⚠ The Revenue Must Catch Up to the BacklogHigh threat

Beyond the concentration risk in the backlog lies an execution risk that the market presses on relentlessly: the revenue must actually catch up to the nearly $100 billion of commitments, and the gap between the two is enormous. A backlog of about $104 billion against annual revenue still in the single-digit billions implies a revenue ramp of extraordinary steepness must materialize — CoreWeave must build the vast, expensive capacity to fulfill the contracts, convert them into recognized revenue on schedule, and do so at margins that justify the effort. The central question analysts keep asking is precisely whether the revenue ramp can keep pace with a backlog that has grown far faster than the business itself.

The conversion is not automatic, and several things can slow or spoil it. Building the capacity requires enormous additional capital and flawless execution at unprecedented scale; delays in data-center construction, power, or chip delivery push revenue out; and margins can disappoint as competition and Nvidia's pricing press, or as customers renegotiate. A backlog that converts more slowly, more expensively, or less profitably than hoped turns the headline number from a promise of growth into a source of disappointment — and because the valuation capitalizes the backlog as though it will convert smoothly, any stumble in the ramp strikes directly at the thesis. CoreWeave's contracts are real and its execution so far has been aggressive and largely successful, so the ramp may well arrive. But an investor should recognize that the sheer speed at which the backlog has outrun the revenue makes the conversion harder, not easier, and that turning $104 billion of promises into profitable revenue — on time, at scale, at acceptable margins — is an unproven feat on which the entire valuation rests, and the one the market is right to watch above all else — converting $99.4B of backlog on ~$6B of trailing revenue¹.