Active remediation, FinOps embedded in engineering. One Optimizer cut daily cloud spend 50% in 9 weeks — no migration, no rebuild, just hygiene.
ZopDev Research · Edition 1 · June 2026 · 23 min read
The State of
Cloud 2026
An Emerging Markets view of global enterprise cloud. How enterprises in India, Southeast Asia, the Middle East, and Latin America actually run cloud — grounded in live telemetry from 25 production cloud accounts, not surveys.
16.9%
of every enterprise cloud dollar is provably recoverable waste, sitting in plain sight. The true figure is likely 30–40%. This is the defensible floor.
Scroll to enter the evidence file
Live telemetry
Spend analyzed$2.22M / yrLive resources25,225Cloud accounts25Enterprises13Regions32Detection rules315+Observation window69 daysHypotheses confirmed10 / 10Recoverable waste16.9%
Orientation
An evidence file, read in exhibits.
Most “State of Cloud” reports are written from Silicon Valley, for Silicon Valley. This one is grounded in primary telemetry from 25 production accounts across 13 emerging-market enterprises. A reading excerpt — highlighted rows are reproduced in full below.
—Executive SummaryIn this excerpt
01Hypothesis ScoreboardIn this excerpt
02The Defensible NumberIn this excerpt · full
03The State of Enterprise CloudIn this excerpt · full
04Where the Waste LivesIn this excerpt · full
05The Detection-to-Action GapFull report
06The Architectural PictureFull report
07An Emerging Markets ViewFull report
08Predictions for 2027Full report
09The GPU ShiftFull report
—Closing · What to Watch NextIn this excerpt
Executive Summary
Cloud has stopped being new. So why is the bill still a mystery?
After fifteen years of migration, the question is no longer whether to be on the cloud. It is why the cloud bill is what it is — and what it could look like if we treated it like a product.
Waste is hygiene, not strategy
16.9% of every cloud dollar is recoverable waste — and 87% of those items are forgotten resources, not bad architectural decisions.
Detection vastly outpaces action
Cloud teams identify waste roughly 30× faster than they fix it. The bottleneck is execution, not awareness.
Three personas, three trajectories
Optimizers cut spend 50% in 9 weeks. Drifters grow 10% a quarter without intent. Scalers grow 4× and bake in tomorrow’s waste. Most enterprises are Drifters.
The Emerging Markets lens reveals different mechanics
Currency exposure, sovereign cloud pressure, and labor-cost economics shape cloud decisions in ways US-centric reports systematically miss.
The GPU shift is the next inflection
Today, AI/GPU spend is under 2% of enterprise cloud bills. By Q2 2027 it crosses 15%. The same anti-patterns return — with three more zeros on every waste number.
Source: cohort telemetry · 25 production accounts · 13 enterprises · 69-day window (Mar 20 – May 27, 2026)
Detection in place, action absent. Waste grows quietly, 5–10% a quarter. The recommendations exist; nobody owns them. The default state of enterprise cloud.
Growth-stage spend. New regions, workspaces, accounts every week — each new service spins up its own NAT gateway, scratch volume, snapshot policy.
Source: cohort run-rate classification · shares of the 13-firm cohort, rounded
25cloud accounts
13enterprises
69days observed
32cloud regions
Instruments: native AWS / Azure / GCP resource-discovery & cost APIs + a 315-rule recommendation engine · $420,554 measured in-window · 36 service types · FMCG, e-commerce, consumer internet, B2B logistics, beverages, creator-economy SaaS
Roughly $1 in every $6 of cloud spend is recoverable through hygiene alone.The economics, in one line — before any architectural change, renegotiation, or repatriation
Chapter 01 · Hypothesis Scoreboard
Ten hypotheses, written before analysis began.
Most industry reports describe what the data showed. We did the opposite: before opening a single dashboard, we wrote ten falsifiable hypotheses. This is the only honest way to do research. Tap any row for the rationale.
Account count and spend are not the same shape. AWS holds 53% of cohort accounts but takes 65.5% of every dollar; Azure punches above its weight on far fewer accounts, and GCP is the long tail concentrated in product and SaaS.
Six of thirteen are single-cloud AWS, four single-cloud Azure, two single-cloud GCP. “Multi-cloud strategy” in industry conversation is often code for failover plans nobody tested — parallel production workloads remain rare.
Three environments hold three-quarters of the provable savings. Remediation effort should be concentrated where the money is, not spread evenly across the estate.
Unattached volumes, orphan snapshots, abandoned IPs — forgotten resources, not bad architectural decisions, dominate the waste ledger. Each takes about five minutes to fix.
The one hypothesis we under-called. Fewer than 1% of detected waste signals are formally applied through any remediation workflow. Detection has been solved; execution has not.
Scheduling alone — nights and weekends off — recovers two-thirds of dev VM cost. Almost no development environment in the cohort had a schedule attached.
Every Databricks workspace spins up its own NAT gateway and Premium SSD scratch volumes — including non-production workspaces where Premium storage buys nothing.
Every developer creates a snapshot before risky changes; almost no one deletes them. Snapshots can exceed live volumes by 14× — charged forever, referenced by nothing.
Not a cost problem — a resilience problem. The reason is never “we made a deliberate choice.” It is always “we ran the migration script in 2021 and never went back.”
For now. By Q2 2027 we predict the average bill is more than 15% AI/GPU — driven by inference, not training. The same anti-patterns return with three more zeros.
16.9 cents of every cloud dollar is recoverable waste, sitting in plain sight. For an enterprise spending $10M annually on cloud, that is $1.69M on the table — recoverable without architectural changes, vendor renegotiations, or workload repatriation.
Across the global enterprise cloud market — near $680B in 2026 — the implied addressable waste is north of $100B annually. In Indian-rupee terms, the addressable waste at a single mid-sized enterprise can exceed ₹50 crore per year, and the currency conversion makes it sharper: the cloud bill is USD-denominated while the revenue defending it is not.
Annualized cloud spend · cohort$2.22M$2,222,930 total
Recoverable savings · proven$375K$375,217 annualized
Waste-to-spend ratio16.9%$375,217 ÷ $2,222,930
Source: cohort telemetry + 315-rule recommendation engine · only high-confidence, provable items counted
We only counted recommendations we could prove with high confidence. We excluded architectural rebuilds, workload repatriation, commitment renegotiations, storage-tier migrations without access evidence, and application-layer efficiency.
Including these would likely double the number — the true recoverable waste is probably 30–40% of cloud spend. But we refuse to claim a number we cannot prove. When we say 16.9%, we mean the central estimate; the true industry-wide value is probably in the 12–22% range.
Method & caveats: Appendix A, C & D of the full report · challenges: contact@zop.dev · 14-day response window
The boardroom question is not “can we cut cloud spend?” It is “why haven’t we already?”16.9% is the headroom you don’t know you have
Spend by provider tells a different story than account count. Azure has fewer accounts in our cohort but commands disproportionate enterprise spend. GCP is the long tail. AWS still takes two-thirds of every cloud dollar.
Share of spend100% = $2.22M / yr
AWS · 65.5%
Azure · 25.7%
8.7%
Share of accounts100% = 25 accounts
AWS · 53%
Azure · 28%
GCP · 19%
AWS — dominant in spend & accountsAzure — punches above its weightGCP — concentrated in product / SaaS
Account count and spend are not the same shape — Azure commands disproportionate spend on far fewer accounts.
AWS only
6 / 13Single-cloud AWS — the modal pattern · 46.2% of cohort
Azure only
4 / 13Single-cloud Azure — concentrated in regulated industries · 30.8%
GCP only
2 / 13Single-cloud GCP — concentrated in product / SaaS · 15.4%
All three
1 / 13Genuinely multi-cloud in production — exactly one enterprise · 7.7%
The “multi-cloud strategy” of industry conversation is often untested failover plans, not parallel production workloads.
| Footprint pattern | Example (anonymized) | Regions | $ / region / mo |
|---|---|---|---|
| Global sprawl, one tenant | Global sporting goods · Azure | 32 | ~$60 |
| Concentrated production | Global CPG · Azure | 3 | ~$5,900 |
| Single-region monolith | Consumer internet · AWS | 2 | ~$12,500 |
| Spread without intent | E-commerce · AWS | 19 | ~$560 |
Two of these patterns reflect deliberate architectural choices. Two reflect accident. Footprint is now a choice — most enterprises have not made it consciously.
Some enterprises cut spend by half in a quarter. Others quietly inflate by 4×. Optimizers shed waste fast, Scalers accumulate it fast, Drifters stay stuck in between. Every cloud bill is on one of these three trajectories.
−50% in 69 days
Active FinOps embedded in engineering: recommendations triaged, owners assigned, savings tracked. No migration. No rebuild. Pure hygiene.
−50%The Optimizer · daily spend
$1,565 → $786 · 69 days · pure hygiene
+10%The Drifter · weekly spend
$5,787 → $6,361 · 8 weeks · no launch, just drift
4.1×The Scaler · weekly spend
$1,429 → $5,918 · 6 weeks · legitimate growth, embedded waste
The “average enterprise cloud bill” is a meaningless number. One customer halves spend while another quadruples it — in the same quarter.
The question every CTO should ask: which trajectory are we on — and is it the one we chose?Chapter 03 · the bimodal economy
Vendor-led narratives focus on big strategic moves — Reserved Instances, Savings Plans, Spot. These account for only 6% of the recommendations we flagged. The real waste is mundane: things nobody remembers creating.
Orphan
87.4%Forgotten resources: unattached volumes, orphan snapshots, abandoned IPs · ~5 min each to fix
Discount
6.3%RIs, Savings Plans, Spot — the entire “strategic” category · 6.3% of items
Rightsizing
3.7%Over-provisioned instances and volumes · 3.7% of items
Idle
1.3%Running but unused resources · 1.3% of items
Schedule
0.9%Missing on/off schedules for dev & test · 0.9% of items
The biggest cloud waste problem isn’t that you bought the wrong thing. It’s that you forgot to delete the right thing.
01
Provision generously, clean up rarely
The friction of provisioning is near zero; the friction of de-provisioning is enormous.
02
No one owns the lifecycle
Accounts are organized by team, environment, or application. Almost never by lifecycle.
03
Snapshots compound silently
Every dev creates a snapshot before risky changes. Almost no one deletes them. Snapshots can exceed live volumes by 14×.
Source: recommendation engine · 315+ rules applied uniformly across all 25 accounts
Chapter 04 · continued
The hall of shame.
Fix only these ten things, and you recover 80% of the waste. None are strategic decisions to revisit — they are hygiene to install.
| # | Anti-pattern | Universality | $ / yr at stake |
|---|---|---|---|
| 01 | Orphan EBS snapshots — no source volume, charged forever | 100% of AWS users | $80K+ |
| 02 | Unattached EBS volumes — single volumes leak $2,500/yr | 100% of AWS users | $25K+ |
| 03 | Premium SSD in non-prod Databricks (Azure) | 100% of Databricks on Azure | $66K+ |
| 04 | EC2 not covered by Savings Plans | ~90% of prod fleets | $50K+ |
| 05 | Single-AZ production RDS — not cost, risk | Every AWS enterprise | Resilience risk |
| 06 | Dev/test workloads with no schedule | 100% of dev environments | $40K+ |
| 07 | Windows VMs not using Azure Hybrid Benefit | Common in regulated industries | 40% on license |
| 08 | On-demand for stateless / batch workloads | Common in cloud-native shops | $30K+ |
| 09 | x86 where Graviton/ARM would work | 80% of compute fleets | $25K+ |
| 10 | Stopped instances still incurring EBS charges | Found in every account | $5K+ |
If your platform team cannot recite this list from memory, you have a 16.9% problem.
Recap · The evidence so far
Four numbers to carry into the boardroom.
10/10hypotheses confirmed
All ten pre-registered hypotheses confirmed. The conventional wisdom is right — the will to act is what’s missing.
16.9%recoverable waste
Provably recoverable today, without architectural change. The true figure is likely 30–40%.
65.5%of spend on AWS
Genuine multi-cloud is 1 in 13. Footprint is mostly accidental — a choice most firms haven’t made consciously.
87.4%of waste is orphans
Forgotten resources, not strategy. Fixing ten anti-patterns recovers ~80% of the total.
Closing · What to Watch Next
Three trends. Three counter-trends. One bet.
One bet, willing to be publicly wrong: by the 2027 edition, more than half of our cohort will have crossed into Optimizer territory — not because tooling improved, but because the next downturn made cloud cost a board priority. The bet is on the cycle, not the technology.
Three trends
- GPU spend crosses 15% of enterprise cloud, up from under 2% today.
- Auto-remediation displaces detection-only tooling in procurement.
- Sovereign cloud becomes board-level globally, not just in Europe.
Three counter-trends
- Repatriation gains traction for steady-state workloads in currency-pressured economies.
- EM enterprises increasingly question 3-year RI commitments on FX exposure.
- “Quiet” Drifter cost growth outpaces flashy AI bills as the dominant budget overrun.
“The most important thing you can do with this report is challenge it.” We published our methodology so you can replicate, contest, and improve on it.