There is a notable lack of evidence on what constitutes good practice in remote quantitative data collection from research participants with dementia. During the COVID-19 pandemic face-to-face research became problematic, especially where participants were older and more at risk of infection. The DETERMIND-C19 study, a large cohort study of people with dementia, switched to telephone data collection over this period. This paper explores the experiences of researchers who collected quantitative data over the telephone from people with dementia during the first COVID-19 lockdowns in England. The aim was to learn from these experiences, share insights and inform future research practice across disciplines. Seven DETERMIND researchers were interviewed about the processes and challenges of collecting quantitative data from people with dementia over the telephone compared to face-to-face. Data were analysed using reflexive thematic analysis. Two themes were developed: first the telephone adds an extra layer of confusion to an already cognitively complex interaction. Second, researchers found it difficult to recognise subtle cues that signalled participants’ rising emotion over the telephone in time to prevent distress. The researchers employed strategies to support participants which may not have conformed to the strict conventions of structured interviewing, but which were informed by person-oriented principles. Whilst in practice this may be a common approach to balancing the needs of participants and the requirements of quantitative research, it is rare for studies to openly discuss such trade-offs in the literature. Honest, reflective reporting is required if the practice of remote data collection from people with dementia is to progress ethically and with integrity.