How AWS is changing the Cloud cost conversation

UK organisations spend an estimated £2.8 billion annually on cloud resources they never use. Flexera's 2025 report found that 27% of cloud spend is wasted. Apply that to the CMA's £10.5 billion UK IaaS and PaaS market valuation. It is not a rounding error. It is a structural problem.

The challenge runs deeper than forgotten instances. Database workloads now represent one of the largest persistent cost categories and, until recently, sat outside flexible commitment pricing. AI workloads add further complexity, and token-based billing creates cost patterns that traditional forecasting cannot anticipate.

Industry analysts warn that organisations scaling AI initiatives routinely underestimate costs by an order of magnitude. Yet cost management tools have not stood still. AWS used re:Invent 2025 to announce its most significant expansion of financial management capabilities in years. Database Savings Plans, AI-powered cost analysis, and standardised billing data address the gaps that manual processes cannot close.

The question for UK IT leaders is not whether to act on cloud cost control. It is whether current practices can keep pace with how cloud consumption is changing.

What AWS launched and why it matters

Database Savings Plans launched in December 2025, bringing flexible commitment pricing to nine database services. These include Aurora, RDS, DynamoDB, ElastiCache, Neptune, and DocumentDB. Savings reach up to 35% on serverless deployments and up to 20% on provisioned instances. Unlike Reserved Instances, these plans let you switch database engines, regions, and instance families within a single commitment. That flexibility matters when workload patterns shift faster than annual planning cycles allow.

The significance is practical. Database-as-a-Service spending is growing at over 15% annually, driven by analytics and AI use cases. Previously, organisations seeking commitment discounts managed separate reservation portfolios for each service. Database Savings Plans collapse that complexity into one hourly commitment across all covered services. 10x Banking, a UK fintech, achieved a 50% reduction in database operating costs and a sixfold performance improvement after migrating to Amazon Aurora PostgreSQL.

Cost Optimisation Hub now consolidates over 18 types of savings recommendations into a single dashboard. The new Cost Efficiency Metric gives organisations a daily score based on actual spend against potential savings. Getting different teams to track the same optimisation metric has historically been anything but simple.

AI-powered analysis changes the FinOps conversation

Amazon Q Developer now handles natural language cost queries across eight domains. You can ask why costs increased last week, which accounts drove the change, and what savings opportunities exist. The tool explores your cost data, forms hypotheses, and shows each API call for transparency. It moves FinOps from spreadsheet analysis to conversational investigation.

Graviton5 processors add a hardware dimension to the cost story. AWS claims up to 40% better price-performance compared to non-Graviton instances, with over 90,000 customers already running Graviton workloads. Up to 60% less energy for equivalent performance connects cost reduction to sustainability commitments.

FOCUS 1.2 Data Exports, generally available since November 2025, standardise billing data across AWS, Azure, GCP, and Oracle Cloud. Fourteen additional columns cover invoice reconciliation, capacity reservation tracking, and virtual currency support. For organisations running hybrid or multi-cloud estates, a common billing format turns monthly close processes from days into hours.

How Softcat helps you take control

We hold AWS Premier Tier Services Partner status. That is the highest level available. At re:Invent 2025, we were named EMEA AWS Public Sector Consulting Partner of the Year. Our FinOps practice follows the FinOps Foundation's Crawl-Walk-Run framework, combining commercial optimisation with technical rightsizing. NHS Digital achieved 25% savings across AWS and Azure. The AA reduced public cloud spend by 27%. These are verified outcomes from UK organisations managing real cloud estates.

Our approach covers both sides of the FinOps equation. Commercial optimisation addresses Savings Plans, Enterprise Discount Programmes, and framework agreements. Technical optimisation targets rightsizing, eliminating idle resources, and workload scheduling.

Making Cloud spend work harder

Cloud cost management has moved beyond simple compute rightsizing. Database workloads, AI inference costs, and multi-cloud billing complexity demand tools and practices that didn’t exist two years ago. AWS has built them. The organisations capturing value will be those that deploy them deliberately.

Whether you are establishing FinOps disciplines or extending cost governance to AI workloads, Softcat can help. We will evaluate your cloud estate and plan an approach that turns waste into investment capacity. Please get in touch with our Sales team.