[{"data":1,"prerenderedAt":621},["ShallowReactive",2],{"blog-cross-border-software-seo":3,"blog-prev-cross-border-software-seo":425,"blog-next-cross-border-software-seo":515},{"id":4,"title":5,"author":6,"body":7,"canonical":408,"category":409,"cover":408,"date":410,"description":411,"draft":412,"extension":413,"keywords":408,"meta":414,"navigation":415,"noindex":412,"ogImage":408,"path":416,"seo":417,"seoDescription":408,"seoTitle":408,"stem":418,"tags":419,"updated":408,"__hash__":424},"blog\u002Fblog\u002Fcross-border-software-seo.md","跨境电商软件的 SEO:出海和国内双市场怎么打","元析智能",{"type":8,"value":9,"toc":396},"minimark",[10,14,19,22,25,28,31,63,66,69,75,86,91,105,108,112,115,193,196,222,225,228,234,240,243,261,264,267,270,275,289,294,311,317,320,323,326,332,338,344,350,353,356,387,390,393],[11,12,13],"p",{},"跨境电商软件(ERP、listing 工具、广告管理工具)的客户群体特殊:既要做国内市场(服务跨境卖家),又要做海外市场(直接卖海外卖家)。SEO 要双线作战,语言、搜索引擎、外链策略都不一样。这篇文章写我们做跨境软件 SEO 的经验,包括踩过的坑。",[15,16,18],"h2",{"id":17},"双语-seo-的难点","双语 SEO 的难点",[11,20,21],{},"先说最难的部分:不是翻译,是关键词对不上。",[11,23,24],{},"中文\"亚马逊 listing 优化工具\",英文对应词可能是\"Amazon listing optimization tool\",也可能是\"Amazon SEO tool\",也可能是\"listing optimizer\"。这三个词在 Google 上的搜索量差 3 倍,意图也有差别。",[11,26,27],{},"我们做过的踩坑:给一个 listing 工具客户做英文站,直接翻译中文关键词,3 个月流量起不来。后来用 Ahrefs 重新挖词,发现真正的高搜索量词是\"Amazon product description generator\"和\"listing builder\",我们一开始没用。",[11,29,30],{},"双语 SEO 的关键:",[32,33,34,42,57],"ul",{},[35,36,37,41],"li",{},[38,39,40],"strong",{},"不要翻译关键词,要重新调研",":中英文用户搜索习惯完全不同",[35,43,44,47,48,52,53,56],{},[38,45,46],{},"不同语言独立站",":不要一个站多语言切换,子目录(",[49,50,51],"code",{},"\u002Fen\u002F","、",[49,54,55],{},"\u002Fzh\u002F",")或子域名(ccTLD)都行",[35,58,59,62],{},[38,60,61],{},"hreflang 标签必须正确",":错的 hreflang 比 没有 hreflang 还糟,会让 Google 把英文页当中文页索引",[15,64,65],{"id":65},"关键词调研工具的差异",[11,67,68],{},"国内市场和海外市场的关键词工具完全是两套:",[11,70,71,74],{},[38,72,73],{},"国内市场",":",[32,76,77,80,83],{},[35,78,79],{},"5118:百度关键词,中文最准",[35,81,82],{},"百度关键词规划师:官方数据,但搜索量范围",[35,84,85],{},"站长工具:查中文词还行",[11,87,88,74],{},[38,89,90],{},"海外市场",[32,92,93,96,99,102],{},[35,94,95],{},"Ahrefs:英文词最准,数据全",[35,97,98],{},"SEMrush:欧洲小语种覆盖好",[35,100,101],{},"Google Keyword Planner:免费但数据粗",[35,103,104],{},"Ubersuggest:入门级,数据不够细",[11,106,107],{},"我们双线做时,英文用 Ahrefs,中文用 5118,两边数据对不上是常态。比如同功能的工具,中文搜索量 300\u002F月,英文可能 3000\u002F月。预算分配要跟着搜索量走,不能平均分。",[15,109,111],{"id":110},"google-和百度算法差异对内容策略的影响","Google 和百度算法差异对内容策略的影响",[11,113,114],{},"Google 和百度的算法差异,直接影响内容怎么写:",[116,117,118,134],"table",{},[119,120,121],"thead",{},[122,123,124,128,131],"tr",{},[125,126,127],"th",{},"维度",[125,129,130],{},"Google",[125,132,133],{},"百度",[135,136,137,149,160,171,182],"tbody",{},[122,138,139,143,146],{},[140,141,142],"td",{},"外链权重",[140,144,145],{},"高",[140,147,148],{},"2025 年后明显降低",[122,150,151,154,157],{},[140,152,153],{},"内容长度偏好",[140,155,156],{},"长内容(2000+ 字)排名好",[140,158,159],{},"中等长度(800-1500 字)够用",[122,161,162,165,168],{},[140,163,164],{},"AI 内容态度",[140,166,167],{},"看质量,不一刀切",[140,169,170],{},"严格识别,批量降权",[122,172,173,176,179],{},[140,174,175],{},"新站沙盒期",[140,177,178],{},"6-9 个月",[140,180,181],{},"3-6 个月",[122,183,184,187,190],{},[140,185,186],{},"E-E-A-T",[140,188,189],{},"权威度信号重要",[140,191,192],{},"看重备案、ICP",[11,194,195],{},"实操建议:",[32,197,198,204,210,216],{},[35,199,200,203],{},[38,201,202],{},"英文站写长内容",":2000-3000 字,深入展开,Google 偏好",[35,205,206,209],{},[38,207,208],{},"中文站写短内容",":800-1500 字,直击重点,百度偏好",[35,211,212,215],{},[38,213,214],{},"同一主题写两版",":不要翻译,要按各自的算法偏好重写",[35,217,218,221],{},[38,219,220],{},"作者署名和权威度",":英文站重视 E-E-A-T,作者要有 LinkedIn、GitHub 等可信背书;中文站备案和 ICP 是基础",[15,223,224],{"id":224},"竞品分析怎么做",[11,226,227],{},"跨境软件的竞品分析比纯国内或纯海外复杂,要分两层:",[11,229,230,233],{},[38,231,232],{},"国内市场竞品",":用 5118 查竞品的关键词、流量、外链。看竞品在百度上的排名词,找空位。",[11,235,236,239],{},[38,237,238],{},"海外市场竞品",":用 Ahrefs 或 SEMrush 查竞品的 Google 排名、外链来源、内容策略。",[11,241,242],{},"我们的做法:",[244,245,246,249,252,255,258],"ol",{},[35,247,248],{},"列出国内外各 5-10 个直接竞品",[35,250,251],{},"拉竞品的 organic keywords(自然排名词)",[35,253,254],{},"找出竞品有排名你却没有的词(content gap)",[35,256,257],{},"分析竞品内容类型,看哪些类型他们做你没做",[35,259,260],{},"分析竞品外链来源,看能不能找同类渠道(不是抄,是找同类)",[11,262,263],{},"跨境软件的竞品分析有个特殊性:国内竞品和海外竞品经常不是同一批。国内做得好的(比如店小秘),海外不一定强;海外强的(Helium 10),国内不一定有。这意味着双线都有空位,关键是判断哪边机会大。",[15,265,266],{"id":266},"外链建设在出海场景的特殊性",[11,268,269],{},"跨境软件的外链建设是国内和海外两套玩法,难度差很多:",[11,271,272,74],{},[38,273,274],{},"国内市场外链",[32,276,277,280,283,286],{},[35,278,279],{},"2025 年后买卖外链基本失效",[35,281,282],{},"主要靠行业媒体(亿邦动力、雨果网)发软文",[35,284,285],{},"知乎、小红书做内容,带官网链接",[35,287,288],{},"ERP 工具可以做行业榜单收录",[11,290,291,74],{},[38,292,293],{},"海外市场外链",[32,295,296,299,302,305,308],{},[35,297,298],{},"买卖外链 Google 还能识别一些,但风险高",[35,300,301],{},"主要靠 guest post(行业博客投稿)",[35,303,304],{},"Reddit、HackerNews、Product Hunt 发起讨论",[35,306,307],{},"行业媒体(EcomCrew、Practical Ecommerce)发深度内容",[35,309,310],{},"工具类网站(G2、Capterra)做产品收录",[11,312,313,316],{},[38,314,315],{},"踩坑经验",":我们给一个跨境 ERP 客户做外链,一开始想用国内的\"行业媒体发稿\"思路打海外,在几个跨境电商博客买了 guest post。3 个月后查外链,Google 把其中 2 个判定为 paid link,权重清零。后来改成做 Product Hunt 发布 + Reddit 真实分享,效果反而好。",[11,318,319],{},"跨境软件出海的外链建设,不要图快。Google 对 paid link 的识别越来越强,真实的内容和外链才是长期资产。",[15,321,322],{"id":322},"具体踩坑经验",[11,324,325],{},"总结几个我们踩过的坑:",[11,327,328,331],{},[38,329,330],{},"坑 1:双语站用翻译插件","\n客户图省事,用 WPML 自动翻译英文站。3 个月后流量起不来。原因:翻译质量太差,关键词不对。改回人工重写后,流量 2 个月涨 4 倍。",[11,333,334,337],{},[38,335,336],{},"坑 2:中文策略直接搬到英文","\n中文站写\"亚马逊运营技巧\"这类宽泛词做内容。搬到英文站做\"Amazon selling tips\",竞争激烈排名起不来。改成做\"how to optimize Amazon listing for A9 algorithm\"这种长尾,1 个月进前 10。",[11,339,340,343],{},[38,341,342],{},"坑 3:海外外链买 PBN","\n图便宜买了 PBN(私人博客网络)外链。2 个月后 Google 算法更新,整个站排名掉了 30 名以上。清理掉 PBN 外链、提交 reconsideration,3 个月才恢复。",[11,345,346,349],{},[38,347,348],{},"坑 4:忽视 hreflang","\n一个客户英文站和中文站内容相似度高,Google 把英文页当中文页的副本索引,中文站排名被压。修正 hreflang 后,中文站 1 个月恢复。",[15,351,352],{"id":352},"我们不接的跨境软件",[11,354,355],{},"不是所有跨境软件都适合做 SEO。我们不接:",[32,357,358,364,370,379],{},[35,359,360,363],{},[38,361,362],{},"国内站没建好的",":先做好一边再做另一边",[35,365,366,369],{},[38,367,368],{},"英文站纯翻译的",":不愿人工重写,做不出效果",[35,371,372,375,378],{},[38,373,374],{},"预算不够做双线的",[376,377],"seo",{}," 预算砍半,双线都做不起来,不如集中一边",[35,380,381,384,386],{},[38,382,383],{},"产品没 PMF 的",[376,385],{}," 之前先解决产品问题",[15,388,389],{"id":389},"总结",[11,391,392],{},"跨境软件 SEO 的核心:双线独立做,不要套模板。国内用 5118 挖词、百度偏好短内容、外链靠行业媒体;海外用 Ahrefs 挖词、Google 偏好长内容、外链靠真实 guest post 和社区。踩过的坑:hreflang、PBN、翻译插件、关键词直译。",[11,394,395],{},"如果你做跨境软件,想知道你的双线 SEO 现状,联系我们做个诊断。",{"title":397,"searchDepth":398,"depth":398,"links":399},"",2,[400,401,402,403,404,405,406,407],{"id":17,"depth":398,"text":18},{"id":65,"depth":398,"text":65},{"id":110,"depth":398,"text":111},{"id":224,"depth":398,"text":224},{"id":266,"depth":398,"text":266},{"id":322,"depth":398,"text":322},{"id":352,"depth":398,"text":352},{"id":389,"depth":398,"text":389},null,"SEO","2025-12-22","跨境电商软件(ERP、listing、广告工具)要同时打国内和海外市场。双语 SEO、Google 和百度算法差异、外链建设,我们做跨境软件 SEO 的踩坑经验。",false,"md",{},true,"\u002Fblog\u002Fcross-border-software-seo",{"title":5,"description":411},"blog\u002Fcross-border-software-seo",[420,421,422,423],"跨境电商SEO","出海","双语SEO","软件出海","7ayDI_UFLfbHcY4_m5_xBtF3j8A_U3DLrZQK8DBWWXg",{"id":426,"title":427,"author":6,"body":428,"canonical":408,"category":503,"cover":408,"date":504,"description":505,"draft":412,"extension":413,"keywords":408,"meta":506,"navigation":415,"noindex":412,"ogImage":408,"path":507,"seo":508,"seoDescription":408,"seoTitle":408,"stem":509,"tags":510,"updated":408,"__hash__":514},"blog\u002Fblog\u002Fscreaming-frog-large-crawl-config.md","爬十万个页面，Screaming Frog 配置先改五处",{"type":8,"value":429,"toc":495},[430,433,437,440,443,447,450,453,457,460,463,467,470,473,477,480,483,486,489,492],[11,431,432],{},"2025 年 12 月我们爬一个文档站，12 万个 URL，第一次跑完导出结果一看，8 万条是带参数的重复页，真正要看的没几页。不是工具不行，是配置默认值适合小站，大站照默认跑就是浪费时间。",[15,434,436],{"id":435},"第一处渲染方式","第一处：渲染方式",[11,438,439],{},"默认不执行 JavaScript，抓的是源码。内容靠前端渲染的站，这么抓回来一片空白。开渲染之前先确认必要性，开了之后速度会慢好几倍，十几万页能跑到第二天。",[11,441,442],{},"我的做法是先关掉渲染快速跑一遍，看 HTML 里有没有正文。有就不开，没有才开，并且把渲染超时从默认的几秒调到 10 秒左右，否则异步加载的内容抓不全。",[15,444,446],{"id":445},"第二处抓取范围","第二处：抓取范围",[11,448,449],{},"用包含和排除规则把范围框住，比爬完再筛有效。要排除的东西很固定：会话 ID、排序参数、日历归档、筛选参数、文件下载、后台路径。",[11,451,452],{},"我们那 12 万 URL 里，光带排序参数的就有 5 万多。加一条正则排除之后，实际需要爬的只剩 3 万出头。",[15,454,456],{"id":455},"第三处限速和并发","第三处：限速和并发",[11,458,459],{},"默认并发对生产站有压力，尤其是共享服务器的小站。爬之前先跟运维确认能不能爬、什么时段爬。我们通常把并发降到 2，加上延迟，放在夜里跑。",[11,461,462],{},"曾经有一次没降速，把一个客户的电商站爬得响应变慢，第二天被运维投诉。从那以后，我们的规矩是先小样本跑 200 页，看服务器负载再放量。",[15,464,466],{"id":465},"第四处robots-与-ua","第四处：robots 与 UA",[11,468,469],{},"要不要遵守 robots，取决于目的。做诊断时遵守，能真实反映搜索引擎看到的东西；做覆盖度检查时可以临时不遵守，但要标记出来，别把结果混着看。",[11,471,472],{},"UA 也建议改一下，带上自己的标识，万一爬出问题对方能找到人。",[15,474,476],{"id":475},"第五处导出和存储","第五处：导出和存储",[11,478,479],{},"十万页以上，内存吃紧，结果文件也大。分批爬，按目录拆开，每批单独导出。导出时只留要看的列：状态码、标题、字数、内链数、canonical、索引设置。",[15,481,482],{"id":482},"爬完先看什么",[11,484,485],{},"结果出来别急着导全表，先看四组数：状态码分布、重复标题数、canonical 指向不一致的页数、内链深度大于 4 的页数。",[11,487,488],{},"第一次爬那个文档站，状态码分布里 200 占 78%，剩下的 22% 全是 301。301 集中在两个旧目录，说明改版时跳转没做干净。这种结论从全表里翻是翻不出来的。",[11,490,491],{},"还有个习惯，每次爬完把结果存一份带日期的 CSV，下次对比用。三个月后能看出抓取问题是在变好还是变坏，这比单次报告有用。",[11,493,494],{},"我现在的习惯是爬之前先写一句话：这次要回答什么问题。写不出来就别爬。工具跑得快不代表有意义，配置和目的不清，爬一百万页也只是生成一个巨大的 CSV。",{"title":397,"searchDepth":398,"depth":398,"links":496},[497,498,499,500,501,502],{"id":435,"depth":398,"text":436},{"id":445,"depth":398,"text":446},{"id":455,"depth":398,"text":456},{"id":465,"depth":398,"text":466},{"id":475,"depth":398,"text":476},{"id":482,"depth":398,"text":482},"技术SEO","2025-12-21","第一次爬一个 12 万 URL 的文档站，跑完发现 8 万条是重复参数页，等于白爬。这篇写大站爬取前必须改的五个配置，以及限速怎么设才不压垮生产环境。",{},"\u002Fblog\u002Fscreaming-frog-large-crawl-config",{"title":427,"description":505},"blog\u002Fscreaming-frog-large-crawl-config",[511,512,513],"Screaming Frog","爬虫配置","技术检测","DCwAhp0UzCMLPG83vMS3jN3KCdRE8kuHWlMloMd0GYM",{"id":516,"title":517,"author":6,"body":518,"canonical":408,"category":503,"cover":408,"date":610,"description":611,"draft":412,"extension":413,"keywords":408,"meta":612,"navigation":415,"noindex":412,"ogImage":408,"path":613,"seo":614,"seoDescription":408,"seoTitle":408,"stem":615,"tags":616,"updated":408,"__hash__":620},"blog\u002Fblog\u002Fnew-site-indexing-slow.md","新站三个月不收录，八成不是\"权重\"问题",{"type":8,"value":519,"toc":603},[520,523,526,564,568,571,578,581,584,587,590,594,597,600],[11,521,522],{},"新站确实有观察期，抓取频次低、发现慢，这是真的。但三个月还零收录，先当技术故障查，别用\"新站\"两个字解释一切。我们查过 11 个这种站，7 个是配置问题，2 个是域名历史问题，只有 2 个属于内容太薄、确实要等。",[15,524,525],{"id":525},"排查顺序",[244,527,528,534,540,546,552,558],{},[35,529,530,533],{},[38,531,532],{},"域名历史","：查 archive.org 有没有前身站点，GSC 里有没有手动操作记录。买了个被用过的域名，前面积累的惩罚会带过来。",[35,535,536,539],{},[38,537,538],{},"robots 和 noindex","：新站模板从别的项目拷来的，带一行整站 noindex，这事我见过不下五次。",[35,541,542,545],{},[38,543,544],{},"渲染","：新站偏爱最新框架，初始 HTML 空空如也。curl 一下就知道。",[35,547,548,551],{},[38,549,550],{},"sitemap 有效性","：提交了但里面全是 302 或者需要登录，等于没提交。",[35,553,554,557],{},[38,555,556],{},"蜘蛛有没有来过","：GSC 抓取统计、服务器日志。一次请求都没有，说明发现环节就断了，去做几条真实外链，别等着。",[35,559,560,563],{},[38,561,562],{},"WAF 和防火墙","：新站默认安全策略经常把海外 IP 段限流，Googlebot 首当其冲。",[15,565,567],{"id":566},"_2025-年-12-月那个站","2025 年 12 月那个站",[11,569,570],{},"一个 API 监控工具的官网，2025 年 10 月上线，等了 10 周，GSC 里一个页面都没收录。他们团队已经接受了\"新站要等半年\"的说法。",[11,572,573,574,577],{},"我们打开首页源码，第 8 行是 ",[49,575,576],{},"\u003Cmeta name=\"robots\" content=\"noindex\">","——从上个项目模板拷过来没删。删掉后第 3 天收录首页，两周收录 210 个页面，第 5 周开始有长尾词排名。",[11,579,580],{},"这种情况没什么技术含量，但它是新站不收录里最高频的原因。查一次 URL 检查，十分钟。",[15,582,583],{"id":583},"新站提交的正确顺序",[11,585,586],{},"上线当天做四件事，顺序别反：先把 robots.txt 写好并确认没有全站屏蔽，再把 sitemap 提交到 GSC 和百度搜索资源平台，然后从几个已有站点放几条真实外链（官网导航、公众号文章、行业目录），最后用网址检查把首页和几个核心产品页确认一遍。十天内蜘蛛应该至少来过一次，没有就要按上面的清单回查。",[11,588,589],{},"内容上别一上线就铺 200 篇。新站抓取预算很小，一次放太多页面，蜘蛛会优先抓链接多的页面，深层的长尾页反而要等更久。我们一般建议首月 30 到 50 个页面，集中在产品页和 3 到 5 篇能打的文章上。",[15,591,593],{"id":592},"怎么判断是真的要等","怎么判断是\"真的要等\"",[11,595,596],{},"看日志。蜘蛛来过、抓了、没收录，这才是内容或质量层面的问题，需要时间。蜘蛛一次都没来，那是发现和配置问题，等三年也不会好。",[15,598,599],{"id":599},"立场",[11,601,602],{},"把\"新站没权重\"当诊断结论是偷懒。这句话只能解释排名慢，解释不了零收录。零收录永远是技术问题优先。",{"title":397,"searchDepth":398,"depth":398,"links":604},[605,606,607,608,609],{"id":525,"depth":398,"text":525},{"id":566,"depth":398,"text":567},{"id":583,"depth":398,"text":583},{"id":592,"depth":398,"text":593},{"id":599,"depth":398,"text":599},"2025-12-23","新站收录慢，最常听到的解释是\"新站没权重，等等就好\"。我们查过 11 个三个月以上不收录的新站，7 个是技术问题：robots、JS 渲染、模板 noindex、sitemap 无效。附排查顺序。",{},"\u002Fblog\u002Fnew-site-indexing-slow",{"title":517,"description":611},"blog\u002Fnew-site-indexing-slow",[617,618,619],"新站收录","索引诊断","抓取","IkGm7SUKQYGlkwM3tOtc13B_dxdlCk6x9wRt2r0uxsg",1794088109824]