22 August 2025

The hidden cost of every click: Why smarter data collection is better for privacy and the planet

In summary

• Every click, tag, and dataset has an unseen environmental cost in carbon emissions and water use.
• Smarter data practices, leaner tracking, shorter retention, and efficient infrastructure like CDNs and server-side tagging, reduce waste.
• Why it matters: Cleaner analytics align privacy, performance, and sustainability, cutting costs, building consumer/client trust, and protecting the planet.

Measure your digital carbon, not just your conversions

Book a flight and you’ll see its carbon footprint. Now imagine the same visibility for every AI query or tracking tag. Would we still be so quick to collect it all?

Every digital action has a cost. Even a simple Google search consumes energy, and with AI summaries now embedded in results, that cost is climbing.

For too long, our industry has measured success only in clicks and conversions. But it’s time we started thinking about measuring our digital carbon footprint. Smarter data collection doesn’t just improve privacy and performance, it lightens the load on the planet.

And behind every click, page load, and tracking tag sits a chain of servers, data centres, and pipelines, each with a tangible environmental impact. Just because it’s invisible doesn’t mean it doesn’t exist.

Digital is not “green” by default

I first came across this idea through Gerry McGovern, an author and consultant who’s been writing for years about digital waste. His message is simple but powerful: the internet is not weightless. Every piece of content we store, every tag we fire, and every dataset we hoard has a physical footprint, electricity to process it, servers to store it, and in the case of large data centres, large volumes of water resources are required for cooling.

Here are some sobering numbers to consider when it comes to digital waste:

Every spam email adds around 0.3 grams of CO₂. With more than 300 billion emails sent globally each day, that’s the equivalent emissions of 1.6 million cars on the road for a year.

An average Google search produces about 0.2 grams of CO₂ equivalent. By comparison, a single AI-powered search query can consume 10 times more electricity, making the carbon cost of “just asking the chatbot” much higher than most people realise.

The bottom line? We can take steps to reduce digital waste and reduce the impact we have, if we act more conscientiously when setting up our data collection infrastructure.

Data hoarding comes at a cost

In analytics, there’s a tendency to “collect everything” on the assumption that more data equals better insights, however in our experience often the opposite can be true.

I’d much rather have a clean, lean dataset that’s structured and relevant than a messy, chaotic and poorly structured dataset.. Less noise means faster insights, less processing power required, and, more importantly, a lower environmental impact.

How about your data retention strategy? Do you really need to keep raw event data for five years? Is anyone in your organisation looking at 3+ year old data on a regular basis? Probably not. The longer you retain it, the more carbon cost you’re adding. Smarter collection means setting boundaries: collect what matters, retain it only as long as it’s useful, and discard it after.

The risk of holding onto the wrong data

It’s not just the amount of data that matters, but the type. Personally identifiable information (PII) is especially risky to store.

Once you’ve extracted the insights you need, there’s little reason to keep raw identifiers.

Here is an analogy to think about, let’s say your analytics dataset contains every car registered on the road in the state of New South Wales. And let’s assume you want to build a dashboard detailing how many cars are registered on a yearly basis, along with attributes such as colour, make, model and year. Whilst you may initially require the underlying licence plate data to build out your dashboards, once you’ve built your aggregate data, retaining the underlying plate data (and potentially sensitive PII) is no longer required.

The same rule can apply to any other user data, aggregate what’s valuable and insightful, then delete the rest.

Holding on to PII indefinitely only increases the damage if a breach occurs. A recent 2022 example is the Optus data breach. Whilst poor security practice was part of the issue, another facet for Optus was the fact that some customers who had data exposed hadn’t been active Optus customers for 10 or more years.

As best practice it’s worth considering hashing or encrypting sensitive data if you must keep it for a short period for the purpose of deriving valuable insights,, but the safest option is to delete it once it’s no longer required.

By treating data as perishable rather than permanent, businesses can reduce both their environmental footprint and their exposure to security risks and as a result the possibility of negative PR.

The water we don’t see

Carbon isn’t the only issue. Recent reporting has highlighted just how much water data centres consume just to stay cool, tens of billions of litres annually in places like Victoria and the US. On a dry continent like Australia, and many other parts of the world where water scarcity and drought is an ongoing challenge, that’s a serious concern.

When you layer AI workloads on top, demand skyrockets, and Google’s own investment into nuclear power in the US is a signal of just how much energy they anticipate will be needed to keep up.

Smarter tracking helps everyone

Cleaner analytics isn’t just good for the environment. It also improves user experience and business outcomes:

• Fewer tags = faster pages. Lighter websites mean better SEO and conversion rates.
• Less data = lower risk. Privacy and security improve when you’re not holding onto sensitive user information for long periods.
• Efficient pipelines = lower cost. Processing smaller, cleaner datasets saves money in the cloud.

Even infrastructure choices, like using content delivery networks or server-side tagging, can reduce latency, improve experience, and potentially lower carbon impact by minimising unnecessary requests.

Louder’s recommendations

Whilst this won’t be a definitive list and no doubt it will likely evolve over time as AI and Machine Learning driven insights and the impact of these tools become better understood, here are some areas kick start your own carbon reduction journey:

• Audit your tracking stack – Remove duplicate or legacy tags that add no value.
• Prioritise what matters – Focus only on metrics that drive decisions and outcomes.
• Align privacy with sustainability – Minimising data collection reduces both carbon impact and privacy risk.
• Optimise your infrastructure – Use content delivery networks (CDNs) to keep data closer to users, apply load balancing to streamline traffic, and consider server-side tagging. A single tagging server managing requests is often more efficient, and potentially more carbon-friendly—than hundreds of browser-fired calls.
• Delete what you don’t need – Data has a lifespan. Retain it only as long as it’s useful.
• Make it a KPI – Measure your analytics footprint alongside clicks and conversions.

How Louder approaches digital sustainability

At Louder, we believe that if we’re asking the industry to take responsibility for digital waste, we need to do the same ourselves. That means building sustainability into the way we work, not treating it as an afterthought.

Some of the steps we take include:

• Minimising unnecessary tracking – we apply the same lean analytics principles internally that we recommend to clients, collecting only the data we need.
• Setting retention limits – our default is to retain data only as long as it’s useful, reducing storage requirements and the risks that come with stockpiling user information.
• Optimising infrastructure – we use content delivery networks and server-side tagging to reduce latency, cut data transfer, and improve efficiency.
• Device and infrastructure choices – we extend the lifespan of hardware wherever possible, recognising that up to 80% of a device’s CO₂ footprint comes from its manufacture.
• Embedding culture – we train our teams to be mindful of privacy, efficiency, and waste when working with data, encouraging “collect smarter, not more” thinking in every project.
• Since 2022 we have committed to planting 1,200 more trees every year via Rainforest Rescue
• We also recently published an article on media buying efficiency and how we can lower the carbon footprint of online advertising.
• Leaning on Big Tech to be more transparent when it comes to the carbon impact of advertising.

These aren’t one-off initiatives; they’re part of a shift in mindset. Just like we help clients rethink measurement for a world of privacy reform, we’re also rethinking how analytics and infrastructure decisions can reduce our collective environmental footprint.

Get in touch

Get in touch with Louder to discuss how we can assist you or your business and sign up to our newsletter to receive the latest industry updates straight in your inbox.