AWS cost audit
We cut your AWS bill. The fee is one month of what we save.
From the second month the saving is yours and we are done. No retainer, no engagement fee, and nothing owed if we find nothing worth changing.
Measured against a change list you approve first, where every line carries its own monthly figure. Not against your invoice, which moves for reasons neither of us controls.
- 01
Send two exports
Two CSVs out of Cost Explorer, about two minutes of your time. No IAM role, no access keys, no read-only user.
- 02
Get a change list
Every line carries its own monthly figure. You approve it before anything is touched, and you are free to walk away with it.
- 03
Pay once, from the saving
We implement, then invoice one month of the verified reduction. From the second month the whole saving is yours.
AWS already tells you where you could save. So why is the bill still high?
Cost Explorer, Trusted Advisor, Compute Optimizer and Cost Optimization Hub all produce recommendations, and they are free. Almost nobody running on AWS is short of recommendations. What is scarce is implemented ones.
The gap between the two is not technical. It is four ordinary organisational problems, and every one of them is a reason the list is still sitting there next quarter.
- 01 Nobody owns it
- Cost sits between the people who can change it and the people who feel it. Engineering has the access and no mandate; finance has the mandate and no access. A recommendation that arrives in a console neither side opens by default is not assigned to anyone, and unassigned work does not get done.
- 02 The risk is asymmetric
- “Resize this instance” is a production change. The upside is forty dollars a month and the downside is an incident with your name on it. An engineer who declines that trade in a sprint they are already behind on is not being obstructive, they are doing the arithmetic correctly.
- 03 Individually small, collectively a project
- Each item takes an hour. Forty items take a quarter, once each one has a test, a deploy window and a way back. That is the point at which it stops competing with other tickets and starts competing with the roadmap, which it loses.
- 04 Nothing closes the loop
- The change ships and the invoice moves for six unrelated reasons in the same month, so nobody can show it worked. Work that cannot be shown to have worked does not get repeated, and the next batch of recommendations is ignored a little faster than the last.
This engagement is shaped around that gap rather than around finding more savings. We take the ownership problem off your team by doing the work, we sequence by production risk as well as by size so the cheap and safe changes land first, and the change list you approve beforehand is what the result is measured against — which is what makes it possible to say afterwards whether it worked.