Google Analytics 4 uses data-driven attribution as its default model. This means GA4 analyzes your actual conversion paths and assigns fractional credit to each touchpoint based on how much it statistically contributed to the conversion. If you don't have enough data for the algorithm to work (typically under 400 conversions per month for the conversion action), GA4 falls back to a last-click model. Universal Analytics, which stopped collecting data in July 2023, defaulted to last-click attribution in standard reports. You could change this in the Model Comparison Tool to first-click, linear, time-decay, or position-based, but most businesses never touched it and ran on last-click without realizing it. The shift to data-driven matters because last-click systematically undervalues upper-funnel activities. If someone finds you through organic search, comes back via email, then converts on a branded PPC ad, last-click gives 100% credit to that final ad. Data-driven might assign 40% to the original organic visit, 20% to email, and 40% to the ad based on how users with similar paths actually behaved. At Ottawa SEO, we see this play out constantly in multi-touch SEO campaigns. A blog post that assists conversions weeks later gets zero credit under last-click, making content marketing look less effective than it is. Data-driven fixes this, but only if you have conversion volume. For smaller sites under that 400-conversion threshold, you're stuck with last-click math even in GA4. One gotcha: GA4's attribution models only apply to key events you've marked as conversions. Session-based metrics and standard reports don't use attribution logic at all. You need to specifically look at the Advertising workspace's attribution reports to see the data-driven model in action. Many site owners think they're getting sophisticated attribution across all GA4 reports when they're actually just seeing last-interaction data in most views.