返回案例Back to cases
案例 01 · 包材代工Case 01 · Packaging

妍创包装 · 香水玻璃瓶代工
120 条答案,0% 上线前基线
Yanchuang · Perfume glass bottles
120 answers, 0% pre-build baseline

这是基线案例,不是“优化后成功案例”。测试覆盖 8 类买家问题 × 3 个平台 × 每题 5 轮;现有证据只支持 0% 基线和执行路线,优化后同口径复测尚未完成。This is a baseline case, not a post-optimization success claim. Eight buyer-query groups × three platforms × five runs each; evidence supports the 0% baseline and execution route, while like-for-like post-build re-testing is pending.

0%
起点出现率baseline rate
120
可回查答案auditable answers
3
实测平台platforms tested
5
每题重复轮次runs per query
起点 · 挑战The challenge

好瓶子,海外 AI 里却完全隐形A great bottle, invisible in AI

在 ChatGPT、Perplexity 与 Gemini 的 120 条归档答案中,妍创出现率为 0%。这只能证明企业在该测试范围内未被提到,不能推断产品能力不合格,也不能用来概括所有 AI 平台或当前实时状态。Across 120 archived answers on ChatGPT, Perplexity and Gemini, Yanchuang's mention rate was 0%. This only proves absence in the tested scope; it does not judge product capability or describe every platform or the current live state.

执行路线(待复测)Execution route (re-test pending)

五层找断点,七步逐项修复Use five layers to locate gaps and seven steps to fix them

下列是基于基线的执行计划,不是已经发生的结果。The following is a baseline-informed plan, not a completed result.

🎯

① 锁定问法 + 信源解剖Target & map sources

用真实买家问法定位「香水玻璃瓶」这一战场,并解剖出 AI 在该问法下到底吃哪些信源。
🔒 具体问法清单与信源权重为专有方法proprietary
Pin the perfume-bottle battlefield and map which sources AI actually cites.

📄

② 官网知识化(内容地基)Citable site

在英文站围绕该品类系统化建设可被 AI 抽取的知识内容 + 结构化数据 + llms.txt。
🔒 内容结构与模板为专有方法proprietary
Build AI-extractable knowledge content + schema + llms.txt for the category.

🗂️

③ 攻占信源(目录 + 榜单)Win sources

在 AI 最常引用的几类信源上完成占位与优化,让"答案的来源"里出现妍创。
🔒 具体信源与进榜路径为专有方法proprietary
Place Yanchuang into the source types AI cites most.

💬

④ 真实证据 + 同口径复测Authentic evidence + like-for-like re-test

补充可验证案例、目录与行业信源;按相同平台、问法和轮次复测,比较五类审计输出。Add verifiable cases, directories and industry sources; re-test the same platforms, queries and runs across five outputs.

阶段门禁Stage gates

每一步先验收,再进入下一步Validate each stage before the next

WEEK 1

AI 体检 · 量基线Audit the baseline

真实问法多轮实测,确认起点出现率 0%、信源与竞品格局。Confirm 0% baseline, map sources & rivals.

出现率 0%rate 0%
WEEK 1–2

问题地图 · 锁定主战场Map & target

在香水玻璃瓶相关问法中,选定一个可赢的细分问法集中突破。Pick one winnable perfume-bottle query to focus on.

GATE 2

官网事实可抓、可理解Site facts crawlable and understandable

验收索引、直答内容、结构化数据与页面事实一致性。Validate indexing, direct answers, structured data and page-fact consistency.

GATE 3

企业实体与外部信源一致Entity and sources aligned

核对公司名、官网、品类能力、目录与第三方证据。Validate company name, domain, capability claims, directories and third-party evidence.

GATE 4

同口径复测Like-for-like re-test

复用基线平台、问法与轮次;复测完成前不填写提升比例。Reuse baseline platforms, queries and runs; no uplift is claimed before completion.

状态:待复测Status: pending
已验证数据Verified data

当前只公布基线,不预填结果Publish the baseline, not an invented outcome

出现率Mention Rate
0%
待同口径复测pending
样本Sample
120
8 问法 × 3 平台 × 5 轮8 queries × 3 platforms × 5 runs
实测平台Platforms
3
ChatGPT · Perplexity · Gemini
优化后结果Post-build result
未验证unverified
不宣称not claimed

说明:一次问法 × 一个平台 × 一轮计 1 条;数据为归档项目基线。AI 答案会变化,不承诺固定排名或推荐。One query × one platform × one run = one record. This is an archived project baseline; answers change and no fixed rank or recommendation is promised.

先建立你的真实基线,
再决定修哪里
Build your real baseline,
then decide what to fix

申请专业基线测试 →Request an expert baseline audit →

首选加微信 13925122895(微信同名)WeChat 13925122895