An attribution model in Google Ads is the rule or set of rules that decides which touchpoints in a customer's conversion path receive credit for the sale or conversion. When someone clicks your ad today but doesn't convert until they click again three days later, the attribution model determines whether the first click, last click, or both get credit. Google Ads provides six standard attribution models. Last-click gives 100% credit to the final ad interaction before conversion, which is Google's historical default but often undercredits top-of-funnel campaigns. First-click does the opposite, crediting the initial touchpoint entirely. Linear splits credit evenly across all clicks in the path. Time-decay gives more weight to interactions closer to the conversion time. Position-based (also called U-shaped) assigns 40% to the first interaction, 40% to the last, and splits the remaining 20% among middle touches. Data-driven attribution uses machine learning to assign credit based on actual conversion likelihood of each touchpoint, but requires at least 3,000 ad interactions and 300 conversions in 30 days to activate. Your choice directly impacts which campaigns appear profitable in reporting. If you run both brand and generic search campaigns, last-click will make your brand campaign look like a hero because people often search your brand name right before converting, even if a generic search ad introduced them days earlier. Linear or data-driven models surface the true contribution of prospecting efforts. At Ottawa SEO, we typically start accounts on last-click because it matches how most clients mentally track ROI, then migrate to data-driven once the account hits Google's threshold. For smaller accounts under 300 monthly conversions, position-based is the best compromise since it credits both discovery and closing interactions. The model you choose should align with your actual sales cycle length and how your team evaluates campaign performance. Changing models doesn't affect actual delivery or bidding unless you're using Target CPA or Target ROAS, which will re-optimize based on the new credit distribution.