Insights / Blog Data products, data commerce, and beyond October 5, 2022 Anthony CosgroveCo-founder and Chief Strategy OfficerHarbr When you’re operating in a new category, you spend a lot of time defining who you are. To keep things simple, we call Harbr a data commerce platform. Our customers use the platform to commercialize data products and generate attributable revenue. So far, so simple. To understand how our customers achieve that, we’ve first got to unpack the concept of the data product, as it is fundamental to data commerce at any meaningful scale. There are many ways to define data products. But from working with our customers, who manage over 2,500 data products across Harbr platforms, we’ve come to the following definition: A data product is one or more data-related assets that are subject to active product management to deliver a defined value proposition to a target market. Notice that this definition doesn’t mention selling or commerce. Data commerce requires data products, but data products don’t have to be commercialized. More on this later, but first, let’s explore the concepts in this definition. Data products require the application of product management, not the project management that data is typically subjected to. “A product manager is in charge of a product’s success from the beginning to the end of its lifecycle, while project managers shepherd projects to completion.” — Coursera The purpose of a data product — like any product — is to optimize value throughout its lifecycle, which can be open-ended. Whereas the purpose of a project is to deliver a predefined value proposition that has a clear endpoint. In practice, products and projects are profoundly different, with distinct processes, skills, and success metrics. Crucially, the data product lifecycle overseen by the data product manager is no different to any other product lifecycle. Our customers consistently undertake the following activities as they build compelling data products: Seek to understand the available assets and any constraints attached to them e.g. technical, legal, cost, etc. Hypothesize on the potential value of data products that could be built and ranking them accordingly Prototyping data products and collaborating with would-be consumers to understand and prioritize the specification, use cases, and value proposition Building a minimum viable product (MVP) that addresses a specific business problem that needs to be solved — a defined value proposition This list may give the impression of a linear process — as you complete one step you move to the next — but the reality is that the process is iterative. Our customers regularly go back to earlier stages in the process having captured some relevant insight; much of what we do at Harbr is to enable them to accelerate the overall process. The target outcome is always a market-ready data product that generates value. To achieve this, your data products will benefit from, and cannot really exist without, input from the potential users and consumers — the target market. While this input will be gathered during the product development process, it doesn’t end there. Once a product is launched, there will be more market feedback that can be used to improve the fit, value, competitiveness, or market opportunity. It can also inspire a hypothesis for a new data product. Putting this feedback to good effect requires repeating some or all of the steps above. For anyone well-versed in agile product management, none of this should be news. However, data products do have at least one unique quirk that’s worth being aware of: data consumers almost always customize the data products they consume. Unlike consumers of most other products, consumers of data products will have highly unique needs that must be fulfilled before they can extract any value from a product. Variations across data consumers can include: Unique questions based on their use case Differing skill levels and technical know-how; this can include second-degree data consumers and stakeholders A need to combine the data product with their own data Acknowledging this quirk has profound implications for both the process of data product management and the experience created around a data product. Going beyond commerce Adopting a data product management mindset unlocks a variety of use cases: create, manage, govern, sell, exchange, move, customize, evaluate, distribute. When you productize something, you create all the conditions around the product that allows you to sell it. But you don’t actually have to sell it. For example, a large conglomerate is using its Harbr platform to allow for discovery, collaboration, and distribution of data products across different parts of their organization. They don’t have to sell these products to derive tremendous value from them. Join us on the journey Data businesses aren’t built overnight. And they don’t modernize overnight, either. But if you’re looking for inspiration from companies building scalable, high-margin data businesses that provide immense value to their data consumers, I can think of some great examples — both conveniently built on Harbr: DataHub from Moody’s Analytics CoreLogic’s Discovery Platform Ask me what a data product is this time next year, and I’ll have plenty of new stories about how our customers (and their consumers) are innovating with data products. If you can’t wait that long, drop us a line now.