Optimize: the four levers you have

Outcome

By the end of this lesson, you will be able to name the four levers of cloud cost optimization and rank them by typical impact and time-to-value.

1. Concept

Optimize is the second phase of the FinOps lifecycle. There are exactly four levers. Every optimization initiative is some combination of these four. Ranking them by impact and time-to-value is the most important sequencing decision in a FinOps program.

LEVER                  IMPACT (typical)   TIME-TO-VALUE   RISK
─────────────────────────────────────────────────────────────────
1. Eliminate waste     20–40% of bill     Days–weeks       Low
2. Rightsize           10–30% of bill     Weeks            Medium
3. Rate optimize       5–25% of bill      Weeks–months    Medium–High
4. Architect / design  varies (5–50%+)    Quarters         High

Lever 1: Eliminate waste

Stop paying for things nobody uses. Scheduling non-prod off-hours, terminating idle resources, releasing orphan storage, cleaning forgotten snapshots, deleting unused Elastic IPs. This is the Top 10 from M0.1 L5.

Why first: lowest risk, fastest payback, no architectural debate. Stopping a dev environment at night does not require a redesign. The savings hit the next bill.

Where it ends: when the avoidable-spend KPI drops below 5–8 percent of bill. Below that, the remaining waste is in the architecture, not in operations.

Lever 2: Rightsize

Run things at the right size. Drop oversized instances a class or two. Move RDS instances down a tier when CPU and connections are persistently low. Switch EBS from gp2 to gp3 (same performance, lower cost). Move under-utilized K8s workloads to lower requests / limits.

Why second: higher reasoning load than waste elimination but still mostly mechanical. The rules engine (T2.M2.1) computes most rightsizing candidates automatically. The remaining question is workload risk tolerance.

Where it gets hard: rightsizing databases (cannot be auto-remediated, must be done in a maintenance window) and rightsizing workloads with bursty traffic (the P99, not the average, defines the floor).

Lever 3: Rate optimize

Pay less per unit. Reserved Instances, Savings Plans, Committed Use Discounts (CUDs), Spot, Azure Hybrid Benefit. Negotiate enterprise discount programs. Move steady workloads from on-demand to commitment-backed pricing.

Why third, not first: commitments only pay back on workloads you actually run for the full term. Committing to capacity you then schedule off is a value-destroying move. The right sequence is schedule first → rightsize → commit on the proven floor.

Where it gets dangerous: over-commitment. A 1-year RI on capacity you only run six months returns negative ROI. A 3-year RI on the wrong instance family compounds the mistake. (See M0.3 L2 and T4.M4.7.)

Lever 4: Architect / design

Change the workload to cost less. Replace a 24/7 service with a serverless equivalent that scales to zero. Move an analytics pipeline from EMR to Athena. Re-implement a real-time feature as event-driven. Decompose a monolith so non-critical paths can run on cheaper compute.

Why last: highest impact ceiling, highest risk, longest time-to-value. An architectural change touches code, deployment, on-call, SLOs. The ROI calculation must account for engineering time.

Where it shines: the steady-state floor cost after levers 1–3 have been applied. If you are still spending $200K per month on a workload that could be re-architected to $80K, the prize is there. But sequence it after the easy wins.

The lever ranking is not opinion

   IMPACT ─→
   ┌──────────────────────────────────────────────────────┐
   │ 4 ARCHITECT/DESIGN          (highest ceiling, slowest)│
   │ 3 RATE OPTIMIZE             (mid impact, medium speed)│
   │ 2 RIGHTSIZE                 (mid impact, fast)        │
   │ 1 ELIMINATE WASTE           (consistent, fastest)     │
   └──────────────────────────────────────────────────────┘
   TIME-TO-VALUE →

The diagonal: lever 1 is in the bottom-left (fast, modest), lever 4 is in the top-right (slow, biggest). Most teams should travel up-and-right diagonally over 12 months, not jump straight to lever 4.

The most common sequencing mistake

Buying reservations before scheduling. The reservation locks in capacity you intend to schedule down, destroying the savings on both sides. The fix is mechanical: schedule first, observe the resulting steady-state floor for 60 days, then commit on the floor only.

Where ZopNight focuses

Lever 1 (scheduling) and lever 2 (rightsizing recommendations + autoscaling) are first-class. Lever 3 (commitments) is on the roadmap. Lever 4 (architecture) is out of scope; that is application engineering work, not platform automation.

2. Demo

A real (anonymized) optimization sequence at a mid-size SaaS, applied over 6 months:

Cumulative monthly savings by month 6: $43,494. Total project effort: ~140 engineering hours. Effective rate: $310 saved per engineering hour.

3. Hands-on (8 min)

For your own estate, score yourself on each lever:

LEVER 1: Eliminate waste
[ ] Are non-prod resources scheduled off after hours?
[ ] Is the avoidable-spend KPI below 8% of bill?

LEVER 2: Rightsize
[ ] Are rightsizing recommendations reviewed weekly?
[ ] What's the average gap between "recommendation surfaced" and "remediated"?

LEVER 3: Rate optimize
[ ] What's the current commitment coverage (RI + SP + CUD)?
[ ] Was the most recent commitment purchased on the post-schedule floor?

LEVER 4: Architect/design
[ ] Are there any workloads in active redesign for cost reasons?
[ ] Is engineering time being tracked against cost-driven redesigns?

If lever 1 is incomplete, do not move to levers 2–3. The sequence is not “do all four in parallel.” It is “do 1 first, then 1+2, then 1+2+3, then all four.”

4. Knowledge check

Q1

A team has $200K monthly cloud spend, mostly non-prod always-on, and wants to start optimizing. The first move is:

A. Buy a 3-year Reserved Instance pool to lock in savings
B. Schedule non-prod off-hours and weekends
C. Migrate to Kubernetes
D. Hire a FinOps consultant

Q2

A FinOps team reports: “We applied 30 rightsizing recommendations this month and saved $4,100. We applied 1 schedule and saved $24,000.” Which lever has more headroom to keep finding savings?

A. Lever 1 (scheduling)
B. Lever 2 (rightsizing)
C. Both equally
D. Lever 3 (commitments)

Q3

A CFO proposes locking in a 3-year RI on the current peak compute usage to “guarantee savings.” Best response:

A. Agree, locking in is good
B. Push back. 3-year RIs on current peak are at high risk of over-commitment if usage shifts. The defensible move is 1-year commitment on the post-schedule, post-rightsizing floor: not on current peak.
C. Wait one year
D. Use Spot instead

5. Apply

ZopNight is built around levers 1 and 2 because that is where the fast, repeatable savings live for most estates:

• Lever 1 → Schedules (any time you can stop a resource), Recommendations idle/orphan rules
• Lever 2 → Recommendations rightsizing/category rules, VM Autoscaling
• Lever 3 → coverage tracked in Reports → Purchase Type breakdown
• Lever 4 → out of scope; engineering ownership

Related lessons

• L4: Operate: the discipline that beats one-shot wins (next)
• M0.3: Why scheduling beats commitments
• T4.M4.7: Commitments demystified

Glossary terms touched

Optimize · Eliminate waste · Rightsize · Rate optimize · Architect-design lever