Mewo Cut S3 Costs by 44% and AWS Spend by 19% with Automat-it FinOps

Table of Contents

Mewo case study image

Automat-it’s deep AWS knowledge transformed our cloud infrastructure, immediately cutting our biggest expense, storage, by nearly half. Their FinOps strategy reduced our total AWS bill by 19%, giving us the peace of mind to scale securely while keeping our media library protected.

Summary

 

Mewo runs a large audio media platform on AWS, where storage alone accounted for roughly 40% of the company’s monthly bill. Automat-it ran a full FinOps analysis of Mewo’s usage, restructured how its 80TB music library was stored, and locked down deletion controls to protect the data. S3 costs came in 44% below previous months, and total AWS spend dropped 19%.

 

 

About Mewo

 

Mewo is a France-based media startup operating in the audio and music space. It is redefining the sync licensing ecosystem through an AI-assisted platform built by and for music supervisors.
The startup’s mission is to make sync licensing easier for everyone and ensuring AI-generated music never becomes more convenient to license than real music.

 

The Challenge: Storage Eating Almost Half of the AWS Bill

 

Mewo’s AWS spend was dominated by storage, and the way that storage was configured pushed the cost higher than it needed to be. The situation was not sustainable if the startup was to grow and scale at its existing pace. The core issues included:

  • Storing a huge audio catalog: Roughly 80TB of active media spread across about 24.6 million objects meant storage had become the single biggest line item in monthly spend (approximately 40% of the total AWS bill).
  • How to facilitate customer needs without impractical costs: Any track could be streamed at any time, so Mewo needed to align demand with access and the practicality of cost-efficient storage.
  • A lack of transparency on existing policies impacting the final bill: Without knowing exactly what was, and wasn’t, switched on or enabled, Mewo was likely spending unnecessarily in key areas but was unaware.

 

The Solution: Match the Storage Class to How the Library Is Actually Used

 

Mewo brought in Automat-it, an AWS Premier Partner focused entirely on AWS and startups, to conduct a financial audit and fix the storage economics without putting any of the catalog at risk.

Automat-it’s Cloud FinOps practice started with a full cost analysis tied to Mewo’s specific usage pattern as a media business. They were able to identify key cost triggers to address, including:

  • Everything sat in S3 Standard, the most expensive tier. Audio files were kept in Standard because retrieval was unpredictable so the team had defaulted to the always-hot, always-priced-highest option.
  • Backup copies were accumulating without control. A policy had been switched on to create backup copies of files as protection against loss. Those copies were being stored and billed with no lifecycle rules to contain them.
  • Versioning was multiplying the bill. With versioning enabled, Mewo was paying to keep multiple noncurrent versions of the same file, stacking storage costs for data that was rarely, if ever, needed.

 

After identifying cost-saving opportunities, Automat-it recommended and implemented changes:

 

  1. S3 Intelligent-Tiering as the primary storage class. Because a music library has no predictable access pattern, Intelligent-Tiering was the right fit. Frequently streamed MP3s stay in the Frequent Access tier at the same price Mewo was already paying under S3 Standard.
  2. Automatic tiering for cold and high-quality files. Files not downloaded for 90 days, along with high-quality masters, move automatically to the Archive Instant Access tier. That tier delivers up to 68% savings versus Standard while still allowing instant retrieval. An important feature for a catalog where any track might be requested at any moment.
  3. Lifecycle rules for noncurrent versions. To stop versioning from quietly inflating the bill, Automat-it added a policy that moves noncurrent versions to S3 Glacier Instant Retrieval after 30 days and permanently deletes versions older than 180 days. This clears out redundant copies without touching the live files.
  4. A Deny Bucket Policy to protect every object. Automat-it applied a bucket policy that denies a particular action for all principals. No one can permanently delete an object version, which guards the catalog against both accidental deletion and malicious intrusion.

 

Automat-it’s delivery team collaborated with Mewo on the rollout using free DevOps hours, producing an implementation Solution Proposal, getting Mewo’s sign-off, and completing the agreed work.

 

The Results: 44% Lower S3 Costs, 19% Lower Total Spend

 

After a transition month where redistribution of files across new storage tiers completed by AWS, the numbers in the following month showed the impact:

  • S3 costs down 44%. S3 spend came in 44% below previous S3 spend.
  • Total AWS costs down 19%. Reworking the largest line item pulled Mewo’s overall monthly AWS bill down 19% versus the previous mark.
  • The full 80TB catalog stayed protected. With the Deny Bucket Policy and lifecycle rules in place, Mewo cut cost while removing the risk of anyone permanently deleting a needed file.

 

Start Your Journey with Automat-it

 

Achieve ultimate cost efficiency without slowing your product development or product impact. Start your journey with Automat-it and join the ranks of high-growth startups like Mewo.

Get in touch today

Get 40 FREE Hours From a Senior AI Engineer
We build one capability in your AWS account. You ship AI faster.