EDP NegotiationSavings Plans OptimizationReserved Instances StrategyEC2 Right-SizingS3 Cost ReductionEgress NegotiationMigration CreditsSupport Tier AdvisoryMulti-Cloud LeverageBedrock AI PricingEDP NegotiationSavings Plans OptimizationReserved Instances StrategyEC2 Right-SizingS3 Cost ReductionEgress NegotiationMigration CreditsSupport Tier AdvisoryMulti-Cloud LeverageBedrock AI Pricing
Service · Reserved Instances

AWS Reserved Instance strategy, portfolio-wide.

Standard vs Convertible analysis, RI Marketplace strategy, cross-region planning, size-flexibility optimization, and portfolio management across RDS, Redshift, ElastiCache, OpenSearch, DynamoDB, and legacy EC2 RIs.

$2.4B+
AWS spend reviewed
500+
Engagements
38%
Avg reduction
$340M+
Documented savings
Overview

RIs aren't dead — they
just live in the database tier now.

The introduction of Savings Plans changed the calculus for EC2 commitments. For most organizations, the right answer for EC2, Fargate, and Lambda is now a layered Savings Plans portfolio. But that decision does not extend to the rest of AWS. RDS, Redshift, ElastiCache, OpenSearch, and DynamoDB still rely on Reserved capacity as their primary commitment vehicle — and they often represent 30–50% of the bill.

That makes Reserved Instance strategy a quietly under-managed area. Many FinOps teams shifted attention to Savings Plans and left their database RI portfolios on autopilot. The result is suboptimal coverage in exactly the spend categories that have grown fastest: managed databases, cache, search, and analytics. We rebuild that portfolio with the same rigor we apply to compute.

We also clean up the EC2 RI tail. Most organizations still carry legacy Standard and Convertible RIs that no longer match their workloads. Some convert. Some sell. Some expire. Each disposition has a specific economic answer, and we work through every line item.

What we optimize

Every RI category, every family.

01RDS, Aurora, and Redshift RI strategy+
Database commitments are inherently stickier than compute — schemas don't move overnight. We model utilization at the instance-class level, account for size flexibility within engine families, and structure one and three-year RIs with the right partial-upfront vs all-upfront mix.
02ElastiCache and OpenSearch coverage+
Cache and search workloads run hot and steady. Coverage targets here are often higher than for general-purpose compute. We model the floor with a higher confidence threshold and lean toward three-year commitments where the workload has structural permanence.
03DynamoDB Reserved Capacity+
Reserved Capacity on provisioned-mode DynamoDB tables is among the highest-discount commitments in AWS. We identify which tables warrant Reserved Capacity, which should move to on-demand, and which sit in a hybrid pattern that benefits from partial coverage.
04Standard vs Convertible decision framework+
Convertibility costs roughly 3–5 percentage points of discount but preserves optionality across instance families. We apply a workload-by-workload framework: high-confidence steady-state goes Standard, anything on a migration path or with size-uncertainty goes Convertible.
05RI Marketplace recovery+
Standard EC2 RIs in eligible regions can be sold on the AWS RI Marketplace. We list, price, and manage the sale of unused RIs — typical recovery is 40–70% of remaining value depending on family, region, and term remaining.
06Cross-region and size-flexibility planning+
Regional RIs apply across availability zones; size flexibility lets one RI cover multiple instance sizes in the same family. We structure the portfolio to maximize flexibility-weighted coverage, particularly for workloads that shift between AZs or scale up and down within a family.
Process

Portfolio audit to optimized state in 4–6 weeks.

1.

Inventory and utilization audit

Map every active RI and Reserved Capacity commitment. Score utilization, identify expiring tail, and surface unused or under-attributed RIs.

2.

Coverage modeling

Build database and EC2 coverage models. Identify gaps, model the optimal one-year vs three-year ladder, and quantify the recovery potential from RI Marketplace.

3.

Execution and rebalancing

Sequence purchases and Marketplace listings, set up quarterly rebalancing, and integrate the RI portfolio with your Savings Plans and EDP commitments.

Results

What clients actually save.

Related services

Often combined with RI strategy.

Your RI portfolio
is under-managed.

$2.4B+ in AWS spend reviewed. We rebuild your RI portfolio in 4–6 weeks.

How we deliver

Four phases. One outcome.

01

Diagnostic (week 1)

Cost and Usage Report ingestion, contract review, EDP scorecard. You get a benchmark against 500+ comparable deals.

02

Strategy (weeks 2-3)

Negotiation positions, BATNAs, target outcomes by line item. We build the playbook and the supporting models.

03

Execution (weeks 4-9)

We sit in your seat opposite AWS. You stay in control of the relationship; we shape the deal.

04

Hand-off (week 10+)

Signed terms, internal playbook, monitoring framework. So you can defend the deal at the next renewal yourself.

Questions

Frequently asked. Directly answered.

Standard vs Convertible Reserved Instances — when does each make sense?+

Standard RIs deliver deeper discounts (up to 72%) but lock you to instance family, OS and region. Convertible RIs offer ~54% off and let you change attributes during the term. Use Standard for predictable, long-lived workloads; use Convertible where your architecture is evolving.

How does Size Flexibility work?+

Within an instance family (e.g., m5) and region, an RI applied to one size automatically covers fractional usage at other sizes via Normalized Units. A 1×m5.xlarge RI covers 2×m5.large, and so on. This is the single most under-used RI mechanic.

Can I sell unused Reserved Instances?+

Yes — Standard RIs can be sold on the AWS RI Marketplace. Discounts versus the original price are typical, but it beats holding unused capacity. Convertible RIs are not sellable but can be modified.

Should I still buy RIs in 2026 or move entirely to Savings Plans?+

RIs remain relevant for RDS, Redshift, ElastiCache, OpenSearch and DynamoDB Reserved Capacity — services where Savings Plans don't apply. For EC2, the default answer is Compute Savings Plans unless you have a specific Standard-RI advantage.

What happens to my RIs if I move regions or change instance families?+

Standard RIs become stranded — they only apply within the original region/family. Convertible RIs can be modified to match. Marketplace RIs can be resold. Plan migrations 60-90 days ahead of move to drain or convert positions.