The use of targeted advertisements on social media platforms (eg, Facebook and Instagram) has become increasingly popular for recruiting participants for online survey research. Many of these surveys offer monetary incentives for survey completion in the form of gift cards; however, little is known about whether the incentive amount impacts the cost, speed, and quality of data collection. This experiment addresses this gap in the literature by examining how different incentives in paid advertising campaigns on Instagram for completing a 10-minute online survey influence the response rate, recruitment advertising cost, data quality, and length of data collection. This experiment tested three incentive conditions using three Instagram campaigns that were each allocated a US $1400 budget to spend over a maximum of 4 days; ads targeted users aged 15-24 years in three nonadjacent designated market areas of similar size to avoid overlapping audiences. Four ad creatives were designed for each campaign; all ads featured the same images and text, but the incentive amount varied: no incentive, US $5 gift card, and US $15 gift card. All ads had a clickable link that directed users to an eligibility screener and a 10-minute online survey, if eligible. Each campaign ran for either the full allotted time (4 days) or until there were 150 total survey completes, prior to data quality checks for fraud. The US $15 incentive condition resulted in the quickest and cheapest data collection, requiring 17 hours and ad spending of US $338.64 to achieve 142 survey completes. The US $5 condition took more than twice as long (39 hours) and cost US $864.33 in ad spending to achieve 148 survey completes. The no-incentive condition ran for 60 hours, spending nearly the full budget (US $1398.23), and achieved only 24 survey completes. The US $15 and US $5 incentive conditions had similar levels of fraudulent respondents, whereas the no-incentive condition had no fraudulent respondents. The completion rate for the US $15 and US $5 incentive conditions were 93.4% (155/166) and 89.8% (149/166), respectively, while the completion rate for the no-incentive condition was 43.6% (24/55). Overall, we found that a higher incentive resulted in quicker data collection, less money spent on ads, and higher response rates, despite some fraudulent cases that had to be dropped from the sample. However, when considering the total incentive amounts in addition to the ad spending, a US $5 incentive appeared to be the most cost-effective data collection option. Other costs associated with running a campaign for a longer period should also be considered. A longer experiment is warranted to determine whether fraud varies over time across conditions.