Respondents will film their fridge. They will not negotiate an app store login after a long shift. Asking people to install an app before they can join a study sounds small, but in practice it is one of the largest causes of dropout. Skipping the install is not a nice-to-have. It is methodology protection, because a study only produces insight from people who actually finish.
What an app install really asks of someone
- 1Click the invitation and land on a store page.
- 2Find enough phone storage to download something unfamiliar.
- 3Sign in to a store account and wait for the install to finish.
- 4Open the new app, create another account, and approve permissions.
- 5Finally, find the study and begin the first task.
Every step is a chance to quit. For participants who are not professional research panelists, such as busy parents, older adults, or people on work phones, the odds of getting through all five steps are much lower than clicking a link.
Why this hurts research, not just numbers
Dropout is rarely random. People who are less technical, more time-pressed, or using older phones leave first. Your remaining sample then over-represents the patient and the tech-comfortable, which is a form of selection bias. You may report “consumers” when you actually heard from “consumers who tolerate research apps.”
- Lower completion means smaller samples and higher cost per useful respondent.
- Biased dropout skews who you hear from.
- Recruiters and clients lose confidence when invited respondents never start.
- Day 1 problems can spoil a multi-day study before it begins.
When an app is still justified
There are cases where an app is the right tool: background or passive data capture, offline recording in places with no signal, or highly committed panels who use the app regularly. For a typical multi-day market research board with consumer samples, those needs are rare, and browser-based capture does the job.
What “app-like” should still include
- Progress that makes returning obvious
- Camera and mic capture in the browser
- Support without leaving the study
- Fast load on mid-range Android phones
- Clear, mobile-friendly layouts that need no pinching or zooming
No-install does not mean low quality. A modern board can feel like an app, with large buttons, easy media capture, and quick page loads, while staying a simple link.
How to check a platform’s respondent experience
- 1Open the invitation on a phone with a weak connection, not on your office laptop.
- 2Count the taps from link to first question.
- 3Try a camera task and deny the permission, then see how the product helps you recover.
- 4Return on Day 2 from a reminder and see whether you land where you left off.
Example: a study recruits 40 shift workers for a five-day board. Half have older phones with limited storage. With a link-based board they all start on Day 1 from a text message. With an app requirement, the team would have lost some before the first question.
FAQ