I’ve been thinking about the growing public opposition to new data centre developments.

And I keep coming back to a question.

If someone proposed building an enormous distribution warehouse on exactly the same piece of land, would local residents simply wave it through?

Of course they wouldn’t.

There would still be concerns about the scale of the building. The impact on the landscape. Construction. Infrastructure. And, fundamentally, whether a development of that size belongs in that location at all.

But put the words “data centre” on the planning application and something seems to change.

The debate gets louder.

Suddenly we’re not just discussing planning. We’re discussing AI, Big Tech, electricity consumption, water, carbon and whether the facility creates enough jobs to justify its existence.

Some of those are entirely legitimate questions.

But I do wonder whether we’re sometimes conflating two different debates.

Do people have a problem with data centres?

Or do people have a problem with very large industrial infrastructure being built in locations where they don’t believe very large industrial infrastructure should be built?

Because those aren’t quite the same thing.

Perhaps we should start with location

Data centres are now recognised by the UK Government as critical national infrastructure.

Whether we realise it or not, they underpin an extraordinary amount of everyday life. Banking, healthcare, communications, business applications, cloud services and, increasingly, AI all depend on the infrastructure sitting behind them.

So we’re going to need data centres.

The more interesting question for me is where should we build them?

Where technically and commercially practical, should we be putting greater emphasis on brownfield and former industrial locations?

Former power stations. Former manufacturing sites. Land which has previously supported significant industrial infrastructure.

That doesn’t magically solve the sustainability question.

In fact, the proposed redevelopment of the former Skelton Grange power station site in Leeds demonstrates that point rather neatly. Despite being a former industrial site, the proposed data centre development has still attracted opposition around energy and water consumption.

And perhaps that’s healthy.

Choosing an appropriate site shouldn’t remove the obligation to talk honestly about power, water, carbon and environmental impact.

But it might allow us to separate those legitimate questions from a much simpler one:

Why are we building something this enormous here?

There’s another challenge for our industry

I also think data centres have a communication problem.

A warehouse is easy to understand.

Goods arrive. People work there. Lorries leave. The connection between the building and its purpose is visible.

A data centre isn’t like that.

To most people, it’s a huge, highly secure building, consuming significant electricity, with relatively few people visibly coming and going.

Yet inside those buildings sits infrastructure on which huge parts of our economy and society increasingly depend.

Perhaps, as an industry, we haven’t been good enough at making that connection.

We’ve spent a lot of time talking about PUE, cooling technology, renewable energy and increasingly sophisticated engineering.

All important.

But those aren’t necessarily the first questions a community asks when presented with a huge proposed development.

Their first question may simply be:

Why here?

And I think that’s a fair question.

Maybe we need to change the conversation

Rather than starting the sustainability conversation with how efficiently can we operate this data centre?, perhaps we need to start one step earlier.

Have we chosen the most responsible place to build it in the first place?

I’m certainly not suggesting brownfield is the answer to everything. Data centres have very specific requirements around power, connectivity and infrastructure, and not every former industrial site will work.

Nor should legitimate concerns about energy and water simply be dismissed because a site happens to be brownfield.

But perhaps “brownfield first, where practical” is an approach worth discussing.

And this is something I find particularly interesting from our perspective at Asanti.

Our model is somewhat different from the hyperscale developments that tend to dominate the headlines. We operate a Regional Edge model, providing colocation capacity across a number of UK locations.

Perhaps part of the answer isn’t always to build bigger. Perhaps it’s also to think more intelligently about where the capacity already exists, where new capacity is genuinely needed, and how more distributed digital infrastructure can fit into the communities and regions it serves.

I’m not suggesting edge solves the big planning, energy or sustainability questions. It doesn’t.

But I do think it gives us another perspective on the assumption that meeting increasing demand for compute inevitably means ever larger campuses.

If we genuinely believe data centres are critical national infrastructure, our responsibility cannot stop at building more efficient facilities.

It should include choosing locations responsibly, making sensible use of existing infrastructure, explaining the value of what we’re building and being prepared to have an honest conversation with the communities that will live alongside it.

Because maybe the problem isn’t simply that we’re building data centres.

Maybe we need to think harder about what we build, where we build it, and whether bigger is always the only answer.

I’d be genuinely interested to hear other views, particularly from people outside the data centre industry.

If the same development were a distribution warehouse rather than a data centre, would you feel differently about it? And if so, why?