By Titik Puspa 
In today’s hyper-competitive startup landscape, the imperative for rapid growth has never been more pronounced. To achieve this exponential expansion, a critical pathway for many emerging companies lies in securing the patronage of large enterprises – the financial powerhouses of the technology sector. However, as any seasoned business that has navigated the complex sales cycles of enterprise clients can attest, winning these lucrative contracts is far from a straightforward endeavor. A pervasive misconception among nascent startups, according to Arsalan Tavakoli-Shiraji, co-founder and senior vice president of field engineering at Databricks, a titan in big data and AI solutions, is the belief that their technological innovations will inherently sell themselves.
While the efficacy of the technology is undeniably foundational, Tavakoli-Shiraji emphatically warns that it is merely a prerequisite, not a guarantee, of success. Databricks, now a globally recognized giant in AI and big data storage, boasting a formidable valuation of $190 billion and serving over 70% of the Fortune 500, was once a fledgling startup. In those formative years, Tavakoli-Shiraji played a pivotal role in spearheading the acquisition of its inaugural significant enterprise deals. Reflecting on his experiences, he shared insights at TechCrunch Disrupt, a premier technology conference, noting that the prevailing sentiment among startups aiming for enterprise clients is often an overemphasis on the technical superiority of their offerings. "Everybody when they’re selling to enterprises at startups… we think it’s all about the tech and having the best tech," Tavakoli-Shiraji stated. However, his journey taught him a profound lesson: "people and processes matter the most."
He recounted a particularly illustrative early experience where he secured a meeting with a prospective client, only to discover that the individual he was scheduled to meet was the same executive who had recently inked a monumental deal with his company’s primary competitor. This realization presented an immediate and significant hurdle: "He has no interest in listening to me because he bet his career on that competitor," Tavakoli-Shiraji explained. Furthermore, he recalled instances where promising deals dissolved due to the startup’s unpreparedness to address the rigorous demands of enterprise-level legal, contractual, and security vetting processes. These early setbacks underscored the critical need for a strategic approach that extended far beyond the technological merits of their product.
To help startups avoid these pitfalls and navigate the intricate path to securing substantial enterprise clients, Tavakoli-Shiraji outlined three key strategies:
1. Leverage the Power of Strategic Partnerships: Selling Under a Larger Tech Company’s Umbrella
In its nascent stages, Databricks encountered significant resistance when attempting to secure a meeting with UBS, a prominent global financial institution. The breakthrough came not through a direct sales push, but through a strategic partnership with Microsoft. By becoming an Azure cloud partner, Databricks found its sales agreements subsumed under Microsoft’s established contractual framework. This alliance offered a dual advantage: Microsoft’s formidable sales force actively facilitated meetings and championed Databricks’ offerings to their existing enterprise customer base. "All of a sudden, that meeting wasn’t hard to get," Tavakoli-Shiraji recalled, emphasizing the transformative impact of this collaboration. "It was a huge boon to us that we were able to partner with somebody who had that access and distribution."
This symbiotic relationship proved instrumental in accelerating Databricks’ market penetration. With the active support of Microsoft’s sales teams, the startup began to experience unprecedented demand, leading to a surge in engagement. While this influx of interest presented new challenges, such as prospects requesting features for geographical markets where Databricks had not yet established a presence, these were ultimately considered "good problems to have" – indicators of robust growth and market traction. Today, the dynamic has reversed, with Databricks now serving as the larger entity that collaborates with startups, integrating their products under Databricks’ own stringent terms and advanced cybersecurity protocols. Notable examples of such collaborations include Hightouch, a provider of AI-powered marketing and customer data solutions (in which Databricks has also invested), Addepar, which offers sophisticated data tools for investment professionals, and Tecton, a machine learning startup that Databricks subsequently acquired. This strategic partnership model provides a powerful mechanism for emerging companies to bypass initial market access barriers and gain credibility by association.
2. Addressing the Enterprise Concerns in AI: The Four Cs Framework
When engaging with enterprises about Artificial Intelligence solutions, Tavakoli-Shiraji advocates for a structured approach by addressing what he terms the "four Cs": context, choice, cost, and control. Each of these pillars represents a fundamental concern that enterprise decision-makers grapple with when considering the adoption of AI technologies.
-
Context: This refers to the AI’s ability to be trained and adapted to the specific operational environment of the enterprise. Tavakoli-Shiraji likens this to onboarding a new employee, where the AI must learn and internalize the company’s unique terminology, established processes, strategic priorities, and cultural nuances. A solution that requires extensive custom development to understand a client’s domain specific language or workflows will face significant adoption hurdles. Therefore, demonstrating an AI’s capacity for rapid and effective contextualization is paramount.
-
Choice: In the rapidly evolving field of AI, enterprises are acutely aware of the potential for technological obsolescence and vendor lock-in. The concern for "choice" revolves around the ability to switch AI models or providers without incurring prohibitive costs or operational disruption. This implies that AI solutions should ideally be built on open standards or offer flexible integration capabilities, allowing companies to adapt to new advancements or pivot if a particular model proves suboptimal. The ability to retain agency and avoid being perpetually tied to a single vendor is a critical consideration for long-term strategic planning.
-
Cost: The economic implications of AI implementation are a primary driver of enterprise decision-making. "Cost" encompasses not only the initial licensing or subscription fees but also the ongoing operational expenses associated with running AI models. This includes computational resources, maintenance, and potential future upgrades. Enterprises require clear visibility and predictable cost structures to manage their budgets effectively. Demonstrating a strong return on investment (ROI) and providing cost optimization strategies are therefore essential components of any AI sales proposition.
-
Control: Perhaps the most critical of the four Cs, "control" pertains to the safeguards and governance mechanisms that dictate the AI’s behavior and, crucially, protect sensitive corporate data. Enterprises are increasingly concerned about data privacy, security breaches, and the ethical implications of AI decision-making. Solutions must offer robust security features, compliance with relevant regulations (such as GDPR or CCPA), and transparent audit trails to ensure accountability. The ability to manage and oversee the AI’s operations, preventing unintended consequences or misuse, is non-negotiable for widespread enterprise adoption. By systematically addressing these four Cs, startups can preemptively alleviate enterprise concerns and build a strong foundation of trust and confidence in their AI solutions.
3. Disrupting Legacy Systems: Unlocking Value in Entrenched Infrastructure
Tavakoli-Shiraji often engages CIOs with a disarmingly simple question: "On a scale of one to five, where do you stand with AI adoption?" A "one" signifies rudimentary access to tools like ChatGPT, while a "five" represents a profound integration of AI capabilities that has fundamentally reshaped business processes. Predictably, most executives gravitate towards "a three leaning towards four." However, Tavakoli-Shiraji observes that "99% of the ones I talk to are in bucket one and two." This stark reality highlights a significant gap between aspiration and execution in enterprise AI deployment. Many organizations have yet to fully harness AI agents in ways that deliver tangible value beyond assisting engineering teams with code generation. He posits that "It’s going to take many, many years before people can actually properly deploy AI in a very meaningful way in the enterprise."
This prolonged adoption cycle presents a substantial opportunity for startups, particularly those capable of modernizing the expensive and often unwieldy legacy systems that continue to underpin critical enterprise operations. Tavakoli-Shiraji paints a vivid picture: "If you go to any bank today, they still have [an IBM] AS/400 and mainframes. Nobody wanted to touch it because nobody knew what that code did." These archaic systems, deeply embedded within financial institutions and other large corporations, represent a significant technological debt and a barrier to innovation.
However, the advent of advanced AI agents is fundamentally changing this landscape. These AI systems possess the remarkable ability to "go in and understand and actually decompose what that code does," Tavakoli-Shiraji explains. This newfound capability opens a vast frontier for startup innovation. Products can now be developed to provide modern, user-friendly interfaces for these antiquated systems, breathing new life into decades-old infrastructure. Even more ambitiously, AI can facilitate the complete replacement of these legacy systems, migrating functionalities to more agile, cloud-native platforms. This strategic focus on modernizing and replacing deeply entrenched, yet poorly understood, legacy systems offers startups a compelling value proposition: the ability to unlock significant cost savings, enhance operational efficiency, and enable new business capabilities that were previously unattainable due to technological constraints.
Tavakoli-Shiraji also shared his forward-looking perspectives on the trajectory of enterprise AI adoption, further enriched by an anecdote from Databricks’ early days. He recounted an instance where he humorously admitted to pretending to be a lawyer to secure a crucial enterprise contract, underscoring the lengths to which startups may need to go to navigate complex sales environments and demonstrating the blend of technical acumen and strategic maneuvering required for success. A more in-depth exploration of these insights and predictions can be found in the full interview available below, offering a comprehensive look at the challenges and opportunities in the enterprise AI market.
The accompanying video provides further context on Databricks’ journey and Tavakoli-Shiraji’s expert advice on overcoming common hurdles in closing enterprise AI deals, particularly those that falter after the initial pilot phase. This content, while potentially involving affiliate links for which a commission may be earned, remains independent of editorial decision-making.
