Definition: AI recommendation is not a switch. Brands can move through being discovered, entering the consideration set, and earning an accurate recommendation. Winin identifies which stage a gap sits in and prioritizes work by evidence.
Website summary: Reach solves “can you be seen”; Shortlist solves “are you considered”; Accurate & recommended solves “do they trust you, do they pick you”.
Canonical stage explanations
REACH
Question: Can AI read your information?
Focus on whether information can be discovered and read; absence reduces chances of entering the consideration set.
SHORTLIST
Question: Will it put you on the candidate list?
Being seen is not being considered; “picked a competitor, not you” can still happen.
ACCURATE & RECOMMENDED
Question: Is your brand described accurately — and recommended?
Mention ≠ selection; common issues include mentioned-not-recommended, stale price, discontinued, and other misstatements.
Four outcomes of a customer question (problem framing)
The website uses one pre-purchase question to illustrate four outcomes: Absent / Wrong / Not recommended / Blind spot. See learn/ai-brand-visibility.
What this page is / is not
- Is: Canonical definition of stage gates and measurement semantics.
- Is not: A universal promise of illustrative figures such as “0/9”; homepage diagrams illustrate capability only.
- Glossary version: glossary/reach-shortlist-recommended.
FAQ
Q: Is Reach alone (making content crawlable) enough?
A: No. You still need Shortlist and accurate recommendation reasons—and same-conditions retest to verify direction of change.
Q: Relation to GEO / AEO / SEO?
A: See learn/geo and site FAQ-aligned wording.
Using stage gates in weekly reports
Report three counts—not only “mention rate”:
- Reach-related: on how many models key prompts yield valid, non-stale brand information.
- Shortlist-related: how many answers put the brand on the candidate list.
- Accurate & Recommended-related: how many answers describe correctly and recommend / give recommendation reasons.
This helps leadership separate “not seen at all,” “seen but not shortlisted,” and “shortlisted but wrong reasons or stale price.” Illustrative dashboard scores explain product capability only and are not a promise of customer outcomes.
Four outcomes: learn/ai-brand-visibility; loop: loop.
Canonical source: https://winin.ai · Updated 2026-09-11