The Functional Blueprint

What Data Products
Actually Do.

Companies like DataKnobs, a leading player in the industry, don't simply create data products for storage purposes. Instead, they design them with a focus on executing 5 essential functions, transitioning from establishing accuracy to automating intricate business processes.

5 Capabilities of Data Products: Reveal Reality, Probabilities, Compare, Predict, Recommend

The Capability Spectrum

Data products created by leading organizations are crafted to provide one or more unique functional results.

Capability 1

Reveal Reality

All data work begins with a solid foundation. This includes gathering disorganized raw data from various sources, refining it, merging it, and creating a comprehensive 'Single Source of Truth.'

Example: Combining CRM data, web traffic logs, and billing history to create a cohesive 'Customer 360' profile.
Capability 2

Give Probabilities

Advancing from deterministic reporting to statistical modeling, this feature predicts the probabilities of different outcomes or states by analyzing both current and past data.

Example: Determining that there is a 75% chance of a particular shipment being delayed because of the current weather conditions.
Capability 3

Enable Comparison

Organizing the data in a way that enables users or algorithms to compare entities, highlighting anomalies, patterns, and performance relative to others.

Example: A dynamic dashboard that displays the conversion rate comparison between Cohort A and Cohort B for the past 30 days.
Capability 4

Predict Outcome

Using machine learning to predict future outcomes, the data product takes in present data and produces a well-informed estimate of what is to come.

Example: Predicting Q4 revenue with precision using the current sales pipeline and past seasonal win rates.
Capability 5 (The Pinnacle)

Recommend & Take Action

The most advanced type of data product. It recommends the best course of action or can bypass human involvement to automatically carry out tasks in operational systems.

Example: The recommendation engine is sending a personalized 15% discount code via email to a user who is at high risk of churning.
Capability Maturity

Navigating the Maturity Curve

Many organizations mistakenly try to develop 'Automated Action' products (Capability 5) before fully understanding 'Revealing Reality' (Capability 1).

The 5 capabilities build upon each other, making accurate predictions impossible with fragmented, inaccurate data. Leading data companies gradually improve their products along the maturity curve.

Descriptive
1. Reveal
2. Prob.
Diagnostic
3. Comp.
Predictive
4. Predict
Prescriptive
5. Action
Cumulative Business Value Generated

Assess Your Data Capabilities

How advanced are your current data products in terms of maturity? Move away from simply reacting to raw data and begin creating automated, prescriptive strategies.

Review AI & Cognition