Enterprise
AI infrastructure sales cycles compress to 30 days from six months
DataDirect Networks says hyperscaler and enterprise deals now involve multiple stakeholders and require rapid deployment capabilities.
Key takeaways
- AI infrastructure sales cycles have compressed from six months to as little as 30 days, requiring vendors to deploy working systems in weeks.
- Neocloud deals now involve multiple stakeholders—capacity off-takers, cloud operators, and integration layers—rather than single enterprise buyers.
- DDN customers on Google Cloud saw 70% better training performance and 40% GPU efficiency gains using Managed Lustre.
- DDN is growing threefold while prioritizing technical and implementation hires over salespeople to support rapid deployment timelines.
- Integrated offerings like Enterprise AI HyperPOD with Super Micro aim to reduce AI infrastructure setup time to three hours.
The procurement timeline for AI data infrastructure has collapsed from six months to as little as 30 days, fundamentally changing how storage vendors engage with hyperscalers and enterprises rushing to maximize GPU investments.
Kevin Delane, president and chief revenue officer at DataDirect Networks, told theCUBE Research at SC26 that the compressed timelines reflect AI's elevation of data infrastructure from back-office purchase to business-critical bottleneck. The shift has made performance the primary differentiator in a market where speed directly determines return on GPU capital.
Why it matters
When infrastructure sales cycles shrink by 80 percent while simultaneously adding stakeholders, vendors must fundamentally restructure how they sell and deploy. Companies that can't deliver working AI stacks in weeks rather than quarters will lose deals to competitors who've rebuilt their go-to-market around rapid deployment. This compression also signals that AI infrastructure has moved from strategic planning cycles into operational urgency.
Multiple buyers replace single enterprise decision-makers
The buying process itself has grown more complex even as it accelerates. Neocloud deals that once involved a single enterprise buyer now require coordination among the capacity off-taker, the cloud operator, and the integration software layer, according to Delane. "The sales cycle is more complex for sure. The speed is insane," he said.
DDN is responding by tightening ecosystem partnerships. The company's decade-long relationship with Nvidia anchors its strategy, while joint offerings with Super Micro Computer—including the Enterprise AI HyperPOD—aim to replace custom data center buildouts with integrated AI stacks. At SC26, DDN introduced software-defined EXAScaler deployment to give enterprise customers more hardware flexibility.
Delane said the integrated approach "takes the complexity out of it, allows an enterprise to set up an AI pod within 3 hours."
Performance gains justify the data layer focus
DDN's performance-first positioning appears to be resonating in both private and public cloud environments. On Google Cloud, customers using DDN's Managed Lustre service achieved 70 percent better training performance and a 40 percent improvement in overall GPU efficiency, Delane reported.
The company is growing threefold this year while adding only a small number of salespeople. Instead, DDN is hiring technical and implementation staff who can accelerate customer deployments. Delane said he looks for candidates with grit who fit a direct, results-driven culture: "If people can fit that culture, want to come in, want to grow this company to multiple billions, we have a place for them."
The details were first reported by SiliconANGLE.