Although angel investors and venture capitalists (VCs) both participate in the supply side of the same market providing capital and advice to early-stage companies, they are distinct in several ways. The differences in when they choose to deploy capital are well studied. The differences in when they choose to provide advice are not. Using a sample of 7,980 mentoring decisions by seed-stage investors from which I construct a novel typology of four types of startup activities, I report among the first empirical findings on systematic differences in angel versus VC advice. Angels are more likely than VCs to choose to provide advice on the design and execution of experiments (e.g., will customers buy this product at this price?), whereas angels are less likely than VCs to choose to provide advice on analysis (e.g., what is the size of the total addressable market?). While analysis is a skill that can be learned from studying, hypothesis testing is a skill developed via learning-by-doing. I report evidence consistent with the hypothesis that angels are more likely to choose to provide advice on the design and execution of experiments because they have a skill advantage in that domain due to operational experience. I also provide evidence that is inconsistent with alternative explanations, including financial incentives and selection.