| name | differentiator-scoping-by-provenance-not-signal |
| description | Use when scoping, defining, filtering, or categorizing something whose stated
purpose is to "highlight our strength / differentiator", "show what makes us
different", "surface what we have that [a competitor / a prior vendor / another
team] doesn't", or "lead with our unique value". Triggers during brainstorming
/ requirements / taxonomy design — the moment you're about to lock the
definition of the category the deliverable hangs on. The trap: defining the
category by its broad CONCEPT (the signal) instead of by PROVENANCE (the
source). A concept-based definition silently includes items that match the
concept but come from the same source the competitor already has — so badging
them as "our differentiator" invites "we already had that". Use also when
reviewing such a definition, or when an advisor / reviewer flags that a scope
answer "conflates source with signal".
|
| author | Claude Code |
| version | 1.0.0 |
| date | "2026-05-20T00:00:00.000Z" |
Differentiator scoping — define by provenance, not by the broad signal
Problem
When the goal of a piece of work is to highlight a differentiator — what
your team / product / dataset has that someone else doesn't — the foundational
decision is what counts as that differentiator. The natural move is to define
the category by its broad concept: "behaviour", "engagement", "AI-powered",
"real-time", "first-party". That definition is clean and easy to agree to.
It is also the wrong cut. A differentiator claim is about provenance — the
source of the thing, not the concept of the thing. A concept-based
definition will quietly sweep in items that genuinely match the concept but are
sourced from exactly the system the competitor already has. Badge those as
"our unique strength" and the client can fairly answer: "we already had that."
The conflation is hard to see because the concept word and the
differentiator claim feel like the same set — but they aren't.
Context / Trigger conditions
- A user / client / stakeholder asks to "highlight our strength", "show our
differentiator", "surface what we have that the other agency / competitor / prior
vendor doesn't", "lead with our unique value".
- You are in brainstorming / requirements / taxonomy design and about to lock
the definition of the category, filter, tag, or tier the deliverable hangs on.
- You are choosing between a "broad" and a "narrow" definition and the broad one
looks cleaner / is the recommended-looking option.
- An advisor or reviewer says the scope answer "conflates source with signal",
"over-claims the differentiator", or "the framing is source-specific but the
definition isn't".
Solution
-
Separate the two dimensions explicitly. For the category in question, name:
- Signal / concept — what the thing is (the broad idea).
- Provenance / source — where the data comes from / who can observe it.
These are different axes. The differentiator lives on the provenance axis.
-
Run the leak test. Try to name one item that matches the concept but
is sourced from the thing the competitor already has. If you can — the
concept-based definition is leaking, and badging that item as "our
differentiator" is an over-claim.
-
Define the category on provenance. Cut the scope by what only we can
see / produce, not by the broad concept. If the concept genuinely spans both
provenances and both are valuable, the honest resolution is usually a
two-tier split: a strict "core differentiator" tier (provenance = ours
alone) plus a broader "supporting" tier (real, but shared provenance) —
labelled honestly so the supporting tier is never sold as the unique thing.
-
Re-ask the scoping question with the source dimension made explicit. If
you already asked and got a concept-based answer, that answer was made
without the source dimension on the table — it doesn't count as informed
consent to the conflation. One extra clarifying round-trip is far cheaper than
shipping a mislabeled differentiator to a client who will notice.
Verification
- Every item inside the "differentiator" category passes the leak test: none of
them is sourced from what the competitor already has.
- If a two-tier split was used, the supporting tier is labelled with its true
provenance and is never referred to as the unique strength.
- The person who asked can articulate why each tier is or isn't the
differentiator — in provenance terms, not concept terms.
Example
A dashboard client asked to "surface more on-site behaviour insight — highlight
our strength, because a prior vendor already did the CRM analysis."
- Concept-based definition (the trap, initially recommended and accepted):
"behaviour = anything the user actively does" — web visits, logins,
and campaign/event attendance, checklist completion.
- Leak test: campaign attendance and checklist completion are real
"behaviour" — but they are recorded in the CRM. The prior vendor, who
worked from the CRM, already had them. Badging them "our differentiator"
would have been an over-claim.
- Provenance-based resolution (two-tier): Tier 1 = portal clickstream (web
visits, searches, logins — observable only in our own first-party event
stream, genuinely unseen by CRM-based work); Tier 2 = observed engagement
(attendance, checklist — real, but CRM-sourced, labelled honestly as
supporting). The whole deliverable was then built on the two-tier definition.
The concept-based answer had been recommended by the model and picked by the
user — it took an advisor pass to surface that the differentiator goal demanded
a provenance cut. The fix was one re-asked question.
Notes
- The tell: the concept word ("behaviour", "engagement", "AI", "real-time")
and the differentiator claim ("what others don't have") name different
sets. Whenever a task pairs a broad concept with a "this is what makes us
different" goal, suspect the gap.
- This is a brainstorming / scoping-time skill — it fires before implementation.
Catching it after the taxonomy ships means re-cutting everything downstream.
- Honest labelling beats a bigger-looking category. A smaller, provenance-true
differentiator that survives "we already had that" is worth more than a broad
one that doesn't.
- Related: a two-tier split is also the honest move when a metric or claim mixes
strong and weak evidence — see
single-anchor-point-estimate-needs-range-framing
for the dashboard-copy sibling of "don't let a clean framing overclaim".