文件Experience Platform儀表板指南

受眾深入分析

Last update: Tue Jul 16 2024 00:00:00 GMT+0000 (Coordinated Universal Time)
  • 主題：

建立對象：

  • undefined
  • undefined

從資料模型分析衍生的深入解析可讓您的Adobe Real-time Customer Data Platform資料更易於存取、理解，並更能對決策產生影響。

存取提供對象分析功能的SQL，藉此瞭解您的對象分析，然後產生您自己的分析，進一步探索組成對象的身分和設定檔。 使用現有的Real-Time CDP資料模型SQL作為靈感，根據您獨特的業務需求建立查詢，將原始資料轉換為可採取行動的新見解。

如需有關如何直接透過PLatform UI調整您見解的SQL的詳細資訊，請參閱檢視SQL檔案。

下列深入分析都可供您用作對象儀表板或自訂使用者定義儀表板的一部分。 請參閱自訂總覽，瞭解如何自訂您的儀表板或🔗在Widget程式庫和使用者定義儀表板中建立及編輯新Widget的說明。

下列深入分析全都可供您用作對象儀表板或自訂儀表板的一部分。

客群重疊報告 audience-overlap-report

此深入分析所回答的問題：

  • 特定篩選對象的前50個重疊對象是什麼？
  • 特定篩選對象中最少重疊的50個對象為何？
  • 不同篩選對象的重疊模式會有何改變？
選取以顯示產生此深入分析的SQL
code language-sql 
SELECT source_segment_name,
        source_segment_id,
        overlap_segment_name,
        overlap_segment_id,
        max(source_segment_audience_count) source_segment_audience_count,
        max(overlap_segment_audience_count) overlap_segment_audience_count,
        max(overlap_audience_count) overlap_audience_count,
        CASE
            WHEN (max(source_segment_audience_count) + max(overlap_segment_audience_count) - max(overlap_audience_count)) > 0 THEN (cast(max(overlap_audience_count) AS DECIMAL(18, 2)) / cast((max(source_segment_audience_count) + max(overlap_segment_audience_count) - max(overlap_audience_count)) AS DECIMAL(18, 2))) * 100::DECIMAL(9, 2)
            ELSE 100.00
        END overlapping_percentage
  FROM
    (SELECT adwh_fact_profile_overlap_of_segments.Segment1 source_segment_id,
            adwh_fact_profile_overlap_of_segments.Segment2 overlap_segment_id,
            Sum(count_of_overlap) overlap_audience_count
    FROM qsaccel.profile_agg.adwh_fact_profile_overlap_of_segments
    WHERE qsaccel.profile_agg.adwh_fact_profile_overlap_of_segments.merge_policy_id = 2027892989
      AND qsaccel.profile_agg.adwh_fact_profile_overlap_of_segments.date_key = '2024-01-10'
    GROUP BY qsaccel.profile_agg.adwh_fact_profile_overlap_of_segments.Segment2 ,
              qsaccel.profile_agg.adwh_fact_profile_overlap_of_segments.Segment1) a
  INNER JOIN
    (SELECT sum(count_of_profiles) source_segment_audience_count,
            adwh_dim_segments.segment_name source_segment_name,
            adwh_fact_profile_by_segment_trendlines.merge_policy_id,
            adwh_fact_profile_by_segment_trendlines.segment_Id segment1
    FROM qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines
    JOIN qsaccel.profile_agg.adwh_dim_segments ON qsaccel.profile_agg.adwh_dim_segments.segment_id = qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.segment_Id
    WHERE qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.merge_policy_id = 2027892989
      AND qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.date_key = '2024-01-10'
    GROUP BY qsaccel.profile_agg.adwh_dim_segments.segment_name,
              qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.merge_policy_id,
              qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.segment_id) b ON a.source_segment_id = b.segment1
  INNER JOIN
    (SELECT sum(count_of_profiles) overlap_segment_audience_count,
            adwh_dim_segments.segment_name overlap_segment_name,
            adwh_fact_profile_by_segment_trendlines.merge_policy_id,
            adwh_fact_profile_by_segment_trendlines.segment_Id segment2
    FROM qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines
    JOIN qsaccel.profile_agg.adwh_dim_segments ON adwh_dim_segments.segment_id = adwh_fact_profile_by_segment_trendlines.segment_Id
    WHERE qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.merge_policy_id = 2027892989
      AND qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.date_key = '2024-01-10'
    GROUP BY qsaccel.profile_agg.adwh_dim_segments.segment_name,
              qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.merge_policy_id,
              qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.segment_id) c ON a.overlap_segment_id = c.segment2
  GROUP BY source_segment_name,
          source_segment_id,
          overlap_segment_name,
          overlap_segment_id
  ORDER BY overlapping_percentage DESC
  LIMIT 5;

如需此深入分析的外觀和功能的相關資訊，請參閱對象重疊報表Widget檔案。

客群重疊 audience-overlap

此深入分析所回答的問題：

  • 哪些設定檔對兩個對象都是通用的？
  • 重疊對參與或轉換率有何影響？
  • 如何為重疊的區段量身打造行銷策略？
選取以顯示產生此深入分析的SQL
code language-sql 
SELECT Sum(overlap_col1) overlap_col1,
        Sum(overlap_col2) overlap_col2,
        Sum(overlap_count) Overlap_count
  FROM
    (SELECT 0 overlap_col1,
            0 overlap_col2,
            sum(count_of_overlap)Overlap_count
    FROM qsaccel.profile_agg.adwh_fact_profile_overlap_of_segments
    WHERE qsaccel.profile_agg.adwh_fact_profile_overlap_of_segments.merge_policy_id = 1133248113
      AND qsaccel.profile_agg.adwh_fact_profile_overlap_of_segments.date_key = '2024-01-10'
      AND ((qsaccel.profile_agg.adwh_fact_profile_overlap_of_segments.segment1=1870062812
            AND qsaccel.profile_agg.adwh_fact_profile_overlap_of_segments.segment2=2080256533)
            OR (qsaccel.profile_agg.adwh_fact_profile_overlap_of_segments.segment1=2080256533
                AND qsaccel.profile_agg.adwh_fact_profile_overlap_of_segments.segment2=1870062812))
    UNION ALL SELECT sum(count_of_profiles) overlap_col1,
                      0 overlap_col2,
                      0 overlap_count
    FROM qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines
    LEFT JOIN qsaccel.profile_agg.adwh_dim_segments ON qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.segment_Id = qsaccel.profile_agg.adwh_dim_segments.segment_Id
    WHERE qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.merge_policy_id = 1133248113
      AND qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.date_key = '2024-01-10'
      AND qsaccel.profile_agg.adwh_dim_segments.segment_Id = 1870062812
    UNION ALL SELECT 0 overlap_col1,
                      sum(count_of_profiles) overlap_col2,
                      0 Overlap_count
    FROM qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines
    JOIN qsaccel.profile_agg.adwh_dim_segments ON qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.segment_Id = qsaccel.profile_agg.adwh_dim_segments.segment_Id
    WHERE qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.merge_policy_id = 1133248113
      AND qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.date_key = '2024-01-10'
      AND qsaccel.profile_agg.adwh_dim_segments.segment_Id = 2080256533 ) a;

如需此深入分析的外觀和功能的相關資訊，請參閱對象重疊Widget檔案。

客群人數變化趨勢 audience-size-change-trend

此深入分析所回答的問題：

  • 過去30天、90天或12個月內，對象人數是否出現任何重大上升或下降？
  • 在特定日子內，對象人數會有何改變？
  • 在過去12個月內是否偵測到任何異常或重複的尖峰或下降模式？
選取以顯示產生此深入分析的SQL
code language-sql 
SELECT date_key,
      Profiles_added
  FROM
    (SELECT rn_num,
            date_key,
            (count_of_profiles-lag(count_of_profiles, 1, 0) over(
                                                                ORDER BY date_key))Profiles_added
    FROM
      (SELECT date_key,
              sum(x.count_of_profiles)count_of_profiles,
              row_number() OVER (
                                  ORDER BY date_key) rn_num
        FROM qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines x
        INNER JOIN
          (SELECT MAX(process_date) last_process_date,
                  merge_policy_id
          FROM qsaccel.profile_agg.adwh_lkup_process_delta_log
          WHERE process_name = 'FACT_TABLES_PROCESSING'
            AND process_status = 'SUCCESSFUL'
          GROUP BY merge_policy_id) y ON x.merge_policy_id = y.merge_policy_id
        WHERE segment_id = 1333234510
          AND x.date_key >= dateadd(DAY, -30 -1, y.last_process_date)
        GROUP BY x.date_key) a)b
  WHERE rn_num > 1;

如需此深入分析的外觀和功能的相關資訊，請參閱對象人數變化趨勢Widget檔案。

依身分列出的客群人數趨勢 audience-size-trend-by-identity

此深入分析所回答的問題：

  • 我的對象是否持續成長、穩定或經歷波動？
  • 是否有任何特定身分會隨著時間推移而增加或減少受眾？
  • 在一段時間內，我的身分成長中是否有任何異常？
選取以顯示產生此深入分析的SQL
code language-sql 
SELECT sum(count_of_profiles) AS identities,
        date_key
  FROM qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines x
  INNER JOIN
    (SELECT MAX(process_date) last_process_date,
            merge_policy_id
    FROM qsaccel.profile_agg.adwh_lkup_process_delta_log
    WHERE process_name = 'FACT_TABLES_PROCESSING'
      AND process_status = 'SUCCESSFUL'
    GROUP BY merge_policy_id) y ON x.merge_policy_id = y.merge_policy_id
  INNER JOIN qsaccel.profile_agg.adwh_dim_namespaces z ON x.namespace_id = z.namespace_id
  AND x.merge_policy_id = z.merge_policy_id
  WHERE x.date_key >= dateadd(DAY, -30, y.last_process_date)
    AND x.segment_id = 1333234510
    AND z.namespace_description = 'crmid'
  GROUP BY date_key;

如需此深入分析的外觀和功能的相關資訊，請參閱身分識別介面工具集檔案的對象人數趨勢。

對象規模趨勢 audience-size-trend

此深入分析所回答的問題：

  • 對象人數在一段時間內的變化（包括任何異常）如何？
  • 我如何尋找期間（30天、90天和12個月）內對象人數的整體趨勢？
  • 有哪些對象的主要特性會影響其規模？ 例如，由於電子郵件行銷活動而出現尖峰。
選取以顯示產生此深入分析的SQL
code language-sql 
SELECT date_key,
        sum(count_of_profiles) AS audience_size
  FROM qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines x
  INNER JOIN
    (SELECT MAX(process_date) last_process_date,
            merge_policy_id
    FROM qsaccel.profile_agg.adwh_lkup_process_delta_log
    WHERE process_name = 'FACT_TABLES_PROCESSING'
      AND process_status = 'SUCCESSFUL'
    GROUP BY merge_policy_id) y ON x.merge_policy_id = y.merge_policy_id
  WHERE date_key >= dateadd(DAY, -30, y.last_process_date)
    AND x.segment_id = 1333234510
  GROUP BY date_key,
          segment_id;

如需此深入分析的外觀和功能的相關資訊，請參閱對象人數趨勢Widget檔案。

對象規模 audience-size

此深入分析所回答的問題：

  • 目前的受眾總人數是多少？
  • 目前的對象人數與先前時段或特定對象有何不同？
  • 最近的行銷活動對對象人數有何影響？
選取以顯示產生此深入分析的SQL
code language-sql 
SELECT
  sum(
    qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.count_of_profiles
  ) count_of_profiles
FROM
  qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines
  LEFT OUTER JOIN qsaccel.profile_agg.adwh_dim_segments ON qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.segment_id = qsaccel.profile_agg.adwh_dim_segments.segment_id
WHERE
  qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.segment_id = -1323307941
  AND qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.merge_policy_id = 1914917902
  AND qsaccel.profile_agg.adwh_fact_profile_by_segment_trendlines.date_key = '2024-01-12';

如需此深入分析的外觀和功能的相關資訊，請參閱對象大小Widget檔案。

Customer AI 分數的分佈 customer-ai-distribution-of-scores

此深入分析所回答的問題：

  • 我的Customer AI模型的每個貯體（依選取的對象篩選）的評分分佈為何？
  • 特定對象的高分、中分和低分的得分分佈情況如何？
  • 不同目標對象的評分分佈劃分如何？
選取以顯示產生此深入分析的SQL
code language-sql 
SELECT b.model_name,
      b.model_type,
      c.segment_name,
      c.segment_id,
      CASE
        WHEN score >= 0
          AND score < 25 THEN 'LOW'
        WHEN score >= 25
          AND score < 75 THEN 'MEDIUM'
        WHEN score >= 75
          AND score <= 100 THEN 'HIGH'
        END bucket_name,
      CASE
        WHEN score >= 0
          AND score < 5 THEN '02.50'
        WHEN score >= 5
          AND score < 10 THEN '07.50'
        WHEN score >= 10
          AND score < 15 THEN '12.50'
        WHEN score >= 15
          AND score < 20 THEN '17.50'
        WHEN score >= 20
          AND score < 25 THEN '22.50'
        WHEN score >= 25
          AND score < 30 THEN '27.50'
        WHEN score >= 30
          AND score < 35 THEN '32.50'
        WHEN score >= 35
          AND score < 40 THEN '37.50'
        WHEN score >= 40
          AND score < 45 THEN '42.50'
        WHEN score >= 45
          AND score < 50 THEN '47.50'
        WHEN score >= 50
          AND score < 55 THEN '52.50'
        WHEN score >= 55
          AND score < 60 THEN '57.50'
        WHEN score >= 60
          AND score < 65 THEN '62.50'
        WHEN score >= 65
          AND score < 70 THEN '67.50'
        WHEN score >= 70
          AND score < 75 THEN '72.50'
        WHEN score >= 75
          AND score < 80 THEN '77.50'
        WHEN score >= 80
          AND score < 85 THEN '82.50'
        WHEN score >= 85
          AND score < 90 THEN '87.50'
        WHEN score >= 90
          AND score < 95 THEN '92.50'
        WHEN score >= 95
          AND score <= 100 THEN '97.50'
        END score_bins,
      Sum(CASE
            WHEN score >= 0
              AND score < 25 THEN count_of_profiles
            WHEN score >= 25
              AND score < 75 THEN count_of_profiles
            WHEN score >= 75
              AND score <= 100 THEN count_of_profiles
        END) count_of_profiles
   FROM qsaccel.profile_agg.adwh_fact_profile_by_segment_ai_models a
          JOIN qsaccel.profile_agg.adwh_dim_ai_models b ON a.merge_policy_id = b.merge_policy_id
     AND a.model_id = b.model_id
          JOIN qsaccel.profile_agg.adwh_dim_segments c ON a.segment_id = c.segment_id
   WHERE a.merge_policy_id = 1133248113
     AND a.model_id = 1829081696
     AND a.segment_id = 1870062812
     AND score_date =
         (SELECT MAX(score_date)
          FROM qsaccel.profile_agg.adwh_fact_profile_by_segment_ai_models d
          WHERE d.model_id = a.model_id) GROUP  BY b.model_name,
             b.model_type,
             c.segment_name,
             c.segment_id,
             CASE
               WHEN score >= 0
                 AND score < 25 THEN 'LOW'
               WHEN score >= 25
                 AND score < 75 THEN 'MEDIUM'
               WHEN score >= 75
                 AND score <= 100 THEN 'HIGH'
               END,
             CASE
               WHEN score >= 0
                 AND score < 5 THEN '02.50'
               WHEN score >= 5
                 AND score < 10 THEN '07.50'
               WHEN score >= 10
                 AND score < 15 THEN '12.50'
               WHEN score >= 15
                 AND score < 20 THEN '17.50'
               WHEN score >= 20
                 AND score < 25 THEN '22.50'
               WHEN score >= 25
                 AND score < 30 THEN '27.50'
               WHEN score >= 30
                 AND score < 35 THEN '32.50'
               WHEN score >= 35
                 AND score < 40 THEN '37.50'
               WHEN score >= 40
                 AND score < 45 THEN '42.50'
               WHEN score >= 45
                 AND score < 50 THEN '47.50'
               WHEN score >= 50
                 AND score < 55 THEN '52.50'
               WHEN score >= 55
                 AND score < 60 THEN '57.50'
               WHEN score >= 60
                 AND score < 65 THEN '62.50'
               WHEN score >= 65
                 AND score < 70 THEN '67.50'
               WHEN score >= 70
                 AND score < 75 THEN '72.50'
               WHEN score >= 75
                 AND score < 80 THEN '77.50'
               WHEN score >= 80
                 AND score < 85 THEN '82.50'
               WHEN score >= 85
                 AND score < 90 THEN '87.50'
               WHEN score >= 90
                 AND score < 95 THEN '92.50'
               WHEN score >= 95
                 AND score <= 100 THEN '97.50'
               END;

如需此深入分析的外觀和功能的相關資訊，請參閱分數Widget檔案的Customer AI分佈。

Customer AI 評分摘要 customer-ai-scoring-summary

此深入分析所回答的問題：

  • 我針對特定對象的每個Customer AI模型的分數摘要為何？
  • 我的Customer AI傾向分數如何隨著不同對象而改變？
  • 我的評分摘要與對象概觀中的其他KPI相比如何？
選取以顯示產生此深入分析的SQL
code language-sql 
SELECT model_name,
         model_type,
         segment_name,
         CASE
             WHEN score BETWEEN 0 AND 24 THEN 'LOW'
             WHEN score BETWEEN 25 AND 74 THEN 'MEDIUM'
             WHEN score BETWEEN 75 AND 100 THEN 'HIGH'
         END score_buckets,
         sum(count_of_profiles) count_of_profiles
  FROM QSAccel.profile_agg.adwh_fact_profile_by_segment_ai_models a
  JOIN QSAccel.profile_agg.adwh_dim_ai_models b ON a.merge_policy_id=b.merge_policy_id
  AND a.model_id=b.model_id
  JOIN QSAccel.profile_agg.adwh_dim_segments c ON a.segment_id=c.segment_id
  WHERE a.merge_policy_id=1133248113
    AND a.model_id =1829081696
    AND a.segment_id=1870062812
    AND score_date=
      (SELECT max(score_date)
       FROM QSAccel.profile_agg.adwh_fact_profile_by_segment_ai_models d
       WHERE d.model_id=a.model_id)
  GROUP BY model_name,
           model_type,
           segment_name,
           CASE
               WHEN score BETWEEN 0 AND 24 THEN 'LOW'
               WHEN score BETWEEN 25 AND 74 THEN 'MEDIUM'
               WHEN score BETWEEN 75 AND 100 THEN 'HIGH'
           END;

如需此深入分析的外觀和功能的相關資訊，請參閱Customer AI評分摘要Widget檔案。

身分識別覆蓋 identity-overlap

此深入分析所回答的問題：

  • 對於篩選的對象，身分型別A和身分型別B之間的共同交集為何？
  • 我如何根據特定身分型別的重疊來調整客戶對象，以增強目標式行銷策略？
  • 評估交集區域內的行銷活動績效能獲得哪些深入分析？
  • 根據這些見解，如何才能最佳化未來的行銷工作？
選取以顯示產生此深入分析的SQL
code language-sql 
SELECT Sum(overlap_col1) overlap_col1,
        Sum(overlap_col2) overlap_col2,
        Sum(overlap_count) Overlap_count
  FROM
    (SELECT 0 overlap_col1,
            0 overlap_col2,
            Sum(count_of_profiles) Overlap_count
    FROM qsaccel.profile_agg.adwh_fact_profile_overlap_of_namespace_by_segment
    WHERE qsaccel.profile_agg.adwh_fact_profile_overlap_of_namespace_by_segment.segment_id = 1333234510
      AND qsaccel.profile_agg.adwh_fact_profile_overlap_of_namespace_by_segment.merge_policy_id = 1709997014
      AND qsaccel.profile_agg.adwh_fact_profile_overlap_of_namespace_by_segment.date_key = '2024-01-10'
      AND qsaccel.profile_agg.adwh_fact_profile_overlap_of_namespace_by_segment.overlap_id IN
        (SELECT a.overlap_id
          FROM
            (SELECT qsaccel.profile_agg.adwh_dim_overlap_namespaces.overlap_id overlap_id,
                    count(*) cnt_num
            FROM qsaccel.profile_agg.adwh_dim_overlap_namespaces
            WHERE qsaccel.profile_agg.adwh_dim_overlap_namespaces.merge_policy_id = 1709997014
              AND qsaccel.profile_agg.adwh_dim_overlap_namespaces.overlap_namespaces in ('crmid',
                                                                                          'email')
            GROUP BY qsaccel.profile_agg.adwh_dim_overlap_namespaces.overlap_id)a
          WHERE a.cnt_num>1 )
    UNION ALL SELECT count_of_profiles overlap_col1,
                      0 overlap_col2,
                      0 Overlap_count
    FROM qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines
    LEFT OUTER JOIN qsaccel.profile_agg.adwh_dim_namespaces ON qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines.namespace_id = qsaccel.profile_agg.adwh_dim_namespaces.namespace_id
    AND qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines.merge_policy_id = qsaccel.profile_agg.adwh_dim_namespaces.merge_policy_id
    WHERE qsaccel.profile_agg.adwh_dim_namespaces.namespace_description = 'crmid'
      AND qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines.segment_id = 1333234510
      AND qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines.merge_policy_id = 1709997014
      AND qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines.date_key = '2024-01-10'
    UNION ALL SELECT 0 overlap_col1,
                      count_of_profiles overlap_col2,
                      0 Overlap_count
    FROM qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines
    LEFT OUTER JOIN qsaccel.profile_agg.adwh_dim_namespaces ON qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines.namespace_id = qsaccel.profile_agg.adwh_dim_namespaces.namespace_id
    AND qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines.merge_policy_id = qsaccel.profile_agg.adwh_dim_namespaces.merge_policy_id
    WHERE qsaccel.profile_agg.adwh_dim_namespaces.namespace_description = 'email'
      AND qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines.segment_id = 1333234510
      AND qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines.merge_policy_id = 1709997014
      AND qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines.date_key = '2024-01-10' ) a;

如需此深入分析的外觀和功能的相關資訊，請參閱身分重疊Widget檔案。

依身分識別劃分的設定檔 profiles-by-identity

此深入分析所回答的問題：

  • 在選定對象的設定檔總數中，哪個身分型別所佔比例最高？
  • 選定對象的身分型別之間是否有重大差異？
  • 依對象區分的身分型別整體分佈情況如何？
  • 不同受眾的身分計數是否有任何重大差異或異常？
選取以顯示產生此深入分析的SQL
code language-sql 
SELECT qsaccel.profile_agg.adwh_dim_namespaces.namespace_description,
        sum(qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines.count_of_profiles) count_of_profiles
  FROM qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines
  LEFT OUTER JOIN qsaccel.profile_agg.adwh_dim_namespaces ON qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines.namespace_id = qsaccel.profile_agg.adwh_dim_namespaces.namespace_id
  AND qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines.merge_policy_id = qsaccel.profile_agg.adwh_dim_namespaces.merge_policy_id
  WHERE qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines.segment_id = 1333234510
    AND qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines.merge_policy_id = 1709997014
    AND qsaccel.profile_agg.adwh_fact_profile_by_segment_and_namespace_trendlines.date_key = '2024-01-10'
  GROUP BY qsaccel.profile_agg.adwh_dim_namespaces.namespace_description
  ORDER BY count_of_profiles DESC;

如需此深入分析的外觀和功能的相關資訊，請參閱依身分Widget的設定檔檔案。

排定的啟用 scheduled-activations

此深入分析所回答的問題：

  • 特定平台上特定對象表現最佳的啟用的開始和結束日期為何？
  • 哪些平台最常用於特定對象的已排程啟用？
  • 平台使用方式中是否有任何模式，可指引您做出針對特定受眾設定啟動策略優先順序或使其多樣化的決策？
選取以顯示產生此深入分析的SQL
code language-sql 
SELECT p.destination_platform ,
       p.destination_platform_name AS platform ,
       d.destination_name ,
       d.destination ,
       br.start_date ,
       CASE
           WHEN br.end_date = '9999-12-31' THEN 'Ongoing'
           ELSE br.end_date
       END AS end_date
  FROM qsaccel.profile_agg.adwh_dim_br_segment_destinations br
  JOIN qsaccel.profile_agg.adwh_dim_destination d ON br.destination_id = d.destination_id
  JOIN qsaccel.profile_agg.adwh_dim_destination_platform p ON d.destination_platform_id = p.destination_platform_id
  JOIN
    (SELECT MAX(process_date) AS last_process_date
     FROM qsaccel.profile_agg.adwh_lkup_process_delta_log
     WHERE process_name = 'FACT_TABLES_PROCESSING'
       AND process_status = 'SUCCESSFUL' ) lpd ON lpd.last_process_date BETWEEN br.start_date AND br.end_date
  AND br.segment_id = 1333234510;

如需此深入分析的外觀和功能的相關資訊，請參閱排程啟用Widget檔案。

後續步驟

閱讀本檔案後，您現在瞭解產生儀表板深入分析的SQL，以及此分析解決哪些常見問題。 您現在可以編輯並反複處理SQL，以產生您自己的深入分析。

如需有關如何直接透過PLatform UI調整您見解的SQL的詳細資訊，請參閱檢視SQL檔案。

您也可以閱讀並瞭解產生設定檔、帳戶設定檔和目的地儀表板之深入分析的SQL。

recommendation-more-help
ececc77d-ff44-4382-85ee-a087c8834323