Reserved Instance & Savings Plan Coverage

Coverage tracking shows how much of your compute spend is covered by commitments (reserved instances, savings plans, spot) vs running on-demand. The goal: minimize on-demand spend for steady-state workloads.

What coverage tracking shows

Open Savings & Coverage from the left sidebar in the console. The overview shows your spend split:

Pricing model	Description	Discount vs on-demand
On-demand	Pay-as-you-go, no commitment	0% (baseline)
Reserved instances (RI)	1-year or 3-year commitment to a specific instance type	30–60%
Savings plans (SP)	1-year or 3-year commitment to a $/hr spend level	20–40%
Spot	Bid on unused capacity, can be interrupted	60–90%

A healthy coverage ratio depends on your workload profile, but most teams with steady-state production workloads should target 60–80% coverage (RI + SP combined).

Daily breakdown chart

The daily chart shows how your spend distributes across pricing models each day for the past 90 days. This helps you spot:

Consistent on-demand spend — a flat band of on-demand at the top means you have steady workloads that could be committed
RI/SP expiration cliffs — a sudden jump in on-demand spend when a reservation expires
Spot interruption patterns — days where spot spend drops and on-demand picks up the slack

Example daily breakdown:

Day       | On-demand | RI      | SP      | Spot   | Total
----------|-----------|---------|---------|--------|--------
Mar 15    | $142      | $280    | $195    | $83    | $700
Mar 16    | $138      | $280    | $195    | $87    | $700
Mar 17    | $245      | $180    | $195    | $80    | $700
           ↑ RI expired — on-demand jumped $100

Coverage gap analysis

The Coverage Gaps tab identifies specific instance families and regions where you’re paying on-demand prices for consistent usage. Each gap includes:

Instance family and region — e.g., m5.xlarge in us-east-1
Average daily on-demand hours — how many hours/day this instance type runs on-demand
Consistency score — percentage of days in the past 30 where this instance type was running (90%+ is a strong candidate for commitment)
Potential monthly savings — estimated savings if you committed to an RI or SP

Gaps are ranked by potential savings, so you can focus on the biggest opportunities first.

Commitment planner

The Commitment Planner analyzes your usage patterns and recommends optimal commitments. For each candidate workload, it shows:

1-year vs 3-year comparison

Option	Monthly cost	Savings vs on-demand	Break-even
On-demand	$730/mo	—	—
1-year RI (no upfront)	$475/mo	35% ($255/mo)	Month 1
1-year RI (all upfront)	$438/mo	40% ($292/mo)	Month 1
3-year RI (no upfront)	$328/mo	55% ($402/mo)	Month 1
3-year RI (all upfront)	$292/mo	60% ($438/mo)	Month 1
Compute SP (1yr)	$511/mo	30% ($219/mo)	Month 1
Compute SP (3yr)	$365/mo	50% ($365/mo)	Month 1

Key factors the planner considers

Usage consistency — workloads running <80% of the time are flagged as risky for commitments
Instance family trends — if you’re likely to change instance types (e.g., migrating from m5 to m7), a Compute Savings Plan is more flexible than an RI
Existing commitments — the planner accounts for RIs and SPs you already have, so it only recommends incremental commitments
Expiration timeline — shows when current commitments expire and what the on-demand impact will be

Step-by-step: improving coverage

Open Savings & Coverage and check your current coverage percentage. If it’s below 50% and you have steady-state workloads, there’s likely room to save.
Go to the Coverage Gaps tab. Sort by potential savings.
For the top 3–5 gaps, open the Commitment Planner to see the 1-year vs 3-year analysis.
Decide on your commitment strategy:
- Conservative: 1-year no-upfront RIs or Compute Savings Plans. Lower savings but lower risk.
- Aggressive: 3-year all-upfront RIs for workloads you’re confident will persist. Highest savings.
- Flexible: Compute Savings Plans that apply across instance families — good if you’re actively modernizing.
Purchase the commitments in your cloud provider’s console. Come back to Xplorr after 24 hours — the coverage chart will update to reflect the new commitments.

Common mistakes

Buying 3-year RIs for workloads that might be migrated or decommissioned. If there’s any chance the workload moves to a different instance family, use a Compute Savings Plan instead.
Over-committing. Don’t try to hit 100% coverage. Leave headroom for variable workloads, auto-scaling, and spot usage. 70–80% is a practical target.
Forgetting to track expiration dates. An RI expiring without renewal means an instant cost spike. Set a calendar reminder 60 days before expiration, or use Xplorr’s expiration alerts.
Buying per-instance RIs when Compute Savings Plans are available. Savings Plans are almost always more flexible for the same discount level. Only use instance-specific RIs when the incremental discount (5–10%) justifies the reduced flexibility.

FAQ

Does Xplorr support Azure reservations? Yes. Azure Reserved VM Instances and Azure Savings Plans are tracked the same way. The coverage chart and commitment planner work across all providers.

Can I see coverage for non-compute services? Coverage tracking focuses on compute (EC2, Azure VMs, GCE). For storage commitments (like S3 Intelligent-Tiering or reserved capacity for DynamoDB), check the provider-specific sections in the cost breakdown.

What happens when an RI expires? Xplorr sends an expiration alert 30 days before the RI ends. The coverage chart will show the projected on-demand spike so you can plan a renewal or replacement.

Can I manage RI/SP purchases through Xplorr? Not directly. Xplorr provides the analysis and recommendations — you make the actual purchase in your cloud provider’s console. This keeps Xplorr read-only and avoids accidental commitments.

Key takeaways

Track your coverage ratio and aim for 60–80% for steady-state production workloads.
Use the coverage gap analysis to find the biggest savings opportunities.
Prefer Compute Savings Plans over instance-specific RIs unless you’re certain the workload won’t change instance families.
Monitor expiration dates to avoid unexpected on-demand cost spikes.