In 2012 Gartner released their Analytic Ascendancy Model (Fig. 1), describing the relationship between the difficulty and value of distilling actionable analytics from data. Since then, the analytics world has seen a multitude of tools and platforms that answer the descriptive, diagnostic, and (lately) predictive questions: “What happened?”, “Why did it happen?” and “What could happen?” However, the ability to generate prescriptive analytics has remained firmly rooted in the realm of the data scientists and statisticians in specialized tools—well out of the reach of most business users who all need to rely on experts to answer the question: “How should we make it happen?”
The process of answering this last question has become known colloquially as “The Last Mile” and is considered the Holy Grail of data analysis—because it represents “the decision.” Simply put, this is what fully drives data-driven decisions, eliminating the guesswork and much of the flimsy interpretations. Not so simply, it requires complex business modeling, logic, and math to surface the solution. Due to the complexity of such systems, this type of functionality is typically out of reach for anyone who is not a statistician or data scientist, is rarely found outside of specialized tools, and is seldom surfaced to the common business user within broader analytics and business platforms.
In 2023 an independent survey of data workers and analysts that use BI and analytics toolsets across Finance, Marketing, Sales, and other LoB, showed that more than 50% of such users export analytic results to Excel more than 50% of the time; with 90% using the Excel desktop and 60% saving data to a local disk. When asked why they do this, the top 5 responses were:
They expressed frustration at the inefficiencies this caused; the difficulty of handling multiple static documents and their proliferation (aka “spreadsheet hell”); the inability to automatically tap into the most recent data; concern for weakened data security and governance; and the loss of a “single source of truth.”
The Pyramid Platform has been built around a “Data Factory for Decisions” (Fig. 2)—a view of the end-to-end data ecosystem across enterprise and corporate businesses. The data factory framework envisions the data journey as a closed-loop factory that takes raw data from creation to the end-user decision point. Since Pyramid’s inception, our goal has been to ensure that any business user can ingest any data, fix, and clean it; enrich it with predictive models and values; embellish the data with logic; discover information from the data through visualization and interrogation; and finally, share it with many downstream decision makers. To complete the data factory, Pyramid added Tabulate and Solve.
The latest extensions to our comprehensive platform are designed to close the gaps for users who need to create business models with live data while offering them a venue to construct and execute optimization logic to resolve decision models and deliver a prescriptive solution.
Pyramid’s “Tabulate” app—a browser-based virtual spreadsheet—is designed for business modeling by business users while leveraging the rest of the platform to bridge the gaps in “The Last Mile.” It delivers business modeling with live data through query mashups, multi-query visualization, and flexible formatting. These models can be embedded in presentations or shared via interactive dashboards and formatted reports.
Tabulate empowers analysts and lowers the bar for everyone from the C-suite to frontline workers through a familiar interface and access to the entire 500+ “Excel” formula library to build and deploy business models with live data wherever it resides—such as SAP BW, AWS Redshift, Snowflake, Microsoft cubes, or Pyramid’s own in-memory engine. The direct connectivity speeds time to insight and effectively eliminates risks associated with moving data out of the analytics solution to yet another standalone desktop tool.
With a wide variety of flexible formatting and layout options to meet the business modeling needs of all levels of users, Tabulate allows users to blend query results, extend logic, and add new visualizations. Static spreadsheets and email attachments are now a thing of the past because all content is dynamically embeddable in a live “Present” dashboard and sharable via report bursting from the “Publish” app. Apart from stemming the spreadsheet epidemic, end users no longer have to plow through the spreadsheet to surface the analysis buried inside it!
Pyramid extends Tabulate to include complex decision modeling and prescriptive analytics with its “Solve” plug-in. Solve is an engine that can generate prescriptive and optimization solutions based on user-designed decision models designed through a no-code/low-code paradigm in Tabulate. By leveraging the familiar formulaic and business modeling capabilities of Tabulate, users can build such models with complex logic waterfalls, conditions, decision points, and the utilization of the entire suite of spreadsheet formulas.
Further, Solve delivers true decision intelligence on live data—answering both “What should I do” questions and “What if” simulations—through a simple yet powerful UI experience that can be embedded in a Present dashboard. Now non-technical end users can view machine-generated decisions and easily adjust the inputs to observe changes in those decisions using Pyramid’s ultra-powerful engine, without ever seeing the underlying mechanics.
With Tabulate and Solve, enterprises can now create and deploy sophisticated business and decision modeling alongside a vast array of toolsets for preparing data, building and deploying data science and machine learning, creating and using data visualizations, and performing other analytical processes—all directly on any data set. In doing so, it greatly enhances the decision intelligence capabilities of the platform, bridging major gaps in “The Last Mile” of analytics.
Schedule your demo today to see what Pyramid can do for you.