Hyperscalers stop talking about capex and start talking about time-to-energy
In Q2 2026 earnings calls, Microsoft, Alphabet and Meta shifted emphasis from aggregate capital expenditure to time-to-energy, large-scale networking, power procurement, and how fast infrastructure converts into revenue-generating compute. The metric change reveals what is actually scarce.
Capex was the right headline while chips were the constraint — spending mapped to capacity. It has stopped mapping. A company can commit tens of billions and wait years for interconnection, so the number that predicts revenue is not what was spent but how quickly it becomes powered, networked and billable.
Time-to-energy is an unusually honest metric because it is largely outside the reporting company's control. It depends on utilities, regulators and local approvals, which means executives are now guiding investors on a variable they can influence but not determine — a notable admission from firms accustomed to controlling their own supply chains.
For investors the practical consequence is that capex comparisons between hyperscalers have lost much of their meaning. Two companies spending identically can be years apart in delivered capacity depending on where they built and how early they secured power.
BloombergNEF — AI data center build advances at full speed → · Data Center Knowledge — New data center developments: August 2026 →