CoolerTracker

Your core competitive advantage is helping players cut through variance confusion with quantified, trustworthy scores. Three specific issues are currently undermining that promise. First, fix your 4.5-second First Contentful Paint immediately. Poker players are decision-makers in a high-frequency mindset—they load CoolerTracker to *instantly* see whether their session was unlucky or unskilled. A blank screen for 4.5 seconds breaks that urgency and triggers tab-closing. Likely culprits: unoptimized PT4 data fetching, heavy JavaScript bundles, or render-blocking resources. Lazy-load non-critical data, defer JavaScript parsing, and prioritize rendering your Variance Score widget first. Target under 1.5 seconds to match player expectations. Second, audit your two unnamed interactive elements and assign descriptive aria-labels immediately. If these buttons control data filtering (e.g., 'Filter by tournament type' or 'Sort by cooler frequency'), blind users cannot access them—and that's both an accessibility failure and a missed opportunity, since serious poker players often use assistive tech. Third, audit all text under 12px and bump any explanatory content (luck score methodology, variance metric definitions, performance breakdowns) to at least 14px. Small text might save screen real estate, but it directly erodes trust in your analysis—users need to *comfortably* read why the algorithm says they got unlucky. Prioritize these in order: performance fix first (impacts all users, trial-to-paid), then accessibility (legal + inclusivity), then readability (reduces friction in score interpretation). Together, they'll ensure your Variance and Luck scores land with impact, not confusion.