From Question to Result: A Mental Model for Semantic Analytics

-

 

Part 1: A Thought Experiment: What If Analytics Models Were Semantic, Not Structural?

Part 2 : Using a Semantic Model as a Reasoning Layer (Not Just Metadata)

Let’s take a concrete example:

“Compare forecast vs actual sales by month for bike categories this year.”

Most systems jump straight to execution.

I think that’s backwards.

Step 1: Understand the Question

Before touching data, the system identifies:

  • Measures: Forecast, Actual Sales

  • Time range: This year

  • Time level: Month

  • Product level: Category

  • Filter: Bike-related categories

This is interpretation, not computation.

Step 2: Validate Meaning

Next, the system checks:

  • Are forecast and actual comparable?

  • Is category a valid rollup for both?

  • Is monthly aggregation defined?

  • Are defaults available where ambiguity exists?

If something is unclear, the system can explain why.

Step 3: Decide How to Answer

Only now does execution matter:

  • Cached aggregates

  • Precomputed tuples

  • On-the-fly calculations

Performance becomes a strategy choice, not a constraint leak.

Step 4: Present with Intent

The result isn’t just numbers.

It reflects:

  • The levels chosen

  • The measures compared

  • The assumptions applied

Which makes the output explainable.

Why I Think This Matters

As systems become more AI-driven, ambiguity becomes dangerous.

If we want:

  • Trust

  • Explain ability

  • Safe automation

Then meaning can’t be implicit.

It has to be modelled.

Closing Thought

This isn’t a finished architecture — it’s a line of thinking.

But I’m increasingly convinced that the future of analytics isn’t:

  • Faster queries

  • More charts

  • Bigger models

It’s better representations of meaning.

If you’re thinking along similar lines, I’d love to compare notes.



Comments

Post a Comment

Popular posts from this blog

Using a Semantic Model as a Reasoning Layer (Not Just Metadata)

-
Image
  Part 1: Using a Semantic Model as a Reasoning Layer (Not Just Metadata) Most systems treat semantic models as documentation. But what if they were active participants in query execution? This question came up while thinking about natural language analytics. Natural Language Is Ambiguous by Default If someone asks: “Show me sales by product this year” There are immediate ambiguities: Which sales measure? At what product level? Calendar or fiscal year? Gross or net? Most tools resolve this by: Picking defaults Or asking the user to clarify But what if the system could reason about the question? The Semantic Model as Context Imagine a model that explicitly defines: Valid measures Valid rollups Default aggregation logic Synonyms and aliases Before executing anything, the system can ask: Is this aggregation valid? Are these measures comparable? Does this level change meaning? This shifts failure from: “The numbers look...
Read more

A Thought Experiment: What If Analytics Models Were Semantic, Not Structural?

-
Image
I’ve been thinking a lot about why analytics and forecasting platforms feel harder to use than they should. Not harder to build,  harder to think with . Most modern data stacks are incredibly capable: Fast databases Columnar storage In-memory caches ML forecasting libraries Yet the questions users struggle to answer haven’t changed much: “Compared to what?” “At what level?” “Is this rolled up correctly?” “Why does this number look wrong?” This feels less like a compute problem and more like a meaning problem . Where Meaning Lives Today In most systems I’ve worked on, meaning is scattered across: Database schemas ETL logic Cube definitions BI metadata Tribal knowledge None of this is explicit. If someone asks: “Can we compare forecast vs actual by category this year?” The system doesn’t reason about that question. It executes SQL and hopes the result makes sense. A Different Framing What if we treated analytics as a sem...
Read more

Dev Tunnels with Visual Studio 2022 and Visual Studio 2026

-
Image
  Securely exposing your local development environment to the world — explained Introduction Developers building web APIs and web applications often face a familiar challenge: how to expose a localhost app to external services, collaborators, mobile devices or testing environments without deploying it to production. Enter Dev Tunnels ,  a powerful feature in Visual Studio that allows you to securely share your local development server over the internet with a public or authenticated URL. It dramatically simplifies workflows involving webhooks , remote debugging, and cross-device testing.  In this article, we’ll explore what Dev Tunnels are, how they work in Visual Studio 2022 , what’s new in the context of Visual Studio 2026 , and practical scenarios for using them. What Are Dev Tunnels? Dev tunnels let you expose a local web app running on localhost to the internet using a secure tunnel endpoint. Once created: You get a remote HTTPS URL that maps to your loc...
Read more