Profil LinkedIn Analyste de Données Prompt

I notice that no Excel file is actually attached to this conversation yet. Before I can analyze anything, you'll need to upload it here. However, here is the step-by-step process I will follow **once the file is available**: **Step 1 - Data Ingestion & Validation** - Load the Excel file and inspect all sheets, columns, and row counts - Check data types, missing values, duplicates, and formatting issues - Validate date ranges, engagement metrics, and any text fields **Step 2 - Metric Definition & KPI Selection** - Identify which metrics are available (impressions, clicks, reactions, comments, shares, profile visits, follower growth, CTR, engagement rate, etc.) - Define your primary KPI(s) based on your goal (brand awareness vs. leads vs. thought leadership) - Flag any metric that lacks sufficient data for statistical confidence **Step 3 - Temporal Analysis** - Plot engagement over time (by day, week, month) - Detect seasonality, weekday/weekend patterns, time-of-day effects - Identify sudden spikes or drops linked to external factors **Step 4 - Content Pattern Analysis** - Parse post text for length, hashtags, mentions, emojis, CTA types, question vs. statement style - Classify posts by format (text-only, image, video, carousel, document, link) and topic - Compute average performance per category and test for statistically significant differences **Step 5 - Audience Insight Extraction** - Analyze which content resonates with which audience segments (if demographic data is available) - Examine engagement-to-impression ratios to detect both reach and resonance issues **Step 6 - Correlation & Multivariate Insights** - Run correlations between features (length, media type, time, hashtags) and your KPI(s) - Use simple regression or decision-tree logic to rank which factors matter most - Control for collinearity (e.g., video posts may naturally get more impressions) **Step 7 - Benchmarking & Gap Analysis** - Compare your worst vs. best performing quartile of posts - Identify the specific characteristics that separate them - Estimate realistic performance uplift if you adopted the top-post features **Step 8 - Actionable Recommendations** - Deliver a prioritized list of concrete changes (content mix, posting schedule, format choices, copy style, CTA strategy) - For each recommendation, include expected impact, confidence level, and rationale based on your data **Step 9 - Measurement Plan** - Propose a follow-up testing framework (e.g., A/B style comparison over next 2-4 weeks) - Define which metrics to track, how to normalize them (per-follower, per-impression), and how to determine if changes worked **Step 10 - Iterative Feedback Loop** - After you implement changes, schedule a re-analysis of new data to validate or refine the recommendations - Update the model as more data accumulates --- **Please upload the Excel file now, and I will begin executing this process immediately.** If the file is too large, you can share it via Google Sheets link or drop it into the chat directly.