A bot that actually helps.

An FAQ bot answers the easy ones. A helpful bot feeds on your real docs, hands off to a human gracefully when it can't help, and captures leads — without trapping anyone. This lesson is the difference between a deflection wall and real support.

The mental model

Helpful means three things: real knowledge, a graceful exit to a human, and a way to follow up.

A bot earns trust by knowing your actual documentation, by never trapping someone who needs a person, and by capturing the conversation so a human can follow up. Knowledge, handoff, and capture — that's the whole upgrade from FAQ toy to real support.

The reframeThe escape hatch is a feature, not a failure. A bot that can't reach a human when needed is a trap. Make the handoff easy and obvious — it's what makes people trust the bot enough to use it.

Step 01 · Helpful, or a trap?

Tap each bot behavior — does it genuinely help, or quietly trap the customer?

Step 02 · Set up handoff & capture

Two things make it real support. Switch each on:

Helpfulness:

0 / 3

That's real support: it knows your docs, escapes to a human cleanly, and captures the lead without holding anyone hostage.

Don't trap people to harvest emails. Forcing contact details before any help — or hiding the human option — tanks trust and gets your widget closed. Help first; capture second; always leave the exit visible.

Your challenge: make it genuinely helpful

Feed the bot your real help docs, not just a hand-typed FAQ.
Set up a graceful handoff that passes the conversation to a human.
Add optional lead capture that triggers after it helps.
Test a question it can't answer — confirm a clean handoff.

✓

Lesson complete

You can feed a bot real documentation, hand off to a human gracefully, and capture leads without trapping anyone.

Up next

Lesson 3 · Connected support — your tools & live data

Continue to Lesson 3 →

Open the Playground Back to Support Chatbot