decisioning and analytics engine that supports data preparation, model development, deployment and monitoring across the business.
In practical terms, that single architecture lets teams:
- Run mixed workloads (from traditional reporting to deep learning) on shared, elastic infrastructure;
- Apply consistent governance for data access, models, and decisions, regardless of where they execute; and
- Retire overlapping legacy tools without losing analytical capability.
Benchmarks from The Futurum Group show why this matters for rationalisation. In that study, SAS Viya delivered up to 30x faster performance and 86% better cost-effectiveness than commercial and open-source alternatives. Faster runs on a more efficient platform reduce the infrastructure footprint required for the same or greater analytical output.
Intelligent rationalisation
Rationalisation often has a negative reputation inside IT. It sounds like cost-cutting, programme freezes and painful decommissioning. The smarter approach is to link rationalisation directly to business value and agility.
Using an integrated platform such as SAS Viya, IT leaders can:
- Map existing analytics workloads and identify which ones truly need dedicated resources;
- Right-size environments using workload management and auto-scaling, so that CPU and memory track demand more closely over the day or month; and
- Move high-value models and data flows into a governed environment, then decommission the legacy scripts and servers they replace.
The SAS Viya cost-optimisation paper includes an example where dynamic workload management and new sizing options helped reduce cloud infrastructure costs by more than 70% for a complex analytics estate. Those savings did not come from doing less analytics. They came from concentrating work on a platform that could use cloud resources more efficiently, then scaling down when demand was low.
From projects to a decisioning backbone
The real prize in this great IT rationalisation is not just a lower bill. It is a more coherent decisioning backbone for the organisation.
When analytics and AI share a single platform, it becomes easier to reuse components and keep models in sync. A risk model developed for one line of business can be adapted for another without lifting and shifting data between systems. Marketing, fraud, collections and claims teams can all leverage the same governance features and decision flows.
That is where SAS positions Viya: as an enabler of intelligent rationalisation. The aim is not to stop experimentation, but to give it a home. Data scientists can still use open-source tools, but they deploy them into a common environment. Business teams still get rapid innovation, but on an architecture that IT can support and auditors can understand.
The great IT rationalisation is coming, whether by design or by attrition. Organisations that use this moment to simplify architectures, concentrate analytics on a single platform and treat rationalisation as a strategic move will be the ones that come out stronger on the other side.
