[{"data":1,"prerenderedAt":347},["ShallowReactive",2],{"blog-contact-page-rebuild-review":3,"blog-prev-contact-page-rebuild-review":78,"blog-next-contact-page-rebuild-review":282},{"id":4,"title":5,"author":6,"body":7,"canonical":62,"category":63,"cover":62,"date":64,"description":65,"draft":66,"extension":67,"keywords":62,"meta":68,"navigation":69,"noindex":66,"ogImage":62,"path":70,"seo":71,"seoDescription":62,"seoTitle":62,"stem":72,"tags":73,"updated":62,"__hash__":77},"blog\u002Fblog\u002Fcontact-page-rebuild-review.md","联系页改版复盘：表单从九个字段砍到四个之后","元析智能",{"type":8,"value":9,"toc":55},"minimark",[10,14,19,22,25,28,31,34,37,40,43,46,49,52],[11,12,13],"p",{},"2025 年 12 月我们给一个做流程管理的客户改联系页。改之前表单 9 个字段，提交率 2.1%；改之后 4 个字段，提交率 5.4%。但同一时期无效询盘从 12% 涨到 31%，这个副作用比提升更值得说。",[15,16,18],"h2",{"id":17},"改之前九个字段三屏","改之前：九个字段，三屏",[11,20,21],{},"原来的字段是：姓名、手机、邮箱、公司名、公司地址、行业、员工人数、预算范围、需求描述。移动端要滑三屏，最后一屏是\"需求描述\"的大文本框。页面上没有电话，没有微信，没有真人。",[11,23,24],{},"我们看了热力图，70% 的用户在第一屏之后离开，其中大部分停在\"公司地址\"那个字段。一个 B2B 站的联系页问地址，是想做线下拜访，但线上转化阶段问这个太早了。",[15,26,27],{"id":27},"改了什么",[11,29,30],{},"保留四个字段：姓名、手机、公司名、需求描述。手机保留是因为国内 B2B 必须电话跟进，但要加一句\"仅用于本次咨询\"。",[11,32,33],{},"删掉的五个字段不是不重要，是挪到后面。销售第一次通话时问行业和预算，比让用户自己填准确得多。",[11,35,36],{},"另外加了三个东西：一张负责人的工作照和名字，一个可预约的日历入口，一个可直接拨打的电话。这三样都是\"看得见的人\"，对转化帮助明显。",[15,38,39],{"id":39},"三周后的数据和副作用",[11,41,42],{},"提交率从 2.1% 涨到 5.4%，绝对数量从每周 9 个涨到 22 个。但无效询盘从 12% 涨到 31%，主要是同行调研和只想问价格的学生。",[11,44,45],{},"处理办法是在表单里加一个必选的\"你是\"下拉：甲方企业、渠道商、学生或个人、其他。选后两项的直接进另一个标签，销售不跟。加完这一项，无效询盘降到 19%，提交率只掉了 0.3 个百分点。",[11,47,48],{},"我们还试过把电话放在表单上方，拨打量一周涨了 14 个，但接起来一半是推销。后来把电话改成\"预约演示\"入口，接听质量明显好一些，虽然数量少了一半。",[11,50,51],{},"表单优化本质是转化问题，不是设计问题。字段删掉容易，删完怎么分层才是关键。",[11,53,54],{},"下一步我们准备把预约日历做成默认入口，表单退到第二选择。原因很简单：填表单是\"我留个联系方式等你来\"，预约是\"我选个时间我们聊\"，后者的意向度高一截。这个改动要等数据跑够两个月再定，急不得。",{"title":56,"searchDepth":57,"depth":57,"links":58},"",2,[59,60,61],{"id":17,"depth":57,"text":18},{"id":27,"depth":57,"text":27},{"id":39,"depth":57,"text":39},null,"内容运营","2026-03-24","2025 年 12 月我们给一个客户改联系页，字段从 9 个砍到 4 个，加了真人头像和预约入口。三周后提交率从 2.1% 涨到 5.4%，无效询盘也涨到 31%，这篇讲怎么处理这个副作用。",false,"md",{},true,"\u002Fblog\u002Fcontact-page-rebuild-review",{"title":5,"description":65},"blog\u002Fcontact-page-rebuild-review",[74,75,76],"联系页","转化率","复盘","1C1IoQBS0NHXQurZqM2WzzzKcouVJBzLR_wktXI_Hdo",{"id":79,"title":80,"author":6,"body":81,"canonical":62,"category":269,"cover":62,"date":270,"description":271,"draft":66,"extension":67,"keywords":62,"meta":272,"navigation":69,"noindex":66,"ogImage":62,"path":273,"seo":274,"seoDescription":62,"seoTitle":62,"stem":275,"tags":276,"updated":62,"__hash__":281},"blog\u002Fblog\u002Fdomestic-ai-citation-sources.md","豆包、元宝、文心更爱引平台内容：一次 300 条引用的统计",{"type":8,"value":82,"toc":262},[83,86,89,92,95,195,198,201,208,214,220,223,226,229,232,235,238,241,244,247,250,253,256,259],[11,84,85],{},"国内 AI 搜索引用官网的比例，比海外低得多。2026 年 3 月到 4 月，我们统计了豆包、元宝、文心三个引擎的 300 条引用，官网占 21%，剩余 79% 全在平台上。",[15,87,88],{"id":88},"统计怎么做的",[11,90,91],{},"三个引擎各跑 100 条查询，覆盖 B2B 软件方向的品类词、对比词、场景词。记录每条回答挂出来的来源，按域名分类，同一条回答里同一域名只算一次。",[11,93,94],{},"300 条引用，去重后 217 个来源域名。",[96,97,98,114],"table",{},[99,100,101],"thead",{},[102,103,104,108,111],"tr",{},[105,106,107],"th",{},"来源类型",[105,109,110],{},"占比",[105,112,113],{},"典型站点",[115,116,117,129,140,151,162,173,184],"tbody",{},[102,118,119,123,126],{},[120,121,122],"td",{},"官网",[120,124,125],{},"21%",[120,127,128],{},"客户自己的站",[102,130,131,134,137],{},[120,132,133],{},"问答",[120,135,136],{},"18%",[120,138,139],{},"知乎",[102,141,142,145,148],{},[120,143,144],{},"技术社区",[120,146,147],{},"14%",[120,149,150],{},"CSDN、掘金",[102,152,153,156,159],{},[120,154,155],{},"公众号转载",[120,157,158],{},"12%",[120,160,161],{},"各类转载站",[102,163,164,167,170],{},[120,165,166],{},"门户与媒体",[120,168,169],{},"11%",[120,171,172],{},"36 氪、少数派",[102,174,175,178,181],{},[120,176,177],{},"百科",[120,179,180],{},"8%",[120,182,183],{},"百度百科",[102,185,186,189,192],{},[120,187,188],{},"其他",[120,190,191],{},"16%",[120,193,194],{},"论坛、文档站、GitHub",[15,196,197],{"id":197},"为什么官网占比低",[11,199,200],{},"三个原因。",[11,202,203,207],{},[204,205,206],"strong",{},"平台有内容池。"," 国内引擎跟内容平台有合作，抓取通道和权重都不一样。同样一篇文章，发在平台上比发在自己站上更容易被读到。",[11,209,210,213],{},[204,211,212],{},"中文语料密度。"," 一个 B2B 软件官网通常几十个页面，知乎上关于这个品类的讨论可能是几千条。引擎在语料里找答案，谁多找谁。",[11,215,216,219],{},[204,217,218],{},"官网写得不像答案。"," 我们看的 100 个官网里，有 70 多个的页面结构是\"产品价值 + 功能列表 + 客户 logo\"。这种页面没有能独立回答问题的段落，引擎抓了也没法用。",[15,221,222],{"id":222},"公众号那块要注意",[11,224,225],{},"公众号占 12%，但要看清楚：引用的多是转载到可抓取页面的版本，公众号原文本身在很多引擎里抓不到。",[11,227,228],{},"这意味着两件事。一是发公众号的时候，同步发一份到官网或知乎，让内容有可抓的入口。二是看数据时别把\"公众号发了\"当成\"被收录了\"。",[11,230,231],{},"我们有个客户每月发 8 篇公众号，发了两年，国内 AI 回答里一次没出现过。查下来是内容从没在可抓页面出现过。",[15,233,234],{"id":234},"对软件公司的含义",[11,236,237],{},"官网是事实源，平台是入口。这个分工要想清楚。",[11,239,240],{},"官网负责\"准确\"：产品参数、定价、部署方式、公司信息，写清楚，让引擎有据可查。",[11,242,243],{},"平台负责\"被读到\"：知乎回答、技术社区文章、行业媒体稿，这些决定引擎会不会在回答里提你。",[11,245,246],{},"只做官网的客户，我们说句实话：国内这三个引擎里，你的可见度会一直很低，跟你内容质量没关系。",[15,248,249],{"id":249},"一个可执行的分配",[11,251,252],{},"按我们的经验，国内 GEO 的力气这么分：",[11,254,255],{},"官网占 40%，做事实和结构。平台占 50%，做分发和场景。剩下 10% 做监测。",[11,257,258],{},"不是五五开，是官网四、平台五。因为平台内容还承担了\"被别人引用\"的功能，一篇知乎回答可能同时被三个引擎读。",[11,260,261],{},"这个比例在海外反过来，官网大概占六。做国内和做出海，不能用同一套排期。",{"title":56,"searchDepth":57,"depth":57,"links":263},[264,265,266,267,268],{"id":88,"depth":57,"text":88},{"id":197,"depth":57,"text":197},{"id":222,"depth":57,"text":222},{"id":234,"depth":57,"text":234},{"id":249,"depth":57,"text":249},"行业观察","2026-03-23","统计国内三个引擎 300 条引用，官网只占 21%，知乎 18%、CSDN 14%、公众号 12%。国内做 GEO，官网是事实源，平台才是入口。",{},"\u002Fblog\u002Fdomestic-ai-citation-sources",{"title":80,"description":271},"blog\u002Fdomestic-ai-citation-sources",[277,278,279,280],"国内AI搜索","豆包","元宝","平台内容","4xW_Ff2rK4r-EDskmVvW_gI45YGugT-ojM9hnwmABQ0",{"id":283,"title":284,"author":6,"body":285,"canonical":62,"category":335,"cover":62,"date":336,"description":337,"draft":66,"extension":67,"keywords":62,"meta":338,"navigation":69,"noindex":66,"ogImage":62,"path":339,"seo":340,"seoDescription":62,"seoTitle":62,"stem":341,"tags":342,"updated":62,"__hash__":346},"blog\u002Fblog\u002Fdesign-system-docs-seo.md","设计系统文档公开，能带来开发者流量",{"type":8,"value":286,"toc":329},[287,290,293,296,299,302,305,308,311,314,317,320,323,326],[11,288,289],{},"设计系统文档该不该公开，很多团队在纠结，怕被抄。我的判断是公开，而且这是设计工具最便宜的一批自然流量。2025 年 12 月我们建议一个设计系统工具把组件文档从内网搬到公开域名下，四个月自然访问从每月 900 到 6900。",[15,291,292],{"id":292},"谁会搜组件文档",[11,294,295],{},"三类人：用这套系统的产品团队、想搭自己设计系统的公司、以及写代码时忘了某个组件怎么用的工程师。第三类最值钱，因为他每天都在用。",[11,297,298],{},"搜的词很具体：“按钮组件 禁用态 规范”“表单校验 提示文案 长度”“空状态 插画 尺寸”。这些词单月搜索量几十到几百，但页面数量可以很多，聚起来就是流量。",[15,300,301],{"id":301},"哪几类页面该公开",[11,303,304],{},"组件页、模式页、内容规范页，这三类公开价值最高。组件页写 API、状态、用法；模式页写“什么时候用抽屉什么时候用弹窗”这种判断；内容规范页写文案长度、语气、错误提示写法。",[11,306,307],{},"设计 token 页最难做好。它天然是表格，一屏几百行，搜索引擎能读但用户看不下去。我们给一个客户的做法是按用途拆成 6 页：颜色、间距、字号、圆角、阴影、动效，每页配一段为什么这么定的说明。拆完这 6 页的月访问加起来 1800。",[15,309,310],{"id":310},"一个反常识的发现",[11,312,313],{},"组件文档页的转化路径不是注册，是回访。用户第一次来查一个组件，不会注册。但一个月来五次之后，他会去看这个系统背后的产品是什么。所以我们在这类页面底部放的不是注册按钮，是“这套系统的设计原则”和“背后的工具”。",[11,315,316],{},"我们追踪过一个客户的数据：从组件文档进入、30 天内产生注册的用户，平均访问了 6.2 个文档页。普通博客读者是 1.4 个。",[11,318,319],{},"关键词怎么定也有讲究。组件文档不需要打“设计系统”这种大词，要打组件名加状态名。我们给一个客户的按钮页标题写成“按钮组件：尺寸、状态与禁用规范”，比原来的“Button 组件”多覆盖了三个搜索词。",[15,321,322],{"id":322},"别做的事",[11,324,325],{},"别把文档放在需要登录的地方，那等于没有。别用截图代替可复制的代码，设计师和工程师都要复制。别在文档里放产品广告位，破坏阅读节奏，代价比收益大。",[11,327,328],{},"公开文档的风险是竞品参考，但设计系统本身不是壁垒，执行速度才是。这个账我们算过很多次，结论一致。",{"title":56,"searchDepth":57,"depth":57,"links":330},[331,332,333,334],{"id":292,"depth":57,"text":292},{"id":301,"depth":57,"text":301},{"id":310,"depth":57,"text":310},{"id":322,"depth":57,"text":322},"技术SEO","2026-03-25","把组件库文档从内网搬到公开域名后，一个设计系统工具每月多 6000 次自然访问。哪些页面值得公开、怎么组织，以及 token 页为什么最难做好。",{},"\u002Fblog\u002Fdesign-system-docs-seo",{"title":284,"description":337},"blog\u002Fdesign-system-docs-seo",[343,344,345],"设计系统","组件库","文档SEO","BSgXwt3LLTUZi1oI2vD_ECOsE9H9y8grGaOG9xSjFlQ",1794088107352]