Use Case: Performance Media Agency Post-Buy Audit

Persona: Media Director, Quantitative Media Planner, Performance Agency Lead
Objective: Deliver a rigorous, transparent post-buy campaign audit deck to your brand client within 48 hours of flight completion, identify underperforming stations/dayparts, and reallocate the next flight's budget for maximum visitor response.

---

The Challenge

Agency clients expect clear attribution answers after investing €200k–€2M in linear broadcast:

1. Which stations delivered the lowest Cost Per Incremental Visitor (CPIV)?
2. Did prime-time inventory justify its 4x rate card premium over daytime fringe?
3. How much attribution distortion occurred from overlapping spots airing at the same minute across competing networks?

Legacy SaaS platforms charge €4,000/month, demand raw telemetry access, and deliver black-box scores. SpotLift allows your agency to run 100% in-house, inspectable audits in Python or via CLI.

---

4-Step Audit Workflow

Step 1: Pre-Flight Data Quality Audit

Before presenting findings to clients, verify that client broadcast logs and GA4 telemetry are cleanly synchronized:

spotlift validate \
  --spots client_schaltplan.csv \
  --web ga4_minute_sessions.parquet \
  --tz Europe/Berlin

What this checks:

• Ensures airings in local station time (e.g. CET/CEST) are accurately aligned with UTC GA4 telemetry.
• Detects unrecorded minute gaps that could invalidate response integrals.
• Quantifies prime-time collision density across competing stations.

---

Step 2: Open-Core Diagnostic Attribution

Run the diurnal baseline attribution pass and generate a stakeholder Markdown summary report:

spotlift run \
  --spots client_schaltplan.csv \
  --web ga4_minute_sessions.parquet \
  --tz Europe/Berlin \
  --format all \
  --out audit_results.csv

This generates:

1. audit_results.csv: Spot-by-spot attributed records with baseline, observed, signed lift, and CPIV.
2. audit_results.md: Executive summary table with channel efficiency rankings and collision diagnostics.

---

Step 3: Station Efficiency Ranking & Daypart Pruning

In Python, inspect station-level performance and identify daypart efficiency outliers:

import spotlift as sl
import pandas as pd

# Load validated data
airings = sl.load_airings("client_schaltplan.csv", tz="Europe/Berlin")
telemetry = sl.load_telemetry("ga4_minute_sessions.parquet")

# Attribute with continuous weighted-share collision handling
results = sl.analyze_spots(airings=airings, web_telemetry=telemetry, window_minutes=8)

# Station efficiency scorecard
station_scorecard = (
    results.groupby("channel")
    .agg(
        total_spots=("spot_id", "count"),
        media_spend=("cost", "sum"),
        total_grp=("grp", "sum"),
        net_lift=("incremental_lift", "sum"),
        collision_count=("is_overlapping", "sum"),
    )
    .assign(
        cpiv=lambda df: df["media_spend"] / df["net_lift"],
        collision_pct=lambda df: (df["collision_count"] / df["total_spots"]) * 100.0,
    )
    .sort_values("cpiv")
)

print(station_scorecard)

Sample Output Scorecard:

Channel	Spots	Media Spend	Total GRP	Net Lift	CPIV (€)	Collision Rate	Recommendation
Joyn (CTV)	150	€210,150	374.5	+26,325	€7.98	56.7%	🟢 Scale (+25% budget)
VOX	159	€349,970	398.2	+31,859	€10.98	50.9%	🟢 Efficient (Maintain)
ProSieben	132	€437,580	314.9	+32,445	€13.49	52.3%	⚪ Benchmark Average
RTL	159	€625,210	411.9	+44,696	€13.99	52.8%	🟡 Prune Late Fringe

---

Step 4: White-Label Client Deliverables (SpotLift Pro)

When presenting to the CMO or brand steering committee, generate interactive client HTML dashboards and multi-tab Excel workbooks branded with your agency colors:

pro_results = sl.analyze_omnichannel_pro(
    airings=airings,
    web_telemetry=telemetry,
    export_html="reports/q1_tv_postbuy_deck.html",
    export_excel="reports/q1_tv_postbuy_workbook.xlsx",
    white_label_config={
        "agency_name": "Apex Media Performance",
        "primary_color_hex": "#1E3A8A",
        "logo_url": "https://agency.com/assets/logo.png"
    }
)

The resulting workbook includes dedicated tabs for:

• Executive Summary & Overall Campaign CPIV.
• Station & Daypart Efficiency Matrices.
• Creative Wear-Out Decay Regressions.
• Prime-Time EM Deconvolution breakdown.