Somewhere around the fourth question, a conversation stops feeling like a conversation and starts feeling like an interrogation.
The four-question threshold
One question at a time works well for a short qualification: what do you need, where are you, when do you need it. It reads as a conversation because it is one.
Past four fields the drop-off becomes obvious in the data. Every message is another chance to get distracted, and there is no sense of progress because the customer cannot see how many questions remain.
What a Flow gives you instead
- A real form UI: text fields, dropdowns, date pickers, multiple screens
- Visible progress, so people know the end is in sight
- Conditional logic that shows fields only when relevant
- Calculations between fields, such as a running quote total
- Dropdown options fetched live from your own API
- One submit, with everything validated together
Where the conversational bot still wins
Short interactions, anything where the answer determines the next question in a genuinely branching way, and anything that benefits from feeling human rather than transactional.
A bot is also better when you want to capture people who will not commit to a form: a single question gets answered by people who would have abandoned a six-field screen.
Using both in one journey
- Open with one conversational question to establish intent.
- If they qualify, send the Flow to collect the structured detail.
- Store the submitted fields against the contact as CRM fields.
- Return to conversation for confirmation and next steps.
- Route to a human where the answers warrant it.
Where the data ends up
Flow submissions land as fields on the CRM contact, which means you can segment and automate on them immediately, and optionally into a Google Sheet for teams that work there.
That matters more than it sounds: data captured into a chat transcript is not data you can build a campaign from. Data captured into fields is.