In an increasingly competitive market, where affective values are key for differentiation, understanding user opinions is crucial to ensuring success in product design. Traditionally, designers have collected data on product evaluation using self-report questionnaires, such as Osgood’s Semantic Differential technique, a widely used method for assessing product aesthetics and affect. However, some authors question the temporal stability of responses, suggesting users may not be consistent in their opinions over time. This study presents two case studies where a group of volunteers evaluated two types of products (chair and watering can) using the Semantic Differential technique in separate sessions. Data on response confidence was also collected. The main objective was to determine if statistically significant differences exist between measurements to assess the stability of responses and the reliability of self-report methods over time. Results showed that some attributes were affected by the temporal factor for both products, with a consistency rate of 91.67% for the chair and between 62.50 and 50.00% for watering cans. Our results show that consistency in user opinion over time may depend on the selected product type, highlighting the importance of considering product typology when selecting the evaluation tool, as self-report questionnaires may be suitable for well-known product typologies (i.e., chairs) or certain product attributes.