**Objectives** To identify clinical indicators reflecting potentially treatment-modifiable prognostic factors associated with pain recovery in patients with non-specific neck pain and to develop a prototype clinical compass for visualising patient-specific profiles.
**Design** Secondary analysis of data from a multicentre, prospective, observational prognostic cohort study, with a 6-month follow-up between January 2020 and March 2023.
**Setting** 30 physiotherapy primary care practices.
**Participants** Patients with a new episode of non-specific neck pain.
**Primary and secondary outcome measures** Patients completed questionnaires at baseline, 6 weeks, 3 months and 6 months assessing pain intensity and biological, psychological and social factors. Questionnaire total scores and individual items reflecting selected potentially modifiable factors were tested separately as candidate indicators. Baseline and repeated indicators were examined as effect modifiers of pain trajectories using growth-curve models. Subsequently, modifiable clinical indicators were incorporated as effect modifiers to evaluate their influence on the trajectory of pain intensity over time. Non-redundant significant indicators were integrated into a prototype clinical compass.
**Results** Data from 603 participants were collected. A quadratic growth-curve model best explained the longitudinal change of pain intensity. 24 clinical indicators were significantly associated with the trajectory of pain intensity over time. Two characteristics were excluded as they demonstrated high intercorrelation. Based on these findings, a clinical compass comprising 22 clinical indicators was developed to visualise patient-specific profiles.
**Conclusions** This study provides a prototype clinical compass that may inform prognostic reasoning in primary care physiotherapy, but further evaluation is needed before clinical implementation.
