Navigating the Build vs. Buy Dilemma: A Pivotal First Step in Modernizing Your Data Stack

• Factors to consider for building enterprise-level data technology include assessing in-house capabilities, strategic importance, deployment urgency, and conducting a comprehensive cost analysis.
• User experience is crucial for success, influencing the build vs. buy decision.
• The ultimate goal is to harness data's transformative potential to drive business growth and innovation.

Building data products is a journey that spans the entire data lifecycle, from initial collection and storage to the final stages of analysis and visualization. The toolkit businesses require for this endeavor varies significantly, influenced by the complexity of the data product, the operational scale, and the unique challenges the business seeks to address. 

As a seasoned data executive, I've navigated the strategic crossroads that could pivot a company's trajectory toward either remarkable success or potential stagnation. A critical choice in this journey is deciding between leveraging off-the-shelf data product solutions or dedicating resources to developing bespoke in-house systems. In this article, I’ll dissect the intricacies of this pivotal decision, offering insights from my tenure as a data and business executive that will aid businesses in navigating this complex landscape. 

Whether the data products are designed for internal optimization or intended for external monetization, the underlying decision-making process remains strikingly similar. In my experience, two crucial aspects of data product delivery have stood out:

1. Synthesizing Disparate Data to Create the Data Product

The ability to transform isolated data elements into a cohesive, actionable source of truth is paramount. This unified information becomes the backbone for informed business decisions and streamlined operations. The core question often revolves around whether to construct specific technologies in-house or to opt for pre-built solutions. This decision is not trivial and warrants a detailed analysis of the organization's unique circumstances, including several key considerations:

In-house capability assessment:

The starting point often involves a realistic evaluation of the internal team's expertise and capacity. I faced this very assessment while spearheading the Product 360 initiative at a Fortune 100 healthcare firm. I queried our data science team about our ability to develop a data harmonization machine learning model internally. Given our prior success in crafting digital twins, we were confident in our in-house capabilities to embark on this new venture. This self-assurance stemmed from a solid foundation of relevant experience within our team, affirming our decision to proceed with building the solution.

Strategic importance of the mastering solution:

Another critical aspect is the significance of the mastering solution in the company's competitive landscape. If the solution is integral to the company's value proposition, crafting it internally could provide strategic advantages. In the context of Product 360, an in-house build was feasible. However, we recognized that the solution's application would be primarily for specialized internal business functions aimed at preemptively identifying revenue-generating opportunities. Despite the potential for internal utility, the solution wasn't suited for external monetization. Moreover, the commitment to maintaining the solution was substantial, especially as the company navigated through extensive mergers and acquisitions which required the frequent integration of new products and business lines.

Urgency of deployment:

The immediacy with which the business needs to implement the solution can greatly influence the build-vs-buy decision. Our initial analysis of the business value proposition for Product 360 revealed a significant potential loss—up to $350 million in incremental global revenue—attributable to disjointed product data. The longer the development timeline is extended, the greater the missed opportunities. We estimated building an internal MVP with limited capabilities would take us at least six months of work with three full-time employees vs buying implementation-ready components of the solution will take less than three months.

Comprehensive cost analysis:

Beyond the upfront investment, it's imperative to consider the long-term financial implications, including maintenance, upgrades, and scalability. For Product 360, the in-house development and maintenance of the model necessitated a dedicated team comprised of a senior member, two junior resources, and additional support from the data engineering and UX departments. Even with the possibility of leveraging offshore resources, the ownership costs were projected to be substantial.

Urgency of deployment:

The immediacy with which the business needs to implement the solution can greatly influence the build-vs-buy decision. Our initial analysis of the business value proposition for Product 360 revealed a significant potential loss—up to $350 million in incremental global revenue—attributable to disjointed product data. The longer the development timeline is extended, the greater the missed opportunities. We estimated building an internal MVP with limited capabilities would take us at least six months of work with three full-time employees vs buying implementation-ready components of the solution will take less than three months.

Comprehensive cost analysis:

Beyond the upfront investment, it's imperative to consider the long-term financial implications, including maintenance, upgrades, and scalability. For Product 360, the in-house development and maintenance of the model necessitated a dedicated team comprised of a senior member, two junior resources, and additional support from the data engineering and UX departments. Even with the possibility of leveraging offshore resources, the ownership costs were projected to be substantial.

2. Enhancing User Experience

Beyond the technical accomplishment of data integration, the interaction between the user and the data product plays a crucial role in its success. The most impactful data products foster a dynamic environment of continuous feedback, enabling users to learn from and contribute to the ongoing enhancement of the product, thereby maximizing its value. 

The build vs. buy decision is significantly influenced by the desired level of user engagement and the ability of the data product to seamlessly integrate into existing business workflows. A compelling illustration of this comes from my involvement in P&L restructuring, which highlighted the pivotal role of user experience in these strategic decisions. 

In an era where businesses are reevaluating their P&L models, the ability to offer customers a comprehensive view of your offerings is becoming increasingly crucial. A particularly enlightening conversation with a Regional VP drew a memorable comparison of our vendor-built AI data harmonization platform to Tinder. This unexpected analogy highlighted the platform's intuitive design, which facilitated straightforward navigation and decision-making, akin to making connections on Tinder. This insight underscored the importance of crafting data products that not only serve functional purposes but also deliver intuitive and engaging user experiences.

The Journey to Decide Build vs. Buy

The journey to decide between buying and building a data product is complex, woven with strategic, operational, and financial threads. While off-the-shelf solutions may promise rapid deployment and proven reliability, custom-built systems offer the allure of tailor-made functionality and the potential for a competitive edge.

This journey is not merely about managing data; it's about harnessing data's transformative potential to spur business growth, drive innovation, and enhance customer engagement. Whether the choice leans towards purchasing a pre-built solution or investing in a custom build, the overarching goal remains to establish a robust, user-centric foundation for data-driven decision-making that resonates with your long-term business aspirations.

Hear Elena discuss AI integration strategies for business success on the Data Masters Podcast