Data Delivery Options
Data Monetization Architectural Components
If you are working on a data monetization product chances are you are rightfully focused on designing a useful data delivery layer – after all, this is the component in your stack where your data first enters your customers’ workflows and thus where they begin to realize value from your efforts.
There are many options in composing a delivery layer. A product owner's primary goal when designing a delivery layer should be to maximize the value that their prospective customers can realize from using the product; and, that often means attempting to drive the data product as deeply into the customer’s workflow, and as close to the moment of data activation, as practical.
In other words, your product should attempt to connect the dots between the production of insight and the realization of value in your customer’s organization. Bringing insight and action close together, ideally in a way you can measure, reduces friction for your customer and increases stickiness for your product.
There are many data delivery options available and it’s critical to develop a clear understanding of your customers needs and workflows to design the optimal solution. The following are the most common delivery options and rationales for choosing them. Note: these options aren’t mutually exclusive and most data products will combine several to offer a cohesive solution.
USER INTERFACES
View Only
The BI dashboard is a classic interface that is here to stay for the foreseeable future. This delivery option requires the data product team to have a clear point-of-view on the metrics business users find valuable and the presentation methods that they will connect with. Maybe it’s a collection of benchmarked KPIs organized by topic, or alternative analytical needs like a drill path to support anomaly diagnosis. Companies will often build this delivery channel as a free or entry-level option to showcase the quality and value of their data product to upsell other delivery options.
Self-Directed Analysis
In this data delivery option, users are empowered with the ability to conduct self-directed analysis through interactive interfaces. This involves features like customizable dashboards, intuitive filters, and dynamic visualizations. Alternatively, you can offer a GPT-style chat interface to your data, sitting on a semantic model, to provide high quality analysis. The goal is to give users the flexibility to explore and analyze the data based on their specific needs and questions. This option is ideal for users who prefer a more hands-on approach to data exploration and decision-making.
Analyze + Act
Taking user interaction to the next level, this delivery option not only enables analysis but also facilitates immediate action based on insights gained. Users can analyze data trends and patterns and seamlessly transition to taking informed actions within the same interface. It's a powerful option for those who want to bridge the gap between analysis and implementation, streamlining decision-making processes.
User Write-backs
Write-backs provide a mechanism for users to perform analysis and consequently input or update the dataset based on their findings. This bidirectional interaction ensures that the data product is not only a source of insights but also a platform for users to contribute their findings or updates. An example of a use case might be allowing users to explore, define and persist market segments that can either be used in downstream analysis or activation. This collaborative approach enhances data accuracy and relevance over time.
MACHINE INTERFACES
Data Applications
This strategy involves deploying a workflow to extract data from a customer’s data warehouse, enriching it with ML or joining it to proprietary data sources, and writing back the results to the customer’s datastore OR via a pre-built connector to a SAAS product like a CRM or ERP. It is often used when the data product requires blending customer data and proprietary data or models, with results consumed downstream in data science or activation platforms.
APIs
For seamless integration into existing workflows and applications, APIs play a crucial role. This option allows developers to connect the data product with other tools and systems, enabling a more cohesive and interconnected data ecosystem. It's a preferred choice for organizations looking to embed data functionalities directly into their existing software infrastructure; while offering flexibility in terms of access management and query composition.
Cloud storage, ftp, etc.
For those who prefer traditional data delivery methods, cloud storage, FTP (File Transfer Protocol), and similar options provide a reliable means of accessing and distributing data. This approach is suitable for enterprises with a mature data platform in an industry that is highly accustomed to trading data in flat files.
Data Marketplaces
Companies looking to share or monetize their data can leverage data marketplaces. This delivery option provides a platform for buying and selling datasets is an excellent choice for organizations aiming to extend the reach and impact of their data products beyond their immediate user base, by leveraging the discoverability and ease-of-installation provided by a marketplace.
