Attribution AI Prompts

Step 1: Data audit - audit the marketing analytics stack for tracking completeness. Identify missing events, broken tracking, and attribution gaps. Produce a prioritized list of data quality fixes. Step 2: Performance baseline - establish current performance baselines for all key marketing metrics by channel. Compute YoY and MoM trends. Identify which channels are improving and which are declining. Step 3: Attribution analysis - compare performance across at least three attribution models (last click, first click, linear). Identify which channels are over- or under-credited in the current reporting model. Recommend a more accurate approach. Step 4: Audience analysis - segment customers into actionable groups using RFM or behavioral clustering. Compute LTV, conversion rate, and CAC by segment. Identify the highest-value underserved segment. Step 5: Campaign optimization - for each active channel, identify the top optimization opportunity (budget reallocation, audience refinement, creative refresh, landing page improvement). Prioritize by expected revenue impact. Step 6: Content and organic strategy - audit SEO performance and content effectiveness. Identify top 10 keyword and content opportunities. Build a 90-day content roadmap prioritized by traffic and conversion potential. Step 7: Marketing plan and measurement - produce a 90-day marketing plan: budget allocation by channel, key initiatives, expected outcomes, and a measurement framework with clear success criteria. Include one incrementality test to run in the quarter.

Design an incrementality test to measure the true causal impact of a marketing channel. Channel to test: {{channel}} (e.g. Facebook Ads, email retargeting, TV) Primary metric: {{metric}} (conversions, revenue) Business context: {{context}} Incrementality testing answers: what revenue would we have generated WITHOUT this channel? This is the only way to determine true incremental value beyond what would have happened anyway. 1. Test design options: Geo holdout test: - Split geography into test regions (channel active) and holdout regions (channel paused) - Match regions by: historical performance, demographics, market size - Duration: minimum 4 weeks (longer for lower-frequency purchase categories) - Measure: conversion rate in test regions vs holdout regions User holdout (ghost bidding): - Randomly assign users: treatment (see ads) vs control (ad slot left empty) - Platform-native holdout if available (Facebook Brand Lift, Google Conversion Lift) - Best for: digital channels with user-level targeting Time-based holdout: - Turn off the channel for a period and compare to matched prior period - Weakness: seasonal and macro factors can confound results - Requires: careful selection of the comparison period 2. Sample size and duration: - Required sample: power calculation based on baseline conversion rate and expected lift - Minimum detectable effect: if the channel drives < {{mde}}% lift, you need more data - Duration: long enough to capture a full purchase cycle 3. Measurement: - Lift = (Conversion Rate_test - Conversion Rate_holdout) / Conversion Rate_holdout - Incremental conversions = Lift x Holdout user volume - True CPA = Channel Spend / Incremental Conversions - Compare to attributed CPA: the gap shows how much attribution was overstating the channel 4. Common pitfalls: - Spillover: people in the holdout region still see the ads via other means - Holdout contamination: test and control groups interact (e.g. via social sharing) - Too short a test: brand campaigns need months to show full effect Return: test design recommendation, sample size calculation, measurement plan, and common pitfall mitigations.

Design and interpret a marketing mix model (MMM) for this business. Business: {{business}} Sales / conversion data: {{sales_data}} (weekly, 2+ years) Marketing spend data: {{spend_data}} (by channel, same period) External factors: {{external_factors}} (macroeconomic data, seasonality, competitor actions) 1. What MMM is and when to use it: - MMM uses regression to decompose total sales into: baseline (organic) + each marketing channel's contribution + external factors - Unlike attribution (which tracks individual user paths), MMM works at aggregate level - Best for: understanding the incremental contribution of each channel, including offline (TV, OOH) - Limitation: requires 2+ years of data, is backward-looking, and cannot measure within-campaign personalization 2. Data preparation: - Adstock transformation: marketing spend has a delayed and decaying effect Adstock_t = Spend_t + decay_rate x Adstock_{t-1} - Decay rates by channel: TV (~0.7), Digital display (~0.3), Paid search (~0.1) - Saturation curve: diminishing returns on increasing spend (log or S-curve transformation) - Control variables: seasonality (Fourier terms or dummy variables), price, distribution, promotions 3. Model specification: Sales_t = Baseline + sum(beta_i x Adstock_i_t) + beta_price x Price_t + Seasonal + Error - Estimate using Bayesian regression (allows priors on channel effectiveness) - Model diagnostics: R-squared, MAPE on holdout period, residual checks 4. Output interpretation: - Baseline %: share of sales occurring without any marketing - Contribution % per channel: what share of incremental sales each channel drove - mROAS (marginal ROAS): the return on the last dollar spent in each channel - Saturation point: at what spend level does each channel show diminishing returns? 5. Budget optimization: - Using the fitted saturation curves: what budget allocation maximizes total sales at the current total budget? - What is the revenue uplift from the optimal allocation vs current allocation? Return: MMM methodology explanation, data preparation steps, model output interpretation, contribution table, mROAS by channel, and optimized budget allocation.

Analyze marketing attribution using multiple models to understand channel contribution. Customer journey data: {{journey_data}} (user_id, touchpoint, timestamp, channel, converted: Y/N) Conversion event: {{conversion_event}} Lookback window: {{lookback_window}} days 1. Attribution model comparison: Build all five standard models and compare: Last click: 100% credit to the final touchpoint First click: 100% credit to the first touchpoint Linear: equal credit to all touchpoints in the path Time decay: more credit to recent touchpoints (decay factor: 0.5 per day) Position-based (U-shaped): 40% to first, 40% to last, 20% split across middle For each model: credits per channel (absolute and %). 2. Channel contribution shift across models: - Which channels gain the most credit in first-click vs last-click? - First-click favors awareness channels (social, display); last-click favors intent channels (search) - Create a heatmap: channels x attribution models showing credit share 3. Path analysis: - Top 10 converting path sequences (e.g. Display → Organic Search → Paid Search → Convert) - Average path length (touchpoints) for converting vs non-converting journeys - Most common first touchpoints for converters 4. Assisted vs direct conversions: - % of conversions with only one touchpoint vs multi-touch paths - Channels that primarily assist (appear in paths but not at end) vs channels that primarily close 5. Data-driven attribution (if sample size permits): - Shapley value attribution: fair allocation based on marginal contribution of each channel - Requires: enough paths where each channel appears with and without others - Provides the most theoretically correct attribution 6. Budget implication: - If we moved budget based on last-click attribution: which channels would be over or under-invested? - Recommended budget allocation change based on the most appropriate model Return: model comparison table, path analysis, assisted vs direct breakdown, and budget implications.