# Before/After比較グラフの作成 import matplotlib.pyplot as plt import pandas as pd import numpy as np # 日本語フォント設定（Google Colab用） plt.rcParams[‘font.family’] = ‘DejaVu Sans’ # サンプルデータ: SNS広告キャンペーンの結果 data = { ‘指標’: [‘訪問者数’, ‘CV数’, ‘CVR(%)’, ‘売上(万円)’], ‘Before’: [10000, 50, 0.5, 500], # 施策前 ‘After’: [15000, 120, 0.8, 1200], # 施策後 ‘目標’: [14000, 100, 0.7, 1000] # 目標値 } df = pd.DataFrame(data) # 達成率と増加率を計算 df[‘達成率(%)’] = (df[‘After’] / df[‘目標’] * 100).round(1) df[‘増加率(%)’] = ((df[‘After’] – df[‘Before’]) / df[‘Before’] * 100).round(1) # 4つの指標を2×2のグリッドで表示 fig, axes = plt.subplots(2, 2, figsize=(14, 10)) for i, metric in enumerate(df[‘指標’]): row = i // 2 # 行番号（0または1） col = i % 2 # 列番号（0または1） ax = axes[row, col] # データを取得 before = df.loc[i, ‘Before’] after = df.loc[i, ‘After’] target = df.loc[i, ‘目標’] increase = df.loc[i, ‘増加率(%)’] # 棒グラフを描画 x = [‘Before’, ‘After’, ‘Target’] y = [before, after, target] colors = [‘#E74C3C’, ‘#2ECC71’, ‘#3498DB’] # 赤、緑、青 bars = ax.bar(x, y, color=colors, alpha=0.7, edgecolor=’black’, linewidth=2) # 棒の上に数値を表示 for bar, val in zip(bars, y): height = bar.get_height() ax.text(bar.get_x() + bar.get_width()/2., height, f'{val:,.0f}’, ha=’center’, va=’bottom’, fontsize=12, fontweight=’bold’) # 増加率を表示 ax.text(0.5, max(y) * 0.85, f’+{increase}%’, ha=’center’, fontsize=14, fontweight=’bold’, bbox=dict(boxstyle=’round’, facecolor=’yellow’, alpha=0.7)) ax.set_title(metric, fontsize=14, fontweight=’bold’) ax.set_ylabel(‘Value’, fontsize=12) ax.grid(axis=’y’, alpha=0.3) plt.suptitle(‘SNS Campaign: Before/After Comparison’, fontsize=18, fontweight=’bold’, y=0.995) plt.tight_layout() plt.show() # サマリーを表示 print(“\n【施策効果サマリー】”) print(df.to_string(index=False))

# 月別のROI推移グラフ（Plotly） import plotly.graph_objects as go import numpy as np # 月別データ months = [‘1月’, ‘2月’, ‘3月’, ‘4月’, ‘5月’, ‘6月’] ad_cost = [50, 55, 60, 65, 70, 75] # 広告費（万円） revenue = [150, 180, 210, 260, 280, 320] # 売上（万円） # ROIを計算 roi = [((r – c) / c * 100) for r, c in zip(revenue, ad_cost)] # グラフ作成 fig = go.Figure() # 広告費（棒グラフ・赤） fig.add_trace(go.Bar( x=months, y=ad_cost, name=’広告費’, marker_color=’#E74C3C’, yaxis=’y’ )) # 売上（棒グラフ・緑） fig.add_trace(go.Bar( x=months, y=revenue, name=’売上’, marker_color=’#2ECC71′, yaxis=’y’ )) # ROI（折れ線グラフ・青、右軸） fig.add_trace(go.Scatter( x=months, y=roi, name=’ROI’, mode=’lines+markers’, line=dict(color=’#3498DB’, width=3), marker=dict(size=12), yaxis=’y2′ )) # レイアウト設定 fig.update_layout( title=’月別 広告費・売上・ROI 推移‘, xaxis_title=’月’, yaxis_title=’金額（万円）’, yaxis2=dict( title=’ROI（%）’, overlaying=’y’, side=’right’, showgrid=False ), height=500, hovermode=’x unified’, legend=dict(x=0.01, y=0.99, bgcolor=’rgba(255,255,255,0.8)’), barmode=’group’ ) fig.show() # サマリーを表示 print(f”\n【ROI分析サマリー】”) print(f”平均ROI: {np.mean(roi):.1f}%”) print(f”累計広告費: {sum(ad_cost)}万円”) print(f”累計売上: {sum(revenue)}万円”) print(f”累計利益: {sum(revenue) – sum(ad_cost)}万円”)

# ストーリーボード形式のレポート import matplotlib.pyplot as plt import numpy as np fig = plt.figure(figsize=(16, 12)) # 1. 背景：既存顧客の年齢構成（課題の可視化） ax1 = plt.subplot(3, 3, 1) age_groups = [’10s’, ’20s’, ’30s’, ’40s’, ’50s’, ’60s+’] current_customers = [5, 15, 25, 30, 20, 5] ax1.bar(age_groups, current_customers, color=’#E74C3C’, alpha=0.7) ax1.set_title(‘Background: Age Distribution’, fontweight=’bold’) ax1.set_ylabel(‘Ratio (%)’) ax1.axhline(y=16.7, color=’green’, linestyle=’–‘, label=’Ideal’) ax1.legend() ax1.text(1.5, 28, ‘Young users\nare few!’, fontsize=10, bbox=dict(boxstyle=’round’, facecolor=’yellow’)) # 2. 施策：月別予算 ax2 = plt.subplot(3, 3, 2) months = [‘Apr’, ‘May’, ‘Jun’] budget = [60, 70, 70] ax2.bar(months, budget, color=’#3498DB’) ax2.set_title(‘Campaign: Monthly Budget’, fontweight=’bold’) ax2.set_ylabel(‘Budget (10K JPY)’) ax2.text(1, 55, f’Total\n{sum(budget)}’, ha=’center’, fontsize=14, fontweight=’bold’, bbox=dict(boxstyle=’round’, facecolor=’lightblue’)) # 3. 結果：訪問者数Before/After ax3 = plt.subplot(3, 3, 3) categories = [‘Before’, ‘After’] visitors = [10000, 15000] colors = [‘#E74C3C’, ‘#2ECC71’] bars = ax3.bar(categories, visitors, color=colors, alpha=0.7) ax3.set_title(‘Result: Visitors +50%’, fontweight=’bold’) ax3.set_ylabel(‘Visitors’) for bar, val in zip(bars, visitors): ax3.text(bar.get_x() + bar.get_width()/2, val, f'{val:,}’, ha=’center’, va=’bottom’, fontsize=12) # 4. 結果：CV数Before/After ax4 = plt.subplot(3, 3, 4) cv = [50, 120] bars = ax4.bar(categories, cv, color=colors, alpha=0.7) ax4.set_title(‘Result: CV +140%’, fontweight=’bold’) ax4.set_ylabel(‘Conversions’) for bar, val in zip(bars, cv): ax4.text(bar.get_x() + bar.get_width()/2, val, f'{val}’, ha=’center’, va=’bottom’, fontsize=12) # 5. 結果：ROI推移 ax5 = plt.subplot(3, 3, 5) roi_trend = [200, 250, 300] ax5.plot(months, roi_trend, marker=’o’, linewidth=3, markersize=12, color=’#2ECC71′) ax5.set_title(‘Result: ROI Trend (Improving)’, fontweight=’bold’) ax5.set_ylabel(‘ROI (%)’) ax5.grid(axis=’y’, alpha=0.3) ax5.axhline(y=200, color=’red’, linestyle=’–‘, label=’Target 200%’) ax5.legend() # 6. 成功要因：年齢別リーチ改善 ax6 = plt.subplot(3, 3, 6) new_reach = [10, 35, 25, 20, 8, 2] ax6.bar(age_groups, new_reach, color=’#2ECC71′, alpha=0.7) ax6.set_title(‘Success: New Age Distribution’, fontweight=’bold’) ax6.set_ylabel(‘Ratio (%)’) ax6.text(1, 33, ‘Young users\nincreased!’, fontsize=10, bbox=dict(boxstyle=’round’, facecolor=’lightgreen’)) # 7. 分析：チャネル別パフォーマンス ax7 = plt.subplot(3, 3, 7) channels = [‘Instagram’, ‘Facebook’, ‘Twitter’] performance = [150, 80, 30] ax7.barh(channels, performance, color=[‘#E74C3C’, ‘#3498DB’, ‘#00acee’]) ax7.set_title(‘Analysis: CV by Channel’, fontweight=’bold’) ax7.set_xlabel(‘Conversions’) # 8. 課題：CPA上昇傾向 ax8 = plt.subplot(3, 3, 8) weeks = [‘W1’, ‘W2’, ‘W3’, ‘W4′] cpa = [15000, 14000, 16000, 18000] ax8.plot(weeks, cpa, marker=’o’, linewidth=2, color=’#E74C3C’) ax8.set_title(‘Challenge: CPA Rising’, fontweight=’bold’) ax8.set_ylabel(‘CPA (JPY)’) ax8.axhline(y=15000, color=’green’, linestyle=’–‘, label=’Target’) ax8.legend() ax8.text(2, 17500, ‘Needs\nimprovement’, fontsize=9, bbox=dict(boxstyle=’round’, facecolor=’pink’)) # 9. 提案：予算増額 ax9 = plt.subplot(3, 3, 9) proposals = [‘Current’, ‘Proposed’] budgets = [200, 300] bars = ax9.bar(proposals, budgets, color=[‘#3498DB’, ‘#2ECC71’], alpha=0.7) ax9.set_title(‘Proposal: Budget x1.5′, fontweight=’bold’) ax9.set_ylabel(‘Budget (10K JPY)’) for bar, val in zip(bars, budgets): ax9.text(bar.get_x() + bar.get_width()/2, val, f'{val}’, ha=’center’, va=’bottom’, fontsize=12) ax9.text(0.5, 240, ‘ROI 250%\n→ Increase OK!’, ha=’center’, bbox=dict(boxstyle=’round’, facecolor=’yellow’)) plt.suptitle(‘SNS Campaign Report – Storyboard’, fontsize=20, fontweight=’bold’, y=0.995) plt.tight_layout() plt.show() print(“\nストーリーボード形式で、”) print(“背景→施策→結果→分析→提案の流れを可視化しました！”)

# matplotlibでPDFレポートを生成 from matplotlib.backends.backend_pdf import PdfPages import matplotlib.pyplot as plt import pandas as pd def create_marketing_report(): “””マーケティングレポートをPDFで生成する関数””” # PDFファイル名 pdf_filename = ‘marketing_report_2024.pdf’ # PdfPagesでPDFを作成 with PdfPages(pdf_filename) as pdf: # ========== ページ1: タイトルページ ========== fig = plt.figure(figsize=(11, 8.5)) # レターサイズ ax = fig.add_subplot(111) ax.axis(‘off’) # 軸を非表示 # タイトル ax.text(0.5, 0.7, ‘SNS Campaign’, ha=’center’, fontsize=36, fontweight=’bold’, transform=ax.transAxes) ax.text(0.5, 0.6, ‘Performance Report’, ha=’center’, fontsize=32, fontweight=’bold’, transform=ax.transAxes) # 期間と作成日 ax.text(0.5, 0.45, ‘Period: April – June 2024′, ha=’center’, fontsize=18, transform=ax.transAxes) ax.text(0.5, 0.4, ‘Created: July 1, 2024′, ha=’center’, fontsize=16, transform=ax.transAxes) # 会社名 ax.text(0.5, 0.2, ‘ABC Corporation’, ha=’center’, fontsize=20, transform=ax.transAxes) ax.text(0.5, 0.15, ‘Marketing Department’, ha=’center’, fontsize=16, transform=ax.transAxes) pdf.savefig(fig, bbox_inches=’tight’) plt.close() # ========== ページ2: エグゼクティブサマリー ========== fig, ax = plt.subplots(figsize=(11, 8.5)) ax.axis(‘off’) summary_text = “”” 【Executive Summary】 1. Campaign Overview – Platforms: Instagram & Facebook – Duration: 3 months (Apr-Jun 2024) – Budget: 2,000,000 JPY – Target: Women 20-35 2. Key Results ✓ Visitors: +50% (Target: 107%) ✓ Conversions: +140% (Target: 120%) ✓ ROI: 250% (Target: 200%) ✓ Revenue Contribution: 5,000,000 JPY 3. Success Factors – Precise targeting (20s women) – Effective video creatives – Weekly optimization 4. Recommendations – Increase budget to 3,000,000 JPY (1.5x) – Expand target to 30-40 age group – Start YouTube advertising 【Conclusion】 The campaign exceeded expectations with 250% ROI. We strongly recommend continuation and expansion. “”” ax.text(0.05, 0.95, summary_text, va=’top’, fontsize=11, family=’monospace’, transform=ax.transAxes) pdf.savefig(fig, bbox_inches=’tight’) plt.close() # ========== ページ3: Before/After比較 ========== fig, axes = plt.subplots(2, 2, figsize=(11, 8.5)) metrics = [‘Visitors’, ‘Conversions’, ‘CVR (%)’, ‘Revenue (10K)’] before = [10000, 50, 0.5, 500] after = [15000, 120, 0.8, 1200] for i, (metric, b, a) in enumerate(zip(metrics, before, after)): row, col = i // 2, i % 2 ax = axes[row, col] x = [‘Before’, ‘After’] y = [b, a] colors = [‘#E74C3C’, ‘#2ECC71′] bars = ax.bar(x, y, color=colors, alpha=0.7) ax.set_title(metric, fontsize=14, fontweight=’bold’) # 増加率を計算して表示 increase = ((a – b) / b * 100) ax.text(0.5, max(y) * 0.85, f’+{increase:.0f}%’, ha=’center’, fontsize=12, fontweight=’bold’, bbox=dict(boxstyle=’round’, facecolor=’yellow’)) plt.suptitle(‘Before/After Comparison’, fontsize=16, fontweight=’bold’) plt.tight_layout() pdf.savefig(fig, bbox_inches=’tight’) plt.close() # PDF情報を設定 d = pdf.infodict() d[‘Title’] = ‘SNS Campaign Report’ d[‘Author’] = ‘Marketing Department’ d[‘Subject’] = ‘Q2 2024 Campaign Results’ d[‘Keywords’] = ‘SNS, ROI, Marketing’ d[‘CreationDate’] = pd.Timestamp.now() print(f”\n✓ PDFレポートを作成しました: {pdf_filename}”) print(” このファイルを経営陣や関係者に共有できます！”) # 実行 create_marketing_report()

# シナリオ分析の可視化 import plotly.graph_objects as go from plotly.subplots import make_subplots # 3つのシナリオを定義 scenarios = [‘現状維持’, ‘提案A\n（予算1.5倍）’, ‘提案B\n（予算2倍）’] budgets = [200, 300, 400] # 予算（万円） expected_sales = [500, 750, 1000] # 予想売上（万円） expected_profit = [300, 450, 600] # 予想利益（万円） roi = [250, 250, 250] # ROI（%）は同じと仮定 # 2つのグラフを横に並べる fig = make_subplots( rows=1, cols=2, subplot_titles=[‘予算と売上の比較’, ‘ROI（投資対効果）’], specs=[[{“type”: “bar”}, {“type”: “scatter”}]] ) # 左: 予算と売上の棒グラフ fig.add_trace(go.Bar( x=scenarios, y=budgets, name=’広告予算’, marker_color=’#E74C3C’, text=budgets, textposition=’outside’ ), row=1, col=1) fig.add_trace(go.Bar( x=scenarios, y=expected_sales, name=’予想売上’, marker_color=’#2ECC71′, text=expected_sales, textposition=’outside’ ), row=1, col=1) # 右: ROIの折れ線グラフ fig.add_trace(go.Scatter( x=scenarios, y=roi, mode=’lines+markers+text’, name=’ROI（%）’, line=dict(color=’#3498DB’, width=3), marker=dict(size=15), text=[f'{r}%’ for r in roi], textposition=’top center’ ), row=1, col=2) # レイアウト設定 fig.update_layout( title_text=’予算増額シナリオ分析‘, height=500, showlegend=True ) fig.update_yaxes(title_text=’金額（万円）’, row=1, col=1) fig.update_yaxes(title_text=’ROI（%）’, range=[0, 300], row=1, col=2) fig.show() # 提案を表示 print(“\n【提案】”) print(“=” * 50) print(“提案A（予算1.5倍: 300万円）を推奨します。”) print(“-” * 50) print(f”・追加投資: +100万円”) print(f”・追加売上: +250万円（予想）”) print(f”・追加利益: +150万円（予想）”) print(f”・ROI: 250%を維持”) print(“=” * 50) print(“\n理由: ROIを維持しながら、リスクを抑えた成長が可能。”) print(“提案Bは効果が高いが、リスクも大きいため、”) print(“まずは提案Aで効果を確認後、段階的に拡大を推奨。”)