Startups face increasing pressure to achieve rapid growth, which often requires winning contracts with large enterprises that possess massive budgets in the IT world. However, securing these contracts is not as easy as many startups assume.
Speaking at the TechCrunch Disrupt conference, Arsalan Tavakoli-Shiraji, co-founder and Senior Vice President of Field Engineering at Databricks, revealed a fundamental mistake startups make: the belief that superior technology is enough to sell a product automatically. While technical quality is a prerequisite, Tavakoli-Shiraji emphasized that "the human element and organizational processes are the most important."
The "Umbrella" Strategy: The Power of Partnerships with Giants
Tavakoli-Shiraji reviewed the experience of Databricks (now valued at $190 billion and serving over 70% of Fortune 500 companies) in its early days. The company faced difficulties reaching large clients like the Swiss bank UBS initially, but the turning point was a partnership with Microsoft to become a partner for the Azure cloud platform.
This partnership provided Databricks with two vital advantages:
- Legal Coverage: Sales agreements were handled under the umbrella of Microsoft's terms, which simplified the procedures.
- Distribution Channels and Access: The Microsoft sales team booked meetings and advocated for Databricks products, opening doors that were previously closed to the startup.
Today, Databricks plays the role of an "umbrella" for other startups like Hightouch, Addepar, and Tecton, selling their products under its own terms and security standards.
Addressing AI Concerns: The "4 Cs"
When selling AI solutions to large enterprises, Tavakoli-Shiraji identified four key challenges that startups must address, which he calls "The Four Cs":
- Context: The AI's ability to learn the company's terminology, processes, and priorities, just as a new employee would.
- Choice: Giving enterprises the freedom to navigate between different AI models rather than being locked into a single provider.
- Cost: Providing mechanisms to control the costs of running AI products.
- Control: Establishing safeguards that govern AI behavior and protect sensitive information.
Modernizing Legacy Systems as a Golden Opportunity
Tavakoli-Shiraji pointed to a significant gap in AI adoption within companies; while companies aspire to reach advanced levels of automation, the vast majority (about 99%) are still at primitive levels limited to using tools like ChatGPT for coding only.
He believes the real opportunity for startups lies in modernizing legacy systems that are expensive and complex, such as IBM AS/400 machines and mainframes that companies are afraid to touch due to the complexity of their code. Today's AI agents can understand and deconstruct this old code, opening the door for products that either add modern interfaces to these systems or replace them entirely.
Why does this news matter?
This analysis reveals that success in the enterprise tech sector does not depend solely on software innovation, but on the ability to integrate into existing ecosystems (like Azure) and understand the operational and legal obstacles faced by large companies. It also defines a clear path for AI-specialized startups: moving from merely providing assistance tools to offering radical solutions that address the problem of legacy systems and the challenges of control and cost.