This article is concerned with the transition of a longitudinal survey from a single-mode design to a web-first mixed-mode design and the role that text messages to sample members can play in smoothing that transition. We present the results of an experiment that investigates the effects of augmenting the contact strategy of letters and emails with text messages, inviting the sample members to complete a web questionnaire and reminding them of the invite. The experiment was conducted in a subsample of Understanding Society, a household panel survey in the United Kingdom, in the wave that transitioned from a CAPI-only design to a sequential design combining web and CATI. In the experiment, a quarter of the sample received letters and emails, while the rest received between one and three text messages with a personalized link to the questionnaire. We examine the effect of the text messages on response rates, both at the web phase of a sequential design and at the end of the fieldwork after a CATI follow-up phase, and explore various mechanisms that might drive the increase in response rates. We also look at the effects on the device used to complete the survey and field efforts needed at the CATI stage. The findings indicate that text messages did not help to significantly increase response rates overall, although some subgroups benefited from them, such as panel members who had not provided an email or postal address before. Likewise, the text messages increased web completion among younger panel members and those with an irregular response pattern. We only found a slight and nonsignificant effect on smartphone use and no effect on the web household response rate, a proxy for fieldwork efforts.