A point-in-time, web-grounded benchmark across 25 shared questions, four providers, competitors, and citation evidence.
Executive verdict
Coverage threshold metIt reached 15 of 20 evaluated discovery questions across four providers. Distinct questions and provider-answer events are counted separately.
38%
of grounded discovery answers
Buyer-question reach
15/20
Mentioned by at least one grounded provider
Mixed provider signals
8
Distinct questions where providers disagreed
Unanimous misses
5
Distinct questions where every grounded provider missed
Owned site cited
25%
18 of 73 grounded discovery answers
Opportunity themes
Ranked by missed grounded provider answers, grouped into distinct buyer-question themes.
Missed in 9 of 14 grounded answers.
Luminary is the most frequent replacement.
Missed in 10 of 12 grounded answers.
Two questions were unanimous misses.
Missed in 7 of 15 grounded answers.
Northstar appears, but owned evidence is rarely cited.
Agreement is measured per distinct question. Mention rates describe provider behavior, not answer accuracy.
All mentioned Northstar
7
distinct questions
Providers disagreed
8
distinct questions
All missed Northstar
5
distinct questions
8 points above overall
4 points below overall
7 points below overall
1 point above overall
Answer-event frequency plus distinct-question reach.
15 distinct questions
19 distinct questions
14 distinct questions
8 distinct questions
Recurrence does not prove source quality or influence.
analyticsweekly.example
4 questions · 3 providers · 5 missed answer events
moderndatastack.example
3 questions · 4 providers · 4 missed answer events
saasopsreview.example
2 questions · 2 providers · 2 missed answer events
Prioritized action plan
Each action explains its evidence and links to a representative buyer question. Live reports also link every stored source.
Create an owned comparison page covering deployment model, supported warehouses, migration effort, limitations, and verifiable customer evidence.
Why: Northstar was absent across four distinct warehouse-native buyer questions, while Luminary appeared repeatedly.
Pitch a factual product profile or hands-on review to those publishers, with documentation and customer proof they can independently verify.
Why: Analytics Weekly and Modern Data Stack were cited across six missed answers spanning five buyer questions.
Add prerequisites, setup steps, time-to-value ranges, architecture diagrams, and dated documentation on the owned implementation page.
Why: Models mentioned Northstar across four implementation questions but often cited third-party summaries instead of northstar.example.
Publish a plain-language security and data-residency page with regions, retention controls, subprocessors, certifications, and review dates.
Why: Northstar was unanimously absent from three privacy-first buyer questions for regulated teams.
Add concrete best-fit and not-a-fit sections to the homepage and relevant use-case pages, supported by specific workflows and outcomes.
Why: Five distinct buyer questions produced incidental rather than shortlist-level mentions.
Supporting evidence
Provider agreement and the main competitor/source signal stay visible. Individual provider outcomes remain collapsed until needed.
Brand outcome
0 of 4 grounded AIs mentioned Northstar
Leading competitors
Luminary, MetricFlow
Key cited domain
analyticsweekly.example
OpenAI
Northstar not mentioned
Claude
Northstar not mentioned
Gemini
Northstar not mentioned
Grok
Northstar not mentioned
Brand outcome
1 of 4 grounded AIs mentioned Northstar
Leading competitors
Beacon, MetricFlow
Key cited domain
moderndatastack.example
OpenAI
Northstar mentioned
Claude
Northstar not mentioned
Gemini
Northstar not mentioned
Grok
Northstar not mentioned
Brand outcome
2 of 4 grounded AIs mentioned Northstar
Leading competitors
Luminary
Key cited domain
saasopsreview.example
OpenAI
Northstar mentioned
Claude
Northstar not mentioned
Gemini
Northstar mentioned
Grok
Northstar not mentioned
Execution quality for this benchmark—not a claim that every AI answer is factually correct.
Discovery questions
20
Grounded provider breadth
4
Grounded discovery answers
73
Excluded discovery answers
7
Discovery-answer coverage met the 90% threshold. Failed or ungrounded answers are excluded rather than counted as Northstar misses; brand-specific diagnostic questions are reported separately.
All four providers receive the same frozen question set.
Results come from model APIs with web grounding and can differ from consumer chat interfaces.
Rerun the exact questions after implementation and compare the deltas.
