AWS Cost Audit

Find out exactly where your AWS bill leaks.

An AWS cost audit is an independent, read-only review of your account that maps every place you are overpaying: oversized compute, cold storage on hot tiers, missing Savings Plans, silent data transfer, and resources nobody remembers launching. We look, we size each opportunity, and we hand you a prioritised report. We do not change anything during the audit. Delivered by AWS-certified engineers.

Book a free 30 minute review →See the fix service

Audit scopeREAD ONLY

Sized and prioritised.

ComputeEC2, ECS, Lambda

StorageS3, EBS, snapshots

CommitmentsRIs, Savings Plans

TransferEgress and inter-AZ

WasteIdle and orphaned

ReportEffort vs saving

What The Audit Covers

Six places an AWS bill quietly grows.

Bills rarely balloon from one big mistake. They creep up from six ordinary places, and an audit is the work of checking each one against what you actually run.

01 · COMPUTE

Rightsizing compute

EC2, ECS and Lambda checked against real utilisation. Oversized instances, over-provisioned tasks, and functions with too much memory are the most common single line of savings. We size to what the workload uses, not to what someone guessed at launch.

EC2ECS / FargateLambda

02 · STORAGE

Storage tiering

S3 objects sitting on Standard that belong on Infrequent Access or Glacier, EBS volumes over-provisioned or unattached, and snapshots that pile up forever. Storage is cheap per gigabyte and expensive at scale when nothing tiers or expires.

S3 tiersEBSSnapshots

03 · COMMITMENTS

Reserved capacity and Savings Plans

Steady workloads running at on-demand rates are money left on the table. We check your Savings Plans and Reserved Instance coverage against your actual baseline, and flag both the gaps and any commitments you are not fully using.

Savings PlansReserved InstancesCoverage

04 · TRANSFER

Data transfer costs

The line item that surprises everyone. Cross-AZ chatter, NAT gateway egress, and traffic leaving AWS all add up quietly. We trace where the bytes go and where an architecture change or endpoint would cut the cost.

EgressInter-AZNAT / endpoints

05 · WASTE

Idle and orphaned resources

Unattached EBS volumes, idle load balancers, forgotten dev environments running around the clock, old AMIs, and stale ECR images. Individually small, collectively a standing charge for nothing. We find them and list what is safe to remove.

Idle resourcesOrphaned volumesScheduling

06 · ARCHITECTURE

Architecture-level savings

The biggest wins are usually structural: a self-managed database that could be RDS, a fleet that could be Fargate or Spot, a cron box that could be a scheduled Lambda. We flag where a managed service or a design change lowers the run cost, not just the instance size.

Managed servicesSpot / FargateServerless

The Deliverable

What lands on your desk.

A report you can act on without us, written for the engineer who fixes it and the finance lead who signs it off.

SIZED

Each opportunity in dollars

Every finding sized in dollars where it can be estimated, so you can see the return before you spend a minute on it. We are honest that estimates are estimates: the real number comes from your bill after the change, and we say which findings are precise and which are directional.

RANKED

Prioritised by effort vs saving

Findings ordered so the quick wins are separated from the work that needs an architecture decision. You can clear the easy money in a week and plan the structural changes deliberately, rather than staring at an undifferentiated list.

EVIDENCED

The evidence behind each

Every finding shows the data it came from: the utilisation graph, the coverage gap, the idle resource. No unexplained scanner output, and nothing you have to take on trust. If you challenge a finding, the evidence is right there.

Start Free

The 30 minute version, at no cost.

You do not have to commission a full audit to get value. The free review is the shallow end of the same work.

Book a free 30 minute architecture review and we will look at your setup with you and leave you with at least one concrete cost-saving recommendation, at no cost and with no obligation. The full AWS cost audit is the deep version of that same review: instead of one recommendation in half an hour, you get the whole account mapped, every opportunity sized, and a prioritised report you can hand to your team. Start with the free call, and step up to the full audit only if the account warrants it.

Where The Audit Fits

Find, fix, sustain. Three jobs.

The cost cluster splits the way the security one does. Knowing which service does which job means you buy the one you actually need.

Step 1 · Find

The audit finds the savings

A read-only review that maps and sizes every opportunity. It tells you where the money is; it does not move it. That is this page.

Book a free review

Step 2 · Fix

Optimization implements them

The hands-on work of rightsizing, tiering, buying commitments and re-architecting to capture the savings the audit found, with the results measured.

AWS cost optimization

Step 3 · Sustain

FinOps keeps it down

Tags, budgets with owners, and anomaly alerts so the bill does not creep back the next quarter. An audit is a snapshot; FinOps is the habit.

FinOps practice

If you want the plain-English version of why AWS bills climb in the first place, before any of this, our why is my AWS bill so high explainer is written for founders rather than engineers.

Proof, Published

Real numbers from real engagements.

Not projections. These are outcomes we have published, each written up in full.

UK HealthTech

38% cut, 214k dollars annualised

What we did. Rightsizing, Reserved Instances, and architecture optimizations on a regulated HealthTech account.
The result. A 38% reduction in monthly AWS cost, worth 214k dollars a year.
Written up. The full engagement is in our UK HealthTech cost optimization case study.

Heroku to AWS

60% median reduction

What we did. Migrated applications off Heroku onto AWS ECS Fargate with zero-downtime cutover.
The result. A 60% median cost reduction across those migrations, with better burst handling.
Written up. The full engagement is in our Heroku to AWS ECS case study.

Both are specific engagements, not a promise for yours. What an audit gives you is a figure grounded in your own account, with the evidence behind it, instead of someone else's percentage.

Common Questions

What buyers ask before a cost audit.

If your question is not here, ask it on the free call. We would rather scope it honestly than sell you an audit your account does not need yet.

What access do you need for a cost audit?+

A read-only IAM role, scoped to billing, Cost Explorer, and the resource metadata we need to see what is running and how it is sized. Nothing write-level, and nothing that lets us change your environment. The audit is a look, not a touch. If your policy requires it, we work from an even tighter read-only scope and tell you exactly which permissions each finding needed.

How long does an AWS cost audit take?+

Scope drives it, and we would rather size it honestly than quote a duration before we have seen the account. A single-account environment is usually a short exercise measured in days once we have access. A multi-account organisation with several teams and a mix of legacy and current workloads takes longer, because most of the time goes into understanding what is actually deployed rather than into running tools. We agree the scope and the timeline before we start.

What does a cost audit cost?+

There is a free version and a full version. The free 30 minute architecture review looks at your setup and leaves you with at least one concrete cost-saving recommendation, at no cost and with no obligation. The full audit, which goes deep across the whole account and produces a sized and prioritised report, is scoped on that call, and we share the price before any commitment. We do not publish an invented number here, because a quote that ignored your account would be worthless.

Do you implement the fixes too?+

Yes, but they are separate so the audit stays honest. The audit finds and sizes the savings; implementing them is our cost optimization service, priced separately so nobody is incentivised to inflate the findings. Plenty of clients take the report, action the quick wins themselves, and bring us in only for the changes they do not want to own. If you want it all done for you, the two run back to back.

How much can we realistically save?+

Honestly, it depends on your starting position, and we will not promise a percentage before we have looked. What we can point to is published results: a UK HealthTech client cut costs by 38%, worth 214k dollars annualised, and teams leaving Heroku for AWS ECS have seen a 60% median reduction across our migrations. Those are real engagements, not a guarantee for yours. The audit gives you a sized, account-specific figure instead of a marketing number.

Is our data safe during an audit?+

The audit uses read-only access, so we can see configuration and billing but cannot change or move anything. We look at how resources are sized and spent on, not at the contents of your databases or storage. HAZERCLOUD holds ISO 27001:2022 and ISO 9001:2015 certification, so the team reviewing your account is audited against a recognised information security standard itself.