最全总结，数据分析的标准流程，保藏！
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">数据分析是业务决策的重要工具，但<span style="color: black;">非常多</span>数据分析师在<span style="color: black;">实质</span>操作中却常常感到困惑，不<span style="color: black;">晓得</span><span style="color: black;">怎样</span>将数据转化为有价值的业务洞察。本文将系统性地介绍数据分析的标准流程，<span style="color: black;">帮忙</span>数据分析师理清思路，更有效地支持业务决策。</p><img src="https://p3-sign.toutiaoimg.com/tos-cn-i-axegupay5k/a897734812a34f7dba29af5607ae8817~noop.image?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1725685638&x-signature=ppweK%2F4coLSTyWG5Dod1sRetdMU%3D" style="width: 50%; margin-bottom: 20px;">
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">“数据分析的标准流程是啥？”</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">“为啥感觉自己只取了数，看不到结果？”</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">“分析做到什么地步，才算是对业务有用？”</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">经常有人有这些疑惑，今天<span style="color: black;">咱们</span>来系统讲解下，数据分析的标准流程，帮<span style="color: black;">大众</span>理清思路。做数据分析其实有2种基本流程和6种<span style="color: black;">详细</span><span style="color: black;">状况</span>，<span style="color: black;">咱们</span>一个个讲。</p>
<h1 style="color: black; text-align: left; margin-bottom: 10px;"><span style="color: black;">1、</span>从问题出发的流程</h1>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">此类流程，典型有三步：问题→数据→答案，<span style="color: black;">通常</span>业务<span style="color: black;">分部</span>会沿着这个路径思考。<span style="color: black;">例如</span>你是一个<span style="color: black;">营销</span>主管，你最关心的肯定是自己业绩表现，于是你会先看：</p>问题1：我的团队业绩<span style="color: black;">怎样</span>？数据1：当月达标<span style="color: black;">状况</span>……今年累计达标<span style="color: black;">状况</span>……答案1：<span style="color: black;">日前</span>已达标，超额20%<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">当然，有可能你思考得更深入<span style="color: black;">有些</span>。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">例如</span><span style="color: black;">机构</span><span style="color: black;">日前</span>有一个业绩排名奖：全国每月同比增长排名前5的团队，<span style="color: black;">能够</span>拿一笔奖金。<span style="color: black;">此刻</span><span style="color: black;">已然</span>20号了，你很想<span style="color: black;">晓得</span>自己有<span style="color: black;">无</span>机会拿到这个奖。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">于是，你会进一步分析：</p>问题2：<span style="color: black;">日前</span>排行+<span style="color: black;">将来</span>10天预计增速，能否让我拿到这个奖数据2：<span style="color: black;">截止</span>19日，同比增长排行……<span style="color: black;">将来</span>10天，各团队预计完成<span style="color: black;">状况</span>答案2：从<span style="color: black;">日前</span>行业+<span style="color: black;">将来</span>增速来看，我能/<span style="color: black;">不可</span>拿到奖金<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">重视</span>！问题2比问题1要<span style="color: black;">繁杂</span><span style="color: black;">非常多</span>，<span style="color: black;">由于</span>问题1只需要统计历史数据就好了，问题2得预测<span style="color: black;">将来</span>10天的走势。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">在怎么预测上，可能有好几种<span style="color: black;">办法</span>，<span style="color: black;">例如</span>：</p>简单用前20天趋势，模拟<span style="color: black;">将来</span>10天走势（趋势外推）<span style="color: black;">按照</span>去年同期的走势，模拟<span style="color: black;">将来</span>10天走势（周期性分析）<span style="color: black;">按照</span>手头尚在跟进的客户*预计转化率，<span style="color: black;">举荐</span>结果（业务模型）<span style="color: black;">按照</span><span style="color: black;">营销</span>人数、<span style="color: black;">营销</span><span style="color: black;">花费</span>等数据建模，再预测结果（算法模型）<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">这<span style="color: black;">便是</span><span style="color: black;">咱们</span>常说的：<span style="color: black;">繁杂</span><span style="color: black;">需要</span>。当<span style="color: black;">需要</span>变<span style="color: black;">繁杂</span>的时候，数据分析的流程<span style="color: black;">亦</span>会变长，<span style="color: black;">重点</span>在数据环节，越复杂的<span style="color: black;">办法</span>，需要越多的数据准备。<span style="color: black;">那样</span>，有<span style="color: black;">那些</span>是<span style="color: black;">繁杂</span><span style="color: black;">办法</span>呢？</p>
<h1 style="color: black; text-align: left; margin-bottom: 10px;"><span style="color: black;">2、</span>四种<span style="color: black;">繁杂</span>度下，分析流程</h1>
<h2 style="color: black; text-align: left; margin-bottom: 10px;"><span style="color: black;">繁杂</span>度一级：认识<span style="color: black;">状况</span>。</h2>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">这种最简单，直接统计历史数据<span style="color: black;">就可</span>，<span style="color: black;">例如</span><span style="color: black;">截止</span>1月3日，今年新<span style="color: black;">增多</span>的用户数/累计完成的<span style="color: black;">营销</span>业绩；1月3日时间点上，商品库存有多少等等。<span style="color: black;">重视</span>！单纯<span style="color: black;">陈列</span>数据，并<span style="color: black;">不可</span>解释<span style="color: black;">状况</span>好坏，需要数据+判断标准，<span style="color: black;">例如</span>累计<span style="color: black;">营销</span>业绩+业绩考核标准，<span style="color: black;">这般</span><span style="color: black;">才可</span><span style="color: black;">发掘</span>问题。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">在这种<span style="color: black;">状况</span>下，数据分析流程<span style="color: black;">便是</span>：业务想<span style="color: black;">认识</span><span style="color: black;">状况</span>→统计数据指标+判断标准→描述<span style="color: black;">状况</span>。</p>
<h2 style="color: black; text-align: left; margin-bottom: 10px;"><span style="color: black;">繁杂</span>度二级：<span style="color: black;">原由</span>分析。</h2>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">典型的问题，<span style="color: black;">例如</span>业务问：“<span style="color: black;">为何</span>我的业绩没达标？”<span style="color: black;">重视</span>，此时业务有假设和没假设，处理流程不<span style="color: black;">同样</span>:</p><img src="https://p3-sign.toutiaoimg.com/tos-cn-i-tjoges91tu/87736039f3254054ca002164a39591fd~noop.image?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1725685638&x-signature=zTKUN31XPX9Xee6T2f5ZDzgqFYc%3D" style="width: 50%; margin-bottom: 20px;">
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">总之，想做得深入，<span style="color: black;">必定</span>要业务题假设才行，<span style="color: black;">否则</span>数据自己把指标拆来拆去，<span style="color: black;">特别有</span>可能只输出诸如：“<span style="color: black;">由于</span>人数少了，<span style="color: black;">因此</span>没达标，<span style="color: black;">意见</span>把人数搞高！”一类肤浅结论。</p>
<h2 style="color: black; text-align: left; margin-bottom: 10px;"><span style="color: black;">繁杂</span>度三级：优化表现。</h2>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">典型的问题，<span style="color: black;">例如</span>“我要怎么做，<span style="color: black;">才可</span>让业绩冲<span style="color: black;">第1</span>”。此时，需要把前边两个<span style="color: black;">繁杂</span>度的问题<span style="color: black;">所有</span>做完，<span style="color: black;">才可</span>有结论。</p><img src="https://p3-sign.toutiaoimg.com/tos-cn-i-tjoges91tu/a8de9c5e71c5afc3c1bb0ec6e1a790fc~noop.image?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1725685638&x-signature=PP%2FAp6iyqXx3HAP8U5JDg%2FYHy0k%3D" style="width: 50%; margin-bottom: 20px;">
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">因此</span>，优化表现类流程会<span style="color: black;">尤其</span>长。<span style="color: black;">非常多</span>数据分析师不<span style="color: black;">晓得</span>怎么提业务<span style="color: black;">提高</span><span style="color: black;">意见</span>，其实<span style="color: black;">便是</span><span style="color: black;">由于</span>缺少了前边这几步铺垫，啥<span style="color: black;">状况</span>都不<span style="color: black;">认识</span>，肯定没法直接提<span style="color: black;">意见</span>。</p>
<h2 style="color: black; text-align: left; margin-bottom: 10px;"><span style="color: black;">繁杂</span>度四级：预测走势。</h2>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">在上一节<span style="color: black;">已然</span>举过预测例子，<span style="color: black;">实质</span>上，凡是预测类的问题都很<span style="color: black;">繁杂</span>。起码得先<span style="color: black;">认识</span><span style="color: black;">状况</span>，<span style="color: black;">晓得</span>问题点，<span style="color: black;">晓得</span>业务有<span style="color: black;">无</span>计划做改进动作，收集一大堆信息以后<span style="color: black;">才可</span>做<span style="color: black;">恰当</span>预测。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">此时：</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">倘若</span>不<span style="color: black;">思虑</span>业务动作，<span style="color: black;">那样</span><span style="color: black;">能够</span>直接用趋势外推，分析流程就很短：业务想<span style="color: black;">晓得</span>预测结果 → 观察过往趋势→按过往趋势拟合函数→直接外推结果。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">倘若</span>业务想<span style="color: black;">思虑</span>自己的动作，<span style="color: black;">例如</span>“我追加投入会<span style="color: black;">怎么样</span>”，<span style="color: black;">此时</span>流程就<span style="color: black;">繁杂</span>了，业务想<span style="color: black;">晓得</span>预测结果 → 观察过往趋势→<span style="color: black;">创立</span>模型，量化投入影响→代入参数，预测结果。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">总之，业务上提问越<span style="color: black;">繁杂</span>，需要分析的环节越长，前期铺垫越多，<span style="color: black;">否则</span>很难输出有价值结论。</p>
<h1 style="color: black; text-align: left; margin-bottom: 10px;"><span style="color: black;">3、</span>从数据出发的流程</h1>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">还有一种<span style="color: black;">状况</span>，<span style="color: black;">便是</span>业务端没主动提<span style="color: black;">需要</span>，<span style="color: black;">然则</span>数据分析师需要主动从数据里读出业务含义，<span style="color: black;">发掘</span>业务问题。此时的基本流程是：数据→问题→答案。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">然则</span>！这个流程经常走不下去。<span style="color: black;">由于</span><span style="color: black;">非常多</span><span style="color: black;">非常多</span>数据分析师只看得到数据，对业务<span style="color: black;">状况</span>不<span style="color: black;">认识</span>。<span style="color: black;">因此呢</span>不<span style="color: black;">晓得</span>该怎么<span style="color: black;">诠释</span>这个数。<span style="color: black;">例如</span>：</p><span style="color: black;">晓得</span>累计<span style="color: black;">营销</span>额是多少，但不<span style="color: black;">晓得</span>业务考核标准，<span style="color: black;">因此呢</span>没法<span style="color: black;">诠释</span>“好/<span style="color: black;">欠好</span>”<span style="color: black;">晓得</span><span style="color: black;">营销</span>排行是啥，但不<span style="color: black;">晓得</span>业务的排行奖励细则，<span style="color: black;">因此呢</span>没法看出谁有<span style="color: black;">潜能</span>拿奖<span style="color: black;">晓得</span><span style="color: black;">营销</span><span style="color: black;">欠好</span>，是<span style="color: black;">由于</span>某个<span style="color: black;">制品</span>卖得很差，但不<span style="color: black;">认识</span><span style="color: black;">制品</span>属性，没法深入分析<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">因此呢</span>本篇用了很长篇幅介绍业务视角<span style="color: black;">怎样</span>展开分析，<span style="color: black;">便是</span>为了提醒<span style="color: black;">哪些</span><span style="color: black;">每日</span>对着报表发呆的<span style="color: black;">朋友</span>，多和业务沟通，多加深对业务背景/业务<span style="color: black;">情况</span>的<span style="color: black;">认识</span>。好在，有<span style="color: black;">非常多</span><span style="color: black;">机构</span>业务和数据沟通还是很紧密的，<span style="color: black;">因此呢</span>，<span style="color: black;">能够</span>优化下“数据→问题→答案”的基本流程。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">例如</span>：从异动出发，数据→<span style="color: black;">反常</span>波动→业务沟通→问题确认→深入分析/问题结束。这种<span style="color: black;">通常</span>是数据分析师<span style="color: black;">发掘</span>了指标异动，之后向业务确认。<span style="color: black;">倘若</span>异动是业务主动发起/早就<span style="color: black;">晓得</span>/<span style="color: black;">已然</span>采取<span style="color: black;">办法</span>应对，那就<span style="color: black;">再也不</span>纠结，如果是意外<span style="color: black;">状况</span>，就再深入分析，直到落实问题<span style="color: black;">原由</span>。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">例如</span>：从标杆出发，数据→寻找标杆→业务沟通→可复制性确认/问题结束。这种<span style="color: black;">通常</span>是数据上，<span style="color: black;">发掘</span>某个<span style="color: black;">制品</span>/某个地区/某个<span style="color: black;">途径</span>表现<span style="color: black;">尤其</span>好，此时数据<span style="color: black;">能够</span>向业务主动确认“是不是一个机会点”“值不值得推广”。<span style="color: black;">倘若</span>业务早就<span style="color: black;">晓得</span>了，那就结束。<span style="color: black;">倘若</span>业务感兴趣，就深入分析标杆可复制性，推动标杆推广。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;"><span style="color: black;">例如</span>：从<span style="color: black;">分部</span>联动<span style="color: black;">方向</span>：数据→<span style="color: black;">相关</span><span style="color: black;">状况</span>分析→信息共享→问题确认→深入分析/问题结束。这种<span style="color: black;">通常</span>是做经营分析的时候，<span style="color: black;">例如</span><span style="color: black;">发掘</span><span style="color: black;">营销</span>走势下滑，提醒供应<span style="color: black;">重视</span>积压<span style="color: black;">危害</span>；<span style="color: black;">例如</span><span style="color: black;">发掘</span>营销花钱很猛，关注现金流；<span style="color: black;">例如</span><span style="color: black;">发掘</span>业务计划大活动，提醒客服/售后准备。先共享信息，再看<span style="color: black;">关联</span><span style="color: black;">分部</span><span style="color: black;">是不是</span>有应对，<span style="color: black;">倘若</span><span style="color: black;">无</span>应对再深入分析。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">以上，所有流程总结如下图，方便<span style="color: black;">大众</span>取用：</p><img src="https://p3-sign.toutiaoimg.com/tos-cn-i-tjoges91tu/5dcb3baf844771ad9d73ed0b61e53d31~noop.image?_iz=58558&from=article.pc_detail&lk3s=953192f4&x-expires=1725685638&x-signature=O9kJ6KXQYLsdo4ZcWgcqbN1fPrE%3D" style="width: 50%; margin-bottom: 20px;">
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">本文由人人都是<span style="color: black;">制品</span>经理作者【接地气的陈老师】，<span style="color: black;">微X</span>公众号：【接地气的陈老师】，原创/授权 发布于人人都是<span style="color: black;">制品</span>经理，未经许可，禁止转载。</p>
<p style="font-size: 16px; color: black; line-height: 40px; text-align: left; margin-bottom: 15px;">题图来自Unsplash，基于 CC0 协议。</p>
谢谢、感谢、感恩、辛苦了、有你真好等。 外贸B2B平台有哪些？ 你的见解独到，让我受益匪浅，非常感谢。 期待与你深入交流，共探知识的无穷魅力。
页:
[1]