A market that attracts, and misleads
The figures are well known and spectacular. France now has more than 300 data centres, and the announced investment trajectory exceeds €100 billion over the coming years, spread across hyperscaler projects, colocation operators and sovereign players. The French market has become one of Europe's three hubs, alongside Frankfurt and Amsterdam. Projects keep coming: Marcoussis, Dugny, Noyal-sur-Vilaine, Marseille Phase 2 and beyond, Toulouse, Lyon.
For a construction manufacturer looking at these figures, the logic seems unstoppable. A high-value market, growing exponentially, consuming en masse what we produce: building envelope, insulation, waterproofing, air conditioning, high- and low-voltage systems, steel structures, cabling, technical equipment. So let's go for it.
And this is where the misunderstanding begins. Because this market, while a genuine construction market, follows almost none of the construction industry's rules as you know them.
Why your usual channels don't work
First surprise for a construction brand entering the market: the traditional distribution channel plays no role. Contractors don't buy from a general builders' merchant. Specifiers don't read "Le Moniteur". Decision-makers aren't at BATIMAT. Site visits take place under NDA. Plans are not public.
Second surprise: specification is locked down by a very small number of players. A handful of ultra-specialised engineering firms (ARTELIA, SETEC, EGIS, SPIE Batignolles Génie Civil, the hyperscalers' in-house teams) hold the technical specifications. These teams number in the dozens, not the thousands. They don't browse product catalogues: they write specifications based on feedback and in-house testing, and products not already on their technically validated shortlist don't make it into the tender.
Third surprise: the qualification cycle is long. Between the first approach to an engineering firm and the first reference installed in a significant data centre, expect 18 to 36 months. Tests, test reports, feedback from a first pilot site, cross-validations. This is a commercial investment with a three-year horizon, not a quarterly prospecting campaign.
The entry filter
Before even talking about communication, a construction brand has to answer one question: does my product, solution or service have a reason to be specified in a data centre, beyond the simple fact that there is a building envelope to construct? If the answer is no, this market is not for you. If it is yes, the work begins.
Three types of construction brand facing the data centre market
Our client conversations over the past twelve months reveal three very distinct profiles. Each calls for a radically different commercial stance and communication strategy.
Type 1: brands already present
They already sell significant volumes into data centres, often without having planned it. They were brought in by an engineering firm or a specialist installer who knew their standard range and introduced it on a project. In this category: certain brands of precision air conditioning, UPS systems, high-performance cabling, technical waterproofing, raised access floors, modular partitions, access control…
Their challenge is not to get in, it is to capitalise and scale. Turning an incidental presence into a deliberate positioning. Documenting their references (project by project, within the limits of NDAs), building a dedicated offer recognised as such, assigning sales staff to this segment and, above all, speaking out about their own cases.
Type 2: legitimate brands that haven't yet got through the door
They have a technically relevant solution for the market but no installed reference yet. Evaporative cooling, free cooling, high-performance thermal insulation, technical fire protection, heavy-duty flooring, seismic-resistant structures, air management, clean-room acoustics.
Their challenge: getting through the engineering firms' filter and securing a first pilot reference. This is infiltration work, not mass communication. It requires a technical offer repackaged for the market (dedicated data sheet, performance expressed in the sector's units, specific test reports), a presence at specialist specification events (DCD Paris, Datacloud Global Congress Monaco, France Datacenter's technical days), and direct commercial investment in 20 to 30 key contacts. This is where ABM (Account Based Marketing) comes into play!
Type 3: brands eyeing the market out of imitation
They produce standard building solutions with no differentiating technical feature for the data centre segment: ordinary acoustic insulation, standard tile adhesive, tiling, generic aluminium joinery, non-specialised low-voltage systems.
For them, the data centre market is not a growth driver. It is a decoy. Volumes flow through general contractor packages or go directly to specialist manufacturers. Investing in communication on this segment would be money wasted, in both budget and sales time.
Communicating in this market, in practice
For Type 1 and Type 2 brands, the question is no longer "should we go?" but "how?". And here, almost all the classic construction communication reflexes are ineffective. Here is what we see working.
Produce technical proof, not campaigns
Data centre specifiers don't buy a message. They buy high-quality technical documentation that is readable, structured and quantified. Data sheets with performance expressed in their units (kW/rack, PUE, uptime, tier). Independent test reports. Named references where NDAs allow, or site typologies where they don't ("3 hyperscale Tier III colocations in the Paris region since 2024" is sometimes worth more than a name). EPDs and integrated carbon footprint, increasingly requested by European hyperscalers bound by their climate commitments.
Be at the events that matter, not the ones that are visible
DCD Connect Paris, DCD Energy & Sustainability, Datacloud Global Congress in Monaco, France Datacenter's events. These are the gatherings that drive specification in the European market. BATIMAT plays no role in the data centre segment. Even a minimal presence at these specialist events pays off more than the finest generalist stands.
Produce long-form technical content
White papers, detailed case studies, articles signed by your engineers. Data centres are a subject where long technical content is valued. Specifiers will read 30 pages if those pages give them a method, a figure or a new approach. This is the construction segment where in-depth content pays off fastest, provided you genuinely have something to say technically.
Invest in GEO and AI-driven specification
Our recent article on GEO said it: specifiers are starting to query AI tools ahead of their choices. In a market this technical and this well documented by third-party English-language sources (international trade press, sector studies, hyperscalers' white papers), presence in AI answers is built differently. A French brand that doesn't exist in English in this segment is invisible. Producing bilingual technical content is becoming a prerequisite.
Build an expert voice
In this market, credibility is built by identifiable people, not by institutional discourse. Your application engineers, technical directors and data centre market managers need to speak out, publish, speak at conferences and respond to press requests. Authority transfers from the person to the brand, not the other way round. This is technical social selling and personal branding work, not a corporate campaign.
A market apart, which rewards patient brands
The data centre is not an El Dorado in the sense the term suggests: a territory where simply arriving is enough to prosper. It is a market apart, with its own rules, which rewards brands capable of rethinking their approach from top to bottom and sustaining that approach over time.
The next two or three years will decide positions for the fifteen that follow. Brands that invest in data centre specification now, seriously and methodically, will take positions that are hard to catch up with. Brands that wait and see, or copy classic construction codes onto this segment, will lose their communication budget.
The only honest question to ask today is not "do we go or not?" but: do my product, my brand and my organisation have a genuine reason to be there? If yes, the method exists. If not, there are other markets.
At Convergence, we now support several clients on this positioning. If the subject is on the agenda of your next marketing meetings, we would be delighted to discuss it with you. No brief, no pressure. Just to shed light on the decision.
