What Benn Stancil's Trust-Signal Method Taught Our Context Catalog Agent

Benn Stancil writes a weekly essay at benn.substack.com about data, analytics, and the industry building it. He co-founded Mode Analytics and has spent years thinking carefully about why analytics fails organizations — not technically, but organizationally and epistemically. His writing is unusually honest: he is as likely to argue against his own industry's assumptions as he is to celebrate them.

Reading through his archive, one thread runs through almost everything: the problem with data is not that numbers are wrong. The problem is that people cannot tell when to trust them. And right now, the answer to 'can I trust this?' mostly depends on invisible cues — who built it, when it last ran, whether it looks right — rather than on anything the catalog actually surfaces.

The Trust Problem, Precisely Stated

Stancil puts it plainly in 'The Modern Data Experience': "Can I trust this?" is one of the most frustrating — and common — questions people ask of data. Our current answer relies on implicit signals: 'Who built it? Did it change recently? Does it look right?' That ambiguity drives teams into what he calls 'endless multi-tool chases to confirm results' — Slack threads, ad-hoc SQL checks, analyst email chains — before anyone will act on anything.

The consequence is not just wasted time. In 'What do we do when we get it wrong?', he makes the stakes explicit: 'Despite its sheen of scientific rigor, data is a confidence game; the only currency we have is trust.'

Three Principles That Do the Real Work

• •Explicit over implicit. Trust signals — freshness, authority, ownership, consistency — must be cataloged and visible, not inferred from context clues.
• •Inductive trust over input validation. Confidence in this month's numbers depends on whether they match last month's — a consistency record over time, not a one-time audit of the pipeline.
• •Decisions over verification. His 'method for measuring analytical work' lands on a single metric: 'Analysts should judge their work by how quickly people make decisions with it.'

How a Method Becomes a Skill

The dw-context-catalog agent's job is to make context available to downstream swarm agents and human consumers. But context without trust signals is noise — a table's schema and lineage mean nothing if the consumer cannot tell whether the table is fresh, authoritative, or contested.

Stancil's method gives the agent a precise mandate: do not just retrieve context, retrieve trust-legible context. That means running freshness and staleness checks and surfacing them in human-readable terms, marking authoritative sources with the rationale for why they are authoritative, importing tribal knowledge from Slack and Confluence into the catalog as explicit trust notes, and flagging documentation gaps as the trust debts they actually are.

One of More Than 400

The trust-signal-cataloging skill is one of more than 400 method-named skills across 19 agents in the Data Workers swarm.

A note on this post: This is independent commentary and homage. It distills publicly available writing and talks by Benn Stancil to illustrate a working method, and every quote is drawn from and verified against the primary sources linked above. The skill it describes is named for the method, not the person, and contains no marketing claims attributed to them. Data Workers is not affiliated with, sponsored by, or endorsed by Benn Stancil. If you are Benn Stancil and would like anything adjusted or removed, email hello@dataworkers.io and we will respond promptly.