Analyze return reasons and cut my product return rate

Help me analyze my product returns and reduce the return rate. Return data: [PASTE: order ID or SKU, category, size/variant, return reason code, customer comment, days to return, refund or exchange, channel] Return rate overall and by category: [%] Return policy: [WINDOW, free or paid returns, exchanges offered] Cost per return: [label, processing, % resold at full price, % liquidated or destroyed - or [ADD]] Do this: 1. Rank SKUs and categories by return rate and by total return cost, not just by count. 2. Group reasons into fixable themes: fit or sizing, not as pictured or described, quality or defect, damaged in transit, late delivery, changed mind, and ordering several sizes to keep one (bracketing). 3. Read the customer comments for patterns the reason codes miss (e.g. "runs small in the shoulders", "color darker than the photo"). 4. Note possible policy-abuse patterns (serial returners, worn-and-returned items) as signals to review, not accusations. 5. Recommend fixes by theme: size guides and fit notes, better photos and color accuracy, description corrections, packaging changes, vendor quality claims, exchange-first return flows, or policy changes. 6. Estimate savings if the top three fixes reduce returns on affected SKUs by an assumed amount; state the assumption. Return: - A short summary for leadership. - A SKU table: SKU, return rate, top reason, return cost, recommended fix. - A 60-day test plan with how we'll measure each fix. Use only my data and mark gaps rather than guessing.