{
  "format": "variant-direction-card",
  "version": 1,
  "url": "https://variant-design-skill.vercel.app/d/pulse.json",
  "page": "https://variant-design-skill.vercel.app/d/pulse",
  "title": [
    "Pulse：周报的三种读法",
    "Pulse: three ways to read a weekly report"
  ],
  "summary": [
    "同一周的数据和锁定的品牌，分别先列出要处理的事、并排比较指标，或追查异常的原因。",
    "The same week of data under a locked brand, leading with what to act on, a side-by-side comparison, or tracing why."
  ],
  "brief": [
    "为订阅类产品的每周数据报告重新设计，读者是周一早上的产品负责人；品牌字体和颜色已锁定。",
    "Redesign the weekly analytics report of a subscription app for the product lead on Monday morning, with brand fonts and colors locked."
  ],
  "category": "data",
  "source": "own",
  "kind": "verified",
  "credit": {
    "name": "Variant Design",
    "url": "https://github.com/YuqingNicole/variant-design-skill"
  },
  "reuse": "Inspiration only: borrow the structure and tradeoffs. Images are shown for reference and are not licensed for reuse in your project.",
  "recommendation": {
    "variant": "A",
    "reason": [
      "读者是周一早上的产品负责人、数据只有一周快照，首要问题是“要处理什么”：A 把两条异常放在最前并可逐条标记已看，最贴合这个任务。这是基于页面注释的推断，尚未经用户测试；若周会要横向比较渠道，B 更合适。",
      "The reader is the product lead on Monday morning and the data is a single week, so the first question is what to act on. A puts the two anomalies first and lets each be marked as reviewed, which fits that task best. This is inferred from the page's comments and not tested with users; if the weekly meeting compares channels, B fits better."
    ]
  },
  "directions": [
    {
      "id": "A",
      "name": [
        "Monday Briefing · 周一简报",
        "Monday Briefing"
      ],
      "comparison": {
        "optimizes": [
          "Monday Briefing：两条异常作为编号要点置顶，附带对应指标与日期，可逐条标记“已看”；指标、趋势、渠道退为背景信息。",
          "Monday Briefing: the two anomalies come first as numbered items with their metric and date, each can be marked as reviewed; metrics, trend and channels become background."
        ],
        "tradeoff": [
          "单栏阅读，一屏看不到全部数字；渠道只按转化率排序，不能切换；横向比较能力最弱。",
          "Single-column reading, so not every number fits on one screen; channels sort only by conversion; weakest for side-by-side comparison."
        ],
        "bestFor": [
          "周一早上快速了解“这周要处理什么”、准备站会；读者主要是负责人本人。",
          "Seeing what needs action this week on Monday morning before stand-up; the reader is mainly the lead."
        ]
      },
      "image": "https://variant-design-skill.vercel.app/media/pulse/a.jpg"
    },
    {
      "id": "B",
      "name": [
        "Scorecard · 计分卡",
        "Scorecard"
      ],
      "comparison": {
        "optimizes": [
          "Scorecard：4 个 KPI 同屏并标注 Better/Worse，DAU 图可逐日查看并带均值线，渠道表可按试用、转化、估算付费数排序并显示占比。",
          "Scorecard: four KPIs side by side marked Better/Worse, a day-by-day DAU chart with an average line, and a channel table sortable by trials, conversion and estimated paid with share bars."
        ],
        "tradeoff": [
          "信息密度最高，异常降为侧栏，需要读者自己判断优先级；“估算付费数”是派生数值，需要团队认可这种算法。",
          "Highest density; anomalies move to a side panel so readers judge priority themselves; estimated paid is a derived number the team must agree on."
        ],
        "bestFor": [
          "周会投屏、跨渠道/跨指标比较，读者熟悉这些指标。",
          "Projecting in the weekly meeting and comparing across channels and metrics, for readers who know these metrics."
        ]
      },
      "image": "https://variant-design-skill.vercel.app/media/pulse/b.jpg"
    },
    {
      "id": "C",
      "name": [
        "Evidence Trace · 证据追踪",
        "Evidence Trace"
      ],
      "comparison": {
        "optimizes": [
          "Evidence Trace：选中一条异常后按“信号 → 时间 → 来源”展开证据，高亮对应 KPI、日期与渠道；顶部常驻 KPI 条保证总览不丢。",
          "Evidence Trace: selecting an anomaly expands its evidence as signal → when → where, highlighting the KPI, date and channel; a KPI strip stays on top so the overview is never lost."
        ],
        "tradeoff": [
          "需要点选才能看到完整图表和渠道表（“Full week overview” 页签）；当前数据缺少同比和 cohort 明细，证据链第 3 步对流失异常只能说明“数据未包含”。",
          "The full chart and channel table need a click (Full week overview tab); the data lacks year-over-year and cohort detail, so step 3 of the churn trace can only say the data is not included."
        ],
        "bestFor": [
          "要追查原因、给团队解释“为什么”的场景；数据源扩充（按日渠道、cohort）后价值最大。",
          "Tracing causes and explaining why to the team; most valuable once the data includes daily channel and cohort detail."
        ]
      },
      "image": "https://variant-design-skill.vercel.app/media/pulse/c.jpg"
    }
  ]
}