Hypothetical bias is a common research problem in measuring intentions. This study examines differences in individuals' hypothetical and nonhypothetical willingness to pay (WTP) based on purchase intention level. The purpose of Study 1 is to see if hypothetical bias is affecting all groups of individuals when segmented by purchase intentions. The contingent valuation method (CVM) was used to assess to individuals' purchase intentions and hypothetical WTP for an environmentally friendly (EF) offering. Individuals then participated in an auction that required them to bid on the same offering, thus requiring an actual monetary commitment. Through segmenting individuals into three purchase intention groups (high, moderate, and low), WTP biases were found to be positively related to stated intention level. High and moderate purchase intenders significantly overstated their hypothetical WTP. However, low purchase intenders did not. Further, no significant differences existed in nonhypothetical WTP between the three groups. The price that low purchase intenders were willing to pay through the CVM represented closely what they and the rest of the market were willing to pay in practice. The purpose of Study 2 was to examine several reasons why higher purchase intenders may be overstating their hypothetical WTP, with normative pressure and social desirability bias taken into consideration. Individuals indicated their purchase intentions and hypothetical WTP for an EF and a non‐EF product, and then participated in an auction. The higher purchase intenders again overstated their actual WTP for both products, whereas low purchase intenders did not. Results suggest that both normative pressure and social desirability bias contribute to hypothetical bias for products that carry a normative dimension (EF products). Controlling for these two factors eliminated the differences between hypothetical and nonhypothetical WTP. For products that did not have a normative component (non‐EF products), controlling for social desirability eliminated the gap between the hypothetical and nonhypothetical WTP amounts. A main implication from this research is that hypothetical bias is not a universal phenomenon and does not operate equally in all groups of purchase intenders. Marketers may want to consider only using lower purchase intenders in their pricing estimates for new products, especially for those that carry a normative component. Further, the CVM has received much criticism for measuring hypothetical WTP. However, estimates produced by the CVM were nearly exact for the low purchase intender group and were reliable when controlling for social desirability bias and normative pressure for the higher purchase intenders.