Startups chasing large enterprise customers often assume that strong technology will carry the sale on its own, but Databricks co-founder and senior vice president of field engineering Arsalan Tavakoli-Shiraji says that assumption is a costly mistake. Speaking at TechCrunch Disrupt, he said people and processes ultimately matter more than the product itself, a lesson he learned while helping Databricks land its first major deals. Databricks, now an AI and big data storage company valued at $190 billion with more than 70% of the Fortune 500 as customers, began as a small startup where Tavakoli-Shiraji was responsible for early enterprise wins.
He recalled one early prospect who turned out to be the same person who had signed a major deal with a Databricks competitor, leaving no interest in hearing the startup out because his career was already tied to the rival. Other early deals fell apart because the company was unprepared for legal, contractual, and security questions. Those experiences shaped his advice for younger companies trying to win large accounts.
Partnerships proved decisive for Databricks. After failing to secure a meeting with the Swiss bank UBS, the company became a Microsoft Azure cloud partner, which placed its sales agreements under Microsoft's terms and put Microsoft's sales staff in the position of booking meetings and advocating for the startup. Tavakoli-Shiraji described the access and distribution that came with the partnership as a major boost, and said Databricks soon found itself pulled in many directions, including prospects asking for features in countries where it had not yet expanded.
Databricks now plays the larger-company role for younger vendors, selling their products under its own terms and cybersecurity setup. Among them are Hightouch, which provides AI-powered marketing and customer-data tools and has also received investment from Databricks; Addepar, which offers data tools for investment professionals; and Tecton, a machine learning startup Databricks later acquired.
Tavakoli-Shiraji sorts enterprise concerns into four categories he calls the four Cs: context, choice, cost, and control. Context refers to an AI that can be trained like a new employee, learning a company's terminology, processes, and priorities. Choice is the ability to switch AI models rather than being locked to one provider, cost is about controlling how much an AI product costs to run, and control covers safeguards governing the AI's behavior and the protection of sensitive information.
He also pointed to legacy systems as a major opening for startups. He said most CIOs he speaks with are still in the early stages of AI adoption, with employees simply having access to ChatGPT rather than the company having changed its business processes. Banks still run IBM AS/400 systems and mainframes that nobody wanted to touch because the code was not understood, he said, but AI agents can now examine and break down what that code does, opening the door to products that add modern interfaces to those systems or replace them entirely.
More company and startup news from TechManNews.
