AWS
Case Studies
FinOps
How Unicorne Accelerated the Go-to-Market of Its New FinOps Tool and Increased Its Well-Architected Framework Review Delivery Capacity by 60%
TL;DR
Unicorne, the Québec-based 2025 AWS Canada Rising Star Consulting Partner of the Year, used CloudQuery to automate the AWS Well-Architected Framework Review Cost Optimization and Security pillar and cut audit delivery time from 50 hours to 20 hours per engagement. The team then built Stable, their AWS FinOps SaaS product, on the same CloudQuery data infrastructure, skipping months of data layer engineering. CloudQuery runs nightly across all of Unicorne's customer accounts, syncing AWS inventory into multiple AWS services, such as Amazon S3 and Aurora PostgreSQL.
Cloud Data Is Everywhere Except Where You Need It #
Unicorne is an AWS Advanced Tier Services Partner that runs Well-Architected Framework Reviews across several dozens of client environments every year. The company won "2025 AWS Canada Rising Star Consulting Partner of the Year" at AWS re
It's a problem they share with most of the industry. 84% of organizations say managing cloud spend is their top challenge (Flexera 2025), and with global cloud spending projected to hit $723 billion in 2025 (Gartner), the scale keeps growing. The data exists; it's just buried across dozens of AWS dashboards, Cost Usage Reports, and account silos. You can see the line items, but connecting that spend to specific resources and specific decisions is a different story.
"Cloud environments get more complex every year. More accounts, more services, more teams spinning up resources. The gap between what you think your infrastructure looks like and what it actually looks like keeps growing." Benoit Maheux, Partner and Vice President Growth, Unicorne
Multi-account environments make this worse. Most organizations Unicorne works with run 10 to 15 sub-accounts for production, staging, R&D, and so on. Some run 50. Each account needs its own audit trail, its own cost analysis, its own security review. As Benoit puts it: "It's not like we're finding information which is complex. It's just that it's scattered."
What makes Unicorne different from a typical AWS consultancy is its background. Unicorne started as a software development company building cloud-native applications, so the team understands the challenges that tech companies and ISVs face on AWS from the inside. Not all team members are auditing infrastructure from the outside; they've built and operated it themselves.
How Did Unicorne Automate AWS Well-Architected Reviews? #
An AWS Well-Architected Framework Review (WAFR) evaluates cloud infrastructure across six pillars: Security, Cost Optimization, Reliability, Operational Excellence, Performance Efficiency, and Sustainability. It's one of the most common consulting engagements in the AWS ecosystem, and it's also one of the most labor-intensive.
This isn't work that only consultancies care about. Any company running production workloads across multiple AWS accounts needs to audit security configurations, validate compliance posture, and track cost efficiency on an ongoing basis. Most internal platform and cloud ops teams are already doing some version of this work, checking CIS Benchmarks, reviewing IAM policies, and reconciling cost allocation across accounts. The difference is that Unicorne, to deliver the service to several dozen clients simultaneously, has developed proprietary tools that automate all of these checks, enabling it to do so much faster, more efficiently, and at scale.
Before CloudQuery, each review started with hours of error-prone manual data gathering. Consultants logged into client consoles, ran individual CLI commands, pulled data from Cost Usage Reports, checked CIS Benchmark controls by hand, and stitched everything together account by account, region by region.
"It's a lot of manual work and scripting to be able to extract all the data and analyze it." Eric Pinet, Co-Founder and CEO, Unicorne
The result was a 50+ hour engagement per client, with a significant portion of that time spent on data collection rather than analysis. And that was the bottleneck, in addition to the complexity of analyzing the results in order to issue the best improvement recommendations.
What Changed with CloudQuery? #
CloudQuery now handles data extraction across client AWS environments using cross-account IAM roles. For two of the six WAFR pillars (Security and Cost Optimization), the process is fully automated. CIS Benchmark coverage that previously took hours per account runs in minutes. Cost Usage Report data that required manual consolidation is now centralized.
"We can automate the process for at least two pillars: security and cost optimization. It's fully automated with CloudQuery. We spent 50 hours beforehand. Now it's 20 hours." Eric Pinet, CTO, Unicorne
The 30 hours saved aren't wasted. Unicorne's consultants now spend that time going deeper into the remaining four pillars (Reliability, Operational Excellence, Performance Efficiency, and Sustainability), where human analysis still adds the most value. The reviews are more thorough, not less.
Client AWS Accounts (10-50 sub-accounts)
|
v
CloudQuery (cross-account sync)
|
v
Amazon S3 (structured data)
|
v
Amazon Athena (SQL queries)
|
v
WAFR Reports and Recommendations
Unicorne now runs multiple WAFRs per week. The automation doubled their revenue from consulting engagements. This was achieved not by charging more, but by executing more engagements with the same team.
Before and After: WAFR Delivery
How Did Unicorne Build a FinOps Product on CloudQuery? #
Managing AWS cloud infrastructure and running WAFRs every week gave Unicorne a front-row seat to the same problem across dozens of clients: AWS cost data is opaque at the resource level. AWS Cost Explorer shows you're spending $20,000 on S3. It doesn't tell you which bucket, whether the costs are driven by storage or transfer, or whether your backup frequency is the real culprit.
"If you can't see it, then you can't make those adjustments on it at all." Benoit Maheux, Partner and Vice President Growth, Unicorne
That pattern repeated enough times for Unicorne to decide to productize it. The internal tooling they'd built for consulting engagements became the foundation of Stable, their FinOps SaaS product.
Why Not Build the Data Layer from Scratch? #
Unicorne explored building its own data collection layer before adopting CloudQuery.
"We explored building our own internal tool to do the same thing as CloudQuery. But it was more complicated, expensive, and it wasn't going to help us complete our MVP quickly." Eric Pinet, Co-Founder and CEO, Unicorne
"It's not the core of our solution. So why not use something that already exists and that has been proven?" Benoit Maheux, Partner and Vice President Growth, Unicorne
The team estimated that building a comparable layer from scratch would have consumed several months of engineering time. CloudQuery lets them skip building the product data layer. We were able to focus immediately on value creation with the recommendation engine, cost breakdowns, and alerting, rather than maintaining API integrations and data ingestion.
The FinOps Foundation's 2026 State of FinOps report found that only 14.2% of organizations have reached mature FinOps capabilities. There's a gap between knowing costs matter and having the tooling to act on them. Unicorne filled that gap.
How Does Stable's Architecture Work? #
CloudQuery runs as a containerized job nightly, syncing resource inventories from all connected customer AWS accounts. The data goes to two destinations simultaneously:
- Amazon S3 for historical analysis, batch processing, and recommendations engine
- Aurora PostgreSQL for low-latency queries powering the Stable UI
"For some queries, it's more efficient to have the relational database, PostgreSQL, for better latency in the application. And for S3, it's for all data that can be analyzed once a day." Eric Pinet, Co-Founder and CEO, Unicorne
Customer AWS Accounts
|
v
CloudQuery (nightly sync, containerized)
|
+----+-----------------+
| |
v v
Amazon S3 Aurora PostgreSQL
(historical/bulk) (live app queries)
| |
+----------+-----------+
|
v
Stable Application Layer
(Recommendations, Alerts,
Resource Explorer UI)
Stable's primary users are DevOps engineers, FinOps practitioners, and CTOs at small to mid-size businesses and ISVs. Unicorne charges a percentage of the AWS bill, which makes the ROI transparent for both sides.
"We're comparing before and after, and we keep track of what has been saved. So it's easy to know and show the value we add by what we're implementing in savings." Benoit Maheux, Partner and Vice President Growth, Unicorne
The results across Stable's client base are concrete: one client saw a 67% reduction in cloud costs, another achieved 30% infrastructure cost savings, and a third cut AWS spend by 23%.
What Has CloudQuery Meant for Unicorne's Business? #
Across both tracks (consulting and product), CloudQuery became a multiplier for a team that was already technically strong but limited by the manual overhead of data collection.
On the consulting side, cutting audit time by 60% didn't mean doing less work. It meant doing more engagements with the same headcount, going deeper on the analysis that clients actually pay for, and scaling a practice that previously couldn't grow without proportionally growing the team.
On the product side, saving several months of engineering on the data layer meant Stable got to market faster and the engineering team could focus on what differentiates the product: the recommendations engine, the resource-level cost breakdowns, and the alerting that turns data into action.
When asked to describe CloudQuery in one sentence, Eric said:
"Proven, reliable, and easy to have access to your full cloud inventory." Eric Pinet, Co-Founder and CEO, Unicorne
Learn More About Unicorne #
Visit unicorne.cloud to learn more about their AWS consulting and managed services, or check out Stable to see the team's FinOps platform in action.
Learn More About CloudQuery #
Building a similar workflow for your organization? Schedule a demo with our team or check out the CloudQuery documentation to get started.
Want to talk through your specific use case? Contact us here or join the CloudQuery community to connect with other users.
