What Is an SEO Audit? Process, Checklist & Examples什么是 SEO 审计？流程、清单与示例
An SEO audit is not a list of tool warnings. It is a scoped investigation that connects discoverability, content, experience, authority, and measurement evidence to prioritized action.
SEO 审计不是工具警告列表，而是一项有明确范围的调查：把可发现性、内容、体验、权威性与衡量证据连接到有优先级的行动。
What is an SEO audit?什么是 SEO 审计？
An SEO audit is a scoped, evidence-based review of how a website can be discovered, crawled, indexed, understood, experienced, trusted, and measured in organic search. It compares current behavior with intended behavior, identifies material gaps, tests plausible causes, and produces an owned plan for correction and validation.
SEO 审计是一项有明确范围、以证据为基础的审查，用来判断网站在自然搜索中如何被发现、抓取、索引、理解、体验、信任与衡量。它比较当前行为与预期行为，识别重要差距，检验可能原因，并产出有负责人、可修复且可验证的计划。
Short answer: an SEO audit explains what is limiting organic-search performance, why the evidence supports that conclusion, what to do first, who owns it, and how success will be checked.
简短回答：SEO 审计要说明什么在限制自然搜索表现、为什么证据支持这个结论、先做什么、由谁负责，以及如何验证成功。
An audit is more than a crawl export or score审计不只是抓取导出或评分
A crawler, performance test, or platform score can surface observations, but it cannot know the full business context, intended canonical behavior, content strategy, release history, market priority, or implementation constraints. A useful audit distinguishes an observation from an issue, an issue from an impact hypothesis, and a hypothesis from a proven cause.
爬虫、性能测试或平台评分可以暴露观测，但无法自动了解完整业务背景、预期规范化行为、内容战略、发布历史、市场优先级或实施约束。有效审计会区分观测与问题、问题与影响假设，以及假设与已证实原因。
When should you run an SEO audit?什么时候应该做 SEO 审计？
Run an audit when you need a dependable baseline, before or after a migration, redesign, domain change, international expansion, rendering change, information-architecture rebuild, major content program, analytics change, or unexplained performance shift. Also audit when ownership changes or technical debt has accumulated across releases.
当你需要可靠基线，或在迁移、改版、域名变更、国际扩张、渲染方式变化、信息架构重建、大型内容项目、分析配置变化或无法解释的表现波动前后，应执行审计。当负责人更替或技术债在多次发布中累积时，也需要审计。
A complete audit examines six connected evidence layers完整审计检查六个相互连接的证据层
Crawl paths, robots controls, sitemaps, canonicals, redirects, status codes, rendering, and index state.
抓取路径、robots 控制、站点地图、规范化、重定向、状态码、渲染与索引状态。
Queries, pages, impressions, clicks, CTR, position, devices, countries, features, and change cohorts.
查询、页面、展示、点击、CTR、平均位置、设备、国家、搜索特性与变化分组。
Coverage, usefulness, duplication, cannibalization, freshness, ownership, and page-purpose alignment.
覆盖、实用性、重复、关键词内耗、新鲜度、所有权与页面目的匹配。
Mobile behavior, performance, accessibility, interaction, templates, structured data, and rendering stability.
移动端表现、性能、可访问性、交互、模板、结构化数据与渲染稳定性。
Internal links, external references, backlink patterns, reputation, entity consistency, and trust signals.
内部链接、外部引用、外链模式、声誉、实体一致性与信任信号。
Data quality, annotations, owners, dependencies, impact hypotheses, effort, acceptance, and monitoring.
数据质量、注释、负责人、依赖、影响假设、工作量、验收与监控。
Define scope and expected behavior before collecting issues收集问题前先定义范围与预期行为
Document the business objective, sites and subdomains, markets, languages, devices, templates, URL volume, rendering model, analytics properties, Search Console properties, release history, known incidents, seasonality, conversions, constraints, stakeholders, and deadline. State what is explicitly out of scope.
记录业务目标、站点与子域、市场、语言、设备、模板、URL 数量、渲染模型、分析资源、Search Console 资源、发布历史、已知事故、季节性、转化、约束、利益相关者与截止日期，并明确哪些内容不在范围内。
Then write expected behavior: which URLs should be discoverable, indexable, canonical, localized, rendered, linked, measured, and converted. Without an expectation, a tool can label intentional behavior as an error or miss a business-critical exception.
然后写清预期行为：哪些 URL 应被发现、索引、规范化、本地化、渲染、链接、衡量与转化。没有预期，工具可能把有意行为标成错误，也可能漏掉业务关键例外。
Audit crawl, rendering, canonicalization, and indexation审计抓取、渲染、规范化与索引
Map how users and search engines reach important templates. Review navigation, internal links, orphan candidates, pagination, faceted paths, parameters, redirects, status codes, robots.txt, meta robots, canonicals, XML sitemaps, hreflang where relevant, JavaScript rendering, duplicate routes, soft errors, and index state.
绘制用户与搜索引擎到达重要模板的路径。检查导航、内部链接、疑似孤儿页、分页、筛选路径、参数、重定向、状态码、robots.txt、meta robots、规范化、XML 站点地图、适用时的 hreflang、JavaScript 渲染、重复路由、软错误与索引状态。
Use multiple views because no single source is complete: site crawl, sitemap inventory, analytics landing pages, Search Console pages, server logs when available, CMS export, and database or catalog records. Reconcile sets to find URLs that exist in one system but not another.
由于没有单一来源完整，应组合多个视图：站点抓取、站点地图清单、分析着陆页、Search Console 页面、可用时的服务器日志、CMS 导出，以及数据库或目录记录。对账这些集合，找出只存在于某个系统的 URL。
Audit search performance with cohorts, not isolated averages用分组而不是孤立平均值审计搜索表现
Analyze full periods and compare query-page cohorts by brand status, intent, topic, template, market, device, country, search type, and business priority. Separate changes in demand, impressions, ranking distribution, CTR, landing-page ownership, and outcomes. A falling average position can coexist with more qualified clicks if the query mix changed.
分析完整周期，并按品牌属性、意图、主题、模板、市场、设备、国家、搜索类型与业务优先级比较查询—页面分组。分别观察需求、展示、排名分布、CTR、着陆页归属与结果变化。如果查询组合发生变化，平均位置下降也可能与更高质量点击同时出现。
Google explains that the Search Console Performance report exposes search traffic dimensions and that data can differ by aggregation and filtering. Record property type, dates, search type, filters, dimensions, data state, and last complete date in every finding.
Google 说明 Search Console 效果报告提供搜索流量维度，数据会因聚合与筛选而不同。每个发现都应记录资源类型、日期、搜索类型、筛选、维度、数据状态与最后完整日期。
Audit whether each page earns a distinct search purpose审计每个页面是否承担独特搜索目的
For priority pages, evaluate target audience, question, intent, evidence, originality, completeness, freshness, authorship, internal-link context, external references, conversion path, and maintenance owner. Compare overlapping pages before creating new content. Consolidation, redirection, repositioning, or deletion may be stronger than expansion.
对重点页面评估目标受众、问题、意图、证据、原创性、完整性、新鲜度、作者、内部链接语境、外部引用、转化路径与维护负责人。创建新内容前先比较重叠页面；合并、重定向、重新定位或删除可能比扩写更有效。
| Decision决策 | Use when适用情况 | Validation验证 |
|---|---|---|
| Keep保留 | Distinct purpose and healthy evidence目的独特且证据健康 | Monitor intent and outcomes监控意图与结果 |
| Update更新 | Useful page with fixable gaps有价值但存在可修复缺口 | Annotate release and compare cohort标注发布并比较分组 |
| Merge合并 | Overlapping purpose splits signals重叠目的分散信号 | Check redirects, links, ownership检查重定向、链接与归属 |
| Remove移除 | No purpose, demand, value, or obligation无目的、需求、价值或义务 | Confirm dependencies and status确认依赖与状态 |
Audit the experience search visitors actually receive审计搜索访客实际获得的体验
Test representative templates on real mobile and desktop viewports. Review navigation, reading order, accessibility, intrusive elements, interaction stability, forms, error handling, media, structured data, metadata, language switching, and critical rendering. Use field performance where available and lab tests for reproducible diagnostics.
在真实移动端与桌面端视口测试代表性模板。检查导航、阅读顺序、可访问性、干扰元素、交互稳定性、表单、错误处理、媒体、结构化数据、元数据、语言切换与关键渲染。可用时使用现场性能数据，并用实验室测试获得可复现诊断。
Google's Core Web Vitals documentation currently centers loading, interaction responsiveness, and visual stability through LCP, INP, and CLS. Treat thresholds as experience signals, not a substitute for observing real task completion.
Google 的 Core Web Vitals 文档目前通过 LCP、INP 与 CLS 关注加载、交互响应与视觉稳定性。应把阈值视为体验信号，而不是观察真实任务完成情况的替代品。
Audit the measurement system before trusting trends信任趋势前先审计衡量系统
Verify Search Console property coverage, analytics configuration, channel rules, event and conversion definitions, consent effects, timezones, filters, bot handling, cross-domain behavior, data retention, annotations, dashboard calculations, API extraction, warehouse transformations, and access ownership. A measurement break can imitate an SEO decline.
验证 Search Console 资源覆盖、分析配置、渠道规则、事件与转化定义、同意机制影响、时区、筛选、机器人处理、跨域行为、数据保留、注释、仪表板计算、API 提取、仓库转换与访问所有权。衡量故障可能伪装成 SEO 下滑。
Reconcile a documented sample across interfaces and exports. Preserve missing, partial, delayed, thresholded, or backfilled states. Do not silently convert unknown values to zero or mix source definitions under an unlabeled KPI.
在界面与导出之间对账有记录的样本。保留缺失、部分、延迟、阈值处理或回填状态。不要把未知值静默变成零，也不要把不同来源定义混入无标签 KPI。
Prioritize findings by evidence, reach, value, and effort按证据、覆盖、价值与工作量排序发现
Severity alone is not a roadmap. Score each finding by confidence, affected URL or query cohort, business value, expected direction, effort, dependencies, reversibility, owner capacity, and validation speed. Mark regulatory, security, or migration blockers as hard gates rather than letting them disappear inside an average score.
严重程度本身不是路线图。每个发现应按置信度、受影响 URL 或查询分组、业务价值、预期方向、工作量、依赖、可逆性、负责人能力与验证速度评分。法规、安全或迁移阻断项应设为硬门槛，而不是消失在平均分中。
Plan immediately and define acceptance evidence.
立即规划并定义验收证据。
Run a controlled test or collect another source.
执行受控测试或补充另一来源。
Batch with related maintenance where efficient.
适合时与相关维护任务批量处理。
Monitor or defer; do not inflate the backlog.
监控或推迟，不要膨胀待办列表。
A useful audit delivers decisions, not just findings有效审计交付决策，而不只是发现
The final package should include an executive summary, scope and limitations, evidence inventory, baseline, issue register, affected cohorts, screenshots or queries, impact hypotheses, confidence, priority, owner, dependency, recommended action, acceptance criteria, release plan, and monitoring schedule. Separate observed facts from interpretations and recommendations.
最终交付应包含执行摘要、范围与限制、证据清单、基线、问题台账、受影响分组、截图或查询、影响假设、置信度、优先级、负责人、依赖、建议行动、验收标准、发布计划与监控节奏。应把观察事实、解释与建议分开。
Example: turn an audit into a 90-day plan示例：把审计转化为 90 天计划
- 1Days 1–15 · Stop loss and measurement gaps.第 1–15 天 · 阻止损失并修复衡量缺口。
Resolve blockers, broken templates, accidental directives, tracking gaps, and migration defects.
解决阻断项、损坏模板、意外指令、跟踪缺口与迁移缺陷。
- 2Days 16–45 · Repair high-reach systems.第 16–45 天 · 修复高覆盖系统。
Improve internal links, canonical rules, sitemaps, templates, rendering, and priority content ownership.
改进内部链接、规范化规则、站点地图、模板、渲染与重点内容归属。
- 3Days 46–75 · Consolidate and expand evidence-backed content.第 46–75 天 · 合并并扩展有证据支持的内容。
Merge overlap, refresh declining pages, fill validated intent gaps, and strengthen contextual links.
合并重叠、更新下滑页面、填补已验证意图缺口并加强语境链接。
- 4Days 76–90 · Validate, document, and operationalize.第 76–90 天 · 验证、记录并运营化。
Re-crawl, inspect source data, compare cohorts, close accepted work, and move controls into monitoring.
重新抓取、检查来源数据、比较分组、关闭已验收工作，并把控制项转入监控。
Use multiple tools because each reveals different evidence使用多种工具，因为每种工具揭示不同证据
| Evidence证据 | Typical source典型来源 | What it cannot prove alone单独无法证明 |
|---|---|---|
| Google Search behaviorGoogle 搜索行为 | Search Console | Full crawl state or business impact完整抓取状态或业务影响 |
| On-site outcomes站内结果 | Analytics / warehouse | Search demand or index eligibility搜索需求或索引资格 |
| URL observationsURL 观测 | Crawler and browser tests爬虫与浏览器测试 | How Google indexed every URLGoogle 如何索引每个 URL |
| Server requests服务器请求 | Log files日志文件 | Rendered content or query value渲染内容或查询价值 |
| Content inventory内容清单 | CMS / database | Discoverability or search performance可发现性或搜索表现 |
Analyze Search Console evidence with InfiniSynapse使用 InfiniSynapse 分析 Search Console 证据
Use the InfiniSynapse GSC analysis tool to explore query-page cohorts, clicks, impressions, CTR, and average position. Compare complete periods, identify high-impression opportunities and ownership shifts, and export evidence into the audit issue register. Treat the tool as one evidence layer alongside crawl, content, experience, authority, and outcome data.
使用 InfiniSynapse GSC 分析工具探索查询—页面分组、点击、展示、CTR 与平均位置。比较完整周期，识别高展示机会与归属变化，并把证据导出到审计问题台账。应把该工具视为证据层之一，并与抓取、内容、体验、权威性与结果数据结合。
Analyze GSC audit evidence分析 GSC 审计证据How often should an SEO audit be performed?SEO 审计应该多久做一次？
Do not wait for an annual document to detect critical failures. Monitor uptime, directives, templates, tracking, index coverage, and major performance shifts continuously or on a risk-based schedule. Run focused audits after significant releases and a broader audit at a cadence matched to site size, change frequency, market complexity, and business exposure.
不要等年度文档才发现关键故障。应持续或按风险节奏监控可用性、指令、模板、跟踪、索引覆盖与重大表现波动。在重要发布后执行专项审计，并按站点规模、变更频率、市场复杂度与业务暴露安排更全面审计。
A stable brochure site may need fewer broad reviews than a large ecommerce, marketplace, publisher, or multilingual platform shipping daily. The trigger should be risk and change, not a calendar alone.
稳定的展示型网站可能比每日发布的大型电商、市场、出版或多语言平台需要更少全面复盘。触发条件应是风险与变化，而不只是日历。
Complete SEO audit checklist完整 SEO 审计清单
- ✓Business objective, scope, exclusions, expected behavior, stakeholders, and deadline are documented.业务目标、范围、排除项、预期行为、利益相关者与截止日期均已记录。
- ✓URL inventories from crawl, sitemap, analytics, Search Console, CMS, and logs are reconciled.抓取、站点地图、分析、Search Console、CMS 与日志 URL 清单已对账。
- ✓Crawl, rendering, directives, canonicals, redirects, sitemaps, internal links, and index state are reviewed.抓取、渲染、指令、规范化、重定向、站点地图、内部链接与索引状态已审查。
- ✓Search performance is segmented by complete periods, cohorts, demand, rank, CTR, ownership, and outcome.搜索表现已按完整周期、分组、需求、排名、CTR、归属与结果拆分。
- ✓Content has a distinct audience, intent, evidence, owner, maintenance plan, and conversion path.内容具有独特受众、意图、证据、负责人、维护计划与转化路径。
- ✓Mobile experience, accessibility, performance, structured data, metadata, and interaction are tested.移动体验、可访问性、性能、结构化数据、元数据与交互均已测试。
- ✓Internal authority, external references, backlink patterns, reputation, and trust context are reviewed.内部权威、外部引用、外链模式、声誉与信任语境均已审查。
- ✓Measurement definitions, freshness, gaps, transformations, access, and annotations are validated.衡量定义、新鲜度、缺口、转换、访问与注释均已验证。
- ✓Every priority item has evidence, affected cohort, confidence, owner, dependency, action, and acceptance criteria.每个优先项都有证据、受影响分组、置信度、负责人、依赖、行动与验收标准。
- ✓Re-crawl, source checks, cohort comparisons, release annotations, and monitoring are scheduled.重新抓取、来源检查、分组对比、发布注释与监控已安排。
SEO audit FAQSEO 审计常见问题
A focused audit may take several days; a large multilingual or ecommerce site can require multiple weeks. Scope, data access, crawl size, history, and validation depth determine duration.
专项审计可能需要数天；大型多语言或电商网站可能需要数周。范围、数据访问、抓取规模、历史与验证深度决定时间。
No. Technical discovery and indexation are one layer. A complete audit also examines search performance, content and intent, experience, authority, measurement quality, prioritization, ownership, and validation.
不是。技术发现与索引只是一个层。完整审计还检查搜索表现、内容与意图、体验、权威性、衡量质量、优先级、所有权与验证。
Automation can collect observations and repeat checks, but human judgment is needed for expected behavior, business context, competing evidence, causality, priority, feasibility, and acceptance.
自动化可以收集观测并重复检查，但预期行为、业务背景、冲突证据、因果、优先级、可行性与验收仍需要人工判断。
Convert accepted findings into owned work, define release and rollback plans, annotate changes, validate with the same evidence, monitor cohorts, and close items only when acceptance criteria pass.
把已接受发现转成有负责人的工作，定义发布与回滚计划，标注变更，用相同证据验证，监控分组，并只在验收标准通过后关闭事项。
Official references for audit evidence审计证据的官方资料
Use Google's current SEO Starter Guide, Search Essentials, Search Console Performance report documentation, and Web Vitals guidance to verify definitions and current recommendations. Official guidance is a baseline; your audit must still test the site's intended behavior and real evidence.
请使用 Google 现行的 SEO 入门指南、Google 搜索基本要素、Search Console 效果报告文档与 Web Vitals 指南核实定义与当前建议。官方指南是基线，审计仍必须测试网站的预期行为与真实证据。
A practical what is seo audit evidence worksheet可直接使用的“what is seo audit”证据工作表
Use this analysis sheet to turn what is seo audit data into a repeatable SEO decision. Preserve the exact GSC filters, comparison period, limitations, threshold, action owner, and recheck date.
使用这份分析表，把“what is seo audit”数据转化为可复现的 SEO 决定。保留完整 GSC 筛选条件、比较周期、局限、阈值、行动负责人和复查日期。
State the decision question and save the property, search type, country, device, query, page, and date filters.
写明决策问题，并保存资源、搜索类型、国家、设备、查询、页面和日期筛选。
Use a comparable prior period or cohort; record seasonality, reporting delay, anonymized queries, and export limits.
使用可比较的历史周期或队列，并记录季节性、报告延迟、匿名查询和导出限制。
Define the minimum material change before looking at results, then test whether another explanation fits the pattern.
在查看结果前定义最小实质变化，再检验是否存在其他合理解释。
Assign one reversible action, owner, expected signal, guardrail, and recheck date; preserve the unchanged comparison group.
指定一项可逆行动、负责人、预期信号、保护指标和复查日期，并保留未改动对照组。
The InfiniSynapse Editorial Team structured this what is seo audit guide around a real decision workflow, reviewed the primary or official sources listed on the page, separated evidence from assumptions, and documented limitations before the final editorial check. The worksheet is an original reusable aid, not a benchmark, certification, professional opinion, or guaranteed outcome.
InfiniSynapse 编辑团队围绕真实决策流程组织本篇“what is seo audit”指南，复核页面列出的一手或官方来源，区分证据与假设，并在最终编辑检查前记录局限。这份工作表是原创可复用辅助材料，不是基准、认证、专业意见或结果保证。
Continue with related InfiniSynapse guides:
继续阅读相关 InfiniSynapse 指南：