From CSRD to undeniably more sustainable IT

I recently spoke with a large manufacturing company about CSRD and sustainability. When I asked how much energy their IT consumed and what the resulting emissions were, there was a brief silence. Not because sustainability wasn’t a priority, but because this data wasn’t immediately available. They had to check whether it was stored somewhere within the organization.

Data on which IT assets were causing the consumption, when peaks occurred, and where exactly there was room for improvement was certainly not available. And that is precisely where the limitation lies. CSRD provides an organization with a general picture of sustainability, but does not automatically offer the detailed information needed to make IT more sustainable in a targeted manner.

The CSRD is intended to provide insight into the sustainability of companies. In the section on greenhouse gas emissions, large parts are still based on estimates and sector averages. This is valuable for reporting and comparison. It helps form an overall picture. But anyone who truly wants to become more sustainable year after year needs more than just an average. It must be clear where emissions originate, which systems contribute to them, and what usage patterns are associated with them.

For IT, this is particularly nuanced. Part of the emissions falls under Scope 2: specific indirect emissions, for example, when an organization powers its own equipment. Another part falls under Scope 3: other indirect emissions. Think of cloud usage in all forms and IT resources that the organization does not power itself, such as data plans for mobile devices.

In this blog, I’ll show how hard measurement data helps to tangibly reduce IT emissions. And how you can also demonstrate that you’re becoming more sustainable when direct measurement data on emissions is lacking, for example with cloud usage.

It starts with insight

The fastest way to make IT more sustainable is to start with the largest sources of emissions. Don’t make small improvements everywhere at once; instead, focus on where the greatest gains can be made. Usually, these are the five or ten heaviest consumers in your IT landscape. But you need to know where they are.

You won’t get that insight from averages. For that, you need hard measurement data.

By measuring consumption across all hardware and software layers, you can see where the emissions are coming from. Which applications require the most capacity? Which workloads cause spikes? Which users, transactions, or servers have the greatest impact on your energy consumption?

If you store those measurements over a longer period, something even more valuable emerges: pattern recognition. You can see when emissions are generated, which load is structural, and which capacity is only needed at certain times. That makes it possible to consolidate workloads more intelligently onto fewer servers. Less server overhead directly translates to lower consumption.

It’s not just about less hardware. The load matters too. A physical server that is under very light or, conversely, extremely heavy load operates less efficiently than a server that is utilized optimally.

With insight into the most demanding applications, workloads, transactions, and users, you know where optimization yields the greatest returns. This way, sustainability doesn’t become a broad improvement program, but a targeted investment with measurable returns.

How more efficient software makes your IT more sustainable the fastest

My navigation system often suggests a route that takes a few minutes longer but saves 10% or 15% on fuel. I almost always take that one. It’s the easiest way to reduce my car’s emissions, even when driving on electric power.

For IT assets, that principle works in a strikingly similar way. The fastest route to sustainability is often not buying more capacity, installing more energy-efficient hardware, or moving everything to a different environment. The fastest route is to reduce demand.

Much of the software used today requires more capacity than is necessary. We’re not talking about small margins of 10% or 20%. In practice, halving consumption often proves achievable. Improvements by a factor of ten, a hundred, or more are regularly seen among clients. Not because dramatic mistakes are being made everywhere, but because software, databases, configurations, and infrastructure grow, change, and become more resource-intensive over the years than is strictly necessary.

You can read about how that inefficiency arose in this blog. The good news: there’s a lot to be gained. And often, those gains are closer than you think.

Once it’s clear where most of the overhead comes from, you also know where optimization will have the greatest impact. This applies to your own software, but also to software from vendors. Databases can be cleaned up and accelerated. Configurations can be made smarter. Workloads can be lightened. Many vendors genuinely want to improve their product, especially when you use concrete metrics to show where the load is coming from.

Optimization therefore starts again with insight. Why is a workload heavy? Where are the bottlenecks? Which adjustment reduces demand without affecting productivity? With those answers, sustainability becomes practical, targeted, and measurable.

In the cloud, too, sustainability starts with optimization

We helped a customer moving to the cloud cut the number of required CPUs in half. That made a significant difference in emissions. And saved €200,000 per year.

That makes cloud optimization one of the most practical paths to sustainability. Even when the exact sustainability data from your cloud provider is limited.

Cloud providers aren’t always generous with detailed sustainability information. That doesn’t have to be an obstacle to demonstrably making your cloud workloads more sustainable. For the coming years, it’s sufficient in CSRD reporting to calculate cloud emissions based on global figures and industry standards. If you use fewer resources in the cloud, that calculation will naturally be lower.

This makes optimization immediately relevant. Fewer CPUs, fewer operating hours, less overcapacity, and fewer unnecessary resources result in lower calculated emissions. And usually a lower bill as well.

In addition to optimization, consolidation is important. Servers that are only active part of the day can often be grouped together more efficiently. Not everything needs to be on continuously or run separately. Any resource that isn’t needed doesn’t have to emit CO2 or cost money.

However, cloud consolidation requires a clear overview. You want to know which workloads are running when, what dependencies exist, and where capacity can be safely reduced. So even in the cloud, sustainability starts with insight.

Make IT sustainability concrete

Sciante quickly helps you identify where the greatest gains lie. Not with averages or assumptions, but with concrete insight into consumption, load, workloads, and optimization opportunities.

That yields more than just a greener IT landscape. Less hardware. Fewer licenses. Lower cloud costs. And more control over the choices you make.

Want to discover where your IT can run lighter, more efficiently, and more sustainably? Schedule a no-obligation consultation. Then we’ll work together to determine the best place to start.