Pain intensity is an important outcome in clinical trials of surgery because pain relief is important to patients. Currently, recommended scales are the numeric rating scale 0-10 and visual analogue scale. However, these scales allow for considerable influence of individual imagination, previous experience, and coping skills, limiting proficiency in comparative clinical trials. We aimed to explore postoperative expressions of "how much it hurts"-the first step to improve pain intensity measurement. This was a qualitative study using inductive content analysis: words and visual cues describing pain intensity were collected from (i) existing pain intensity measures by search of COSMIN, PubMed, and Google, (ii) patient interviews recorded and transcribed word-for-word, (iii) clinician interviews transcribed likewise, and (iv) 100 patient telephone interviews with notes taken. After familiarization, the collected expressions were labelled inductively in categories and assembled in tables (case and theme-based matrices). Descriptors fell into 12 categories: intensity (slight/strong), evaluative (negligible/unbearable), cognitive impact (distracting/can be ignored), activity impact (limits some/all activity), sleep impact (can/cannot sleep), examples (like stubbing a toe), physical signs (crying/writhing), associated symptoms (nauseating/tiring), treatment (ice helps/need morphine), affective (annoying/dreadful), discriminative (aching/piercing), and general recovery (hindering recovery/functional interference). Many visual cues were also identified. Literature and recorded interviews gave rise to the categories, and telephone interviews found saturation, providing no further categories. Pain intensity is expressed by terms that fall into 12 categories and by a variety of graphic elements. This advances development of a patient-reported outcome measure of pain intensity for orthopedic trials.