FinOps as a Service

Three ways to run FinOps. One only charges when it works.

Build the function in-house, buy a platform, or bring in an outside team. Koritsu is the outsourced option that bills only on savings verified against your bill. No upfront cost.

Get your Savings Opportunity Report Compare the models

Success fee only. The fee is two months of realised savings, verified against your bill. No upfront cost.

In one line

FinOps as a Service is the outsourced delivery of cloud financial operations, the visibility, optimisation and governance of cloud spend, to a specialist provider rather than building the function in-house. Koritsu delivers it on a success fee. No upfront cost. The fee is two months of realised savings, verified against your bill.

Build, buy, or bring in a team

Most teams weighing FinOps are choosing between three operating models. Each has an honest trade-off in cost and time to value.

Build

Hire in-house

A permanent FinOps hire gives you dedicated focus. It also takes months to recruit, carries a full salary from day one, and a single person rarely covers AWS, GCP and Azure equally well.

Buy

Buy a platform

A platform gives you dashboards quickly. Someone still has to read them, interpret them, and act. In most cases the licence cost lands whether or not the findings get actioned.

Bring in a team

Outsource to a team

An outside team brings the platform and the people together. On a success fee the trade-off is smaller, because you only pay when savings are real and the platform is already built, so visibility starts on day one.

Success-fee FinOps as a Service de-risks the decision. There is no upfront cost to weigh against an uncertain payback, and the platform is already built, so you are not funding a setup phase before the first finding.

Scale the people layer up or down, without a permanent hire

The platform stands on its own. The team scales from there, so you take on as much or as little hands-on support as the moment needs.

Monitor

Platform only

Continuous visibility and anomaly alerting through Kori. Resource-level cost data across every account, with alerts normalised and routed to Slack or email. A good fit when you want eyes on the spend but the team to act sits with you.

Advisor

Platform plus reactive support

Everything in Monitor, plus FinOps support when you need a second opinion. Bring us a finding, a commit decision, or an anomaly, and you get an answer grounded in your own billing data rather than a generic benchmark.

Embedded

A proactive, fractional team

A proactive FinOps team working your environment on a regular cadence. Outsourced FinOps without a permanent hire, surfacing opportunities and sequencing the work before the question reaches the board.

Premium

Fully bundled

Platform, agent and team working as one function. For teams that want FinOps owned end to end, with standardised reporting in English and Brazilian Portuguese and a quarterly executive review.

Engineering-grade, by default

Outsourcing the function does not mean settling for surface-level findings. The same depth runs underneath every tier.

Resource-level resolution

Findings work from exported billing data, so they name the individual resource driving the spend, not just the service category.

Findings name the cause

We point to the code path, service behaviour or design decision behind the cost. A missing cache. An inefficient timeout. Idle capacity behind a steady bill.

Sequenced, then verified

Rightsize before you commit, so you rarely lock in spend you could have removed. Savings are measured against the bill, not projected on a slide.

In practice

There are cases where a single function drives most of a service's bill. With a global financial services group running on AWS, resource-level analysis of the exported billing data surfaced one Lambda function behind a disproportionate share of the Lambda spend. It was holding a high memory allocation while waiting up to 14 minutes on a third-party API, an inefficient shape for a workload billed on memory multiplied by time.

Rightsizing the memory took that function down 66% overnight, with no code change. Reworking it into an event-driven flow took the workload down more than 96% against the original baseline, verified against billing.

Read the full case study

FinOps as a Service, common questions

FinOps as a Service is the outsourced delivery of cloud financial operations to a specialist provider, rather than building the function in-house. It covers visibility into where the spend goes, the optimisation work to bring it down, and the governance to keep it down. Koritsu delivers it on a success fee. No upfront cost. The fee is two months of realised savings, verified against your bill.

A tool surfaces the data. It rarely acts on it. Many teams already have a dashboard nobody has time to read, and that dashboard carries its own licence, hosting and maintenance cost. FinOps as a Service pairs the platform with the people who interpret it, so a finding names the specific cause and the work to realise the saving gets done. If you already run a tool, we typically work alongside it rather than replace it.

It depends on your spend and how constant the focus needs to be. A permanent hire makes sense once cloud spend is large enough to justify a full salary and the work is continuous. Below that, in most cases an outside team is the more proportionate option. You get multi-cloud coverage and a platform that is already built, without recruiting for a role that is hard to fill. There are cases where a hybrid works well, with us embedded alongside an internal owner.

There is no upfront fee. Koritsu works on a success fee only. We deliver a Savings Opportunity Report and the work to realise the savings, and the fee is two months of realised savings, measured against your actual bill. The Savings Opportunity Report is the core deliverable. Ongoing support converts into a tiered monthly subscription, which you can scale up or down.

Visibility typically starts in the first week. We arrive with the platform already built, so there is no two-month dashboard project before the work begins. Resource-level analysis starts on day one, and the first findings usually follow shortly after.

Read-only access, scoped by an explicit allowlist your security team can review before anything is connected. We rarely need more than the billing and metrics data. Legacy accounts and any accounts mid-migration are excluded from the analysis, and from the savings figures.

AWS, GCP and Azure, and any provider that exposes FOCUS billing data. The view is unified across all three, so a multi-cloud estate reads as one picture rather than three separate ones.

Decide which way to run FinOps.

Whether you want a hands-on engagement with a fractional team, or just Kori watching the account and flagging anomalies, it starts the same way. A Savings Opportunity Report on your environment. No upfront cost, billed only on what shows up on the bill.

Get your Savings Opportunity Report