# Desk log KB-VS-SL-MARGIN-20260822

**Status:** First-party sanitized composite; not a customer extract, independent benchmark, third-party dataset, certification, media report, or endorsement
**Page:** https://infinisynapse.com/en/blog/knowledge-base-vs-semantic-layer
**Run ID:** `KB-VS-SL-MARGIN-20260822`
**Date:** 2026-08-22
**Operator:** InfiniSynapse Data Team
**Attestor:** William Zhu, InfiniSynapse cofounder ([GitHub @allwefantasy](https://github.com/allwefantasy))
**Contact for contradictions:** zhuhl@infinisynapse.com

## What this file is

A downloadable record of a compiled-margin contract versus a bound fee memo on a sanitized orders extract. It documents the six-step method, compiled 13.5, memo-exception flags, and artifact inventory. The companion [aggregate CSV](https://infinisynapse.com/blog-media/knowledge-base-vs-semantic-layer/downloads/aggregate-KB-VS-SL-MARGIN-20260822.csv) contains only the published aggregate rows—not the 13,500-row extract. The companion [verify script](https://infinisynapse.com/blog-media/knowledge-base-vs-semantic-layer/downloads/verify-KB-VS-SL-MARGIN-20260822.py) checks those published rows only. No independent reproduction was known as of 2026-08-28.

## Six-step method (reproducible)

1. Pick one authorized source. Prefer a replica or sanitized extract.
2. Write a small memo pack: field notes plus one signed report.
3. Keep or skip the metric contract you already have. Do not invent a warehouse you do not own.
4. Upload the pack, then bind it to that source. Binding is a separate click from upload.
5. Ask one goal. Label pack passages separately from contract filters.
6. Ask the same goal again. A successful replay keeps the same footnote and the same compiled grain.

## Source and goal

| Field | Value |
|---|---|
| Source | Sanitized 13,500-row orders extract |
| Source grain | One row per sanitized order record |
| Source version | Fixed desk extract dated 2026-08-22 |
| Collision | Compiled `margin` = invoice − COGS omitted last month’s marketplace-fee footnote |
| Pack metadata | Two-page finance memo; owner: InfiniSynapse Data Team |
| Bind mapping | Memo pack ↔ authorized orders extract |
| Retrieval configuration | Bound-pack retrieval; configuration values not published |
| Contract-only baseline | Same goal with the compiled measure and no bound memo |
| Goal asked twice | “What is last-month margin?” |

## Results

| Retrieval state | Compiled invoice−COGS | Memo fee exception retrieved | Both objects kept |
|---|---|---|---|
| Contract only | 13.5 | 0 | 0 |
| Knowledge base + semantic pair | 13.5 | 1 | 1 |

The compiled number did not change. The second run retrieved the fee footnote the contract omitted. Wall-clock for the paired rerun was 12 minutes (warehouse time excluded).

## Task artifacts

| Retrieval state | SQL | Memo | Chart | CSV | Fee footnote cited |
|---|---|---|---|---|---|
| Contract only | 1 | 0 | 1 | 0 | No |
| Knowledge base + semantic pair | 1 | 1 | 2 | 1 | Yes |

## Minimum disclosure for independent replication

Disclose source grain, row count, and version; memo and contract versions; compiled metric definition; bind mapping; retrieval and semantic configuration; query, SQL, and passages; the contract-only baseline; all results and failures; run time; and commercial conflicts of interest. Publish confirming and conflicting results.

## What you may cite

- Grain, collision, compiled 13.5, memo exception 0→1, artifact counts 1 / 1 / 2 / 1, fee footnote cited No → Yes, run ID, ~12 min wall-clock

## What you may not claim

- Customer uplift, named-logo case, independent benchmark, certification, raw dataset, official EEAT score, or external validation

## External context (not validation of this run)

- [NIST AI Risk Management Framework](https://www.nist.gov/itl/ai-risk-management-framework) (retrieved 2026-08-28)
- [dbt Semantic Layer](https://docs.getdbt.com/docs/use-dbt-semantic-layer/dbt-sl) (retrieved 2026-08-28)
- [Snowflake Cortex Analyst](https://docs.snowflake.com/en/user-guide/snowflake-cortex/cortex-analyst) (retrieved 2026-08-28)
- [LookML introduction](https://cloud.google.com/looker/docs/what-is-lookml) (retrieved 2026-08-28)
- [UK NCSC secure AI guidelines](https://www.ncsc.gov.uk/collection/guidelines-secure-ai-system-development) (retrieved 2026-08-28)
- [ISO/IEC 11179-3](https://www.iso.org/standard/78914.html) (retrieved 2026-08-28)
- [OWASP Top 10 for LLM Applications](https://owasp.org/www-project-top-10-for-large-language-model-applications/) (retrieved 2026-08-28)