Executive Summary
- Engineering-led new business, Adaptive MDC launched recently, so Chief Editor, Lauren Raybould, sits down with Tom Babbington, Founder & CEO, Adaptive MDC, to talk about his journey and Adaptive MDC’s vision.
- If you are tied to one or two suppliers, you are limited by their manufacturing capacity and their lead times. By being agnostic, Adaptive MDC can look across the market for products with equivalent scope and capability that can be delivered in the timeline the customer needs.
- The most urgent demand is concentrated in Southern and Western Europe. Spain, Portugal, France, and Italy are all going through a period of growth in data centre development, driven by energy prices and PPA availability attracting customers to those markets.
1. In your 15 years in traditional data centre construction before founding Adaptive MDC, was there a lightbulb moment which made you think: there has got to be a better way to do this?
There was no one moment. It was more of a gradual build-up of frustration. The accumulation of watching the same failure mode repeat itself across project after project, regardless of the client, the contractor, or the geography. The delivery model was the problem: fragmented accountability, sequential processes, too many parties with partial responsibility, and no single entity that owned the outcome.
Over a long career, you see enough programmes overrun. You see enough clients absorb costs that could have been avoided. You see enough situations where everyone had a defensible position, and nobody actually owned the result. At some point you stop accepting that as inevitable and start asking what a different structure would look like.
If I am being honest, I cannot say there was a specific project or a specific moment. But there was an inflexion point. Rack densities were getting bigger. Technology was becoming more efficient and took up less space, which tied into being able to put it into smaller, factory-built enclosures. I remember the biggest data centre in Europe in 2006 was six megawatts. That same building was converted to 20 megawatts 15 years later, using much of the same footprint but with far more efficient technology. That trajectory made the case for a different delivery model.
What I can say is that all those accumulated frustrations led up to a point where I thought: if it can be done differently, why are we not doing it differently? Adaptive was founded to build the model that answers that question. The idea is to run manufacturing in a controlled environment, on a design based on a framework we can revise to suit each customer’s specification, so that the delivery and the site preparation happen concurrently rather than sequentially.
2. What is the biggest bottleneck traditional builds face right now, and how does a modular approach fit?
Schedules are always being compressed. Every data centre operator or developer needs to generate revenue from their platform, and the mechanism for doing that is the lease. To get to a lease, they need to have speculatively acquired the power and the land, developed a design concept that is attractive enough to secure a customer but not so specific that it locks them to a single tenant, and then once a customer signs, that becomes the fixed date. Everything from that point has to be delivered against a committed timeline. And that timeline is always getting shorter.
The result is that the traditional build model, where design, procurement, manufacturing, and site preparation run sequentially, is fundamentally misaligned with what the market now requires. Every handover point between those phases adds months. There is no technical reason why those workstreams cannot run concurrently. The reason they do not is the delivery model, not the technology.
Modular addresses this structurally. By running manufacturing in a controlled environment against a design framework that we adapt to each customer’s specification, we can compress the programme significantly. Site preparation runs in parallel with factory build. The result is deployment 20 to 50 percent faster than conventional construction, with projects achievable in as little as 12 months depending on scope.
There are also external constraints that the industry needs to address collectively. Power procurement delays, land-to-live timeline pressure, long-lead mechanical and electrical equipment, and jurisdictional permitting complexity all sit outside any single company’s control. For operators expanding live facilities, there is the additional challenge of adding capacity to infrastructure that cannot be taken offline. These are real bottlenecks, and they are not going away. But the delivery model constraint is the one that is most directly within our power to fix, and it is the one Adaptive was built to address.
3. A core pillar of Adaptive MDC is being vendor-agnostic. Why is it important to avoid being tied to specific equipment suppliers, and how does that benefit a client’s bottom line?
Being vendor-agnostic is about capacity as much as it is about principle. If you are tied to one or two suppliers, you are limited by their manufacturing capacity and their lead times. By being agnostic, we can look across the market for products with equivalent scope and capability that can be delivered in the timeline the customer needs. In an environment where long-lead equipment is already under pressure, that flexibility is not a nice-to-have.
A scoping document is, at its core, a series of preferences: a certain PUE, a certain uptime, a certain power density. Those outcomes can be achieved through clever design and a range of equipment choices. The established vendors have well-earned market share and produce excellent equipment. But no single product is the right choice for every customer and every use case, and the large vendors cannot build everything for everybody. There are alternatives in the market that are frequently underestimated.
For the customer, vendor-agnosticism means the design is driven by their operational requirements rather than by our supply chain relationships. We have no margin incentive to push a particular product. That translates directly to a specification optimised for their project, and to procurement flexibility precisely when it is most needed.
4. Time is money in the sector right now, quite literally. How damaging could a six-month construction delay be for a company?
Any delay is financially and reputationally damaging, and the damage is not contained to the party who caused it. Every data centre owner and operator has a customer, and every construction company has a customer whose customer has a customer. We are all in a hierarchy. If the top gets disappointed with the middle and the middle with the bottom, the next job does not come.
The most direct financial consequence is liquidated damages. A customer will inevitably have LDs or upstream liquidated damages with their off-taker, calculated as a proportion of the lease per day. The scale of those damages depends on the size of the programme and the terms of the agreement, but they can be substantial. Beyond LDs, there is the cost of capital being serviced on infrastructure that is not yet generating revenue, and in some cases there are break options in offtake agreements tied to delivery milestones that create even more exposure.
The ratio also matters. Six months on a 12-month programme is a different order of problem from six months on a 36-month programme. In the current market, where 12-to-18-month timelines driven by signed offtake commitments are increasingly common, the tolerance for delay has reached essentially zero. That is the commercial reality Adaptive’s delivery model is built to address.
5. You’ve launched Adaptive MDC to serve Europe, the Middle East and Africa. Over the next 18 months, where are you seeing the most urgent demand? What’s your ultimate vision for the company’s footprint?
In terms of where we are seeing the most urgent inbound demand right now, it is concentrated in Southern and Western Europe. Spain, Portugal, France, and Italy are all going through a period of growth in data centre development, driven by energy prices and PPA availability attracting customers to those markets. The requests we are getting are coming from those geographies. Ireland has slowed. Germany has also been quieter than expected and seems to have slowed a bit. The Nordics remain active but the tier-one markets like London and Frankfurt are constrained by cost of power, land, and the legislative environment and community backlash.
There is also a growing amount of activity in Eastern Europe, and the Middle East continues to be a strong market. The driver across all of these is the same: wherever energy is available at a competitive price and at scale, data centre development follows.
On the vision for our footprint, there is no reason why we could not deliver at scale globally. Our model is designed to do exactly that: to deliver quickly in multiple locations simultaneously, by running manufacturing in a controlled environment on a design framework that we adapt to each customer’s specification and each site’s requirements. The ambition is to reach a point where we can tell customers what capacity we have and when, rather than waiting for them to come to us. A manufacturing call-off model for data centre capacity.
