drio

Launch Your First App

Configure a sales app that answers a buyer, qualifies them, and books the call.

This is the fastest path to a sales app that converts. You will build one coherent conversation that walks the full spine — answer, qualify, book — instead of a pile of unrelated features. It is written for the team that owns the AI channel, not for engineers building from scratch.

Before You Start

Come in with these four things:

  • one kind of buyer and the decision you want them to make, such as booking a demo or starting a trial
  • one source of truth their questions are answered from, such as your product content, pricing, help center, or a connected API
  • three to five real buying questions they ask in ChatGPT or Claude today
  • a place to book the meeting, such as a scheduling link or your CRM

What You Will Build

By the end of this guide you will have one app that:

  • answers your buyer's top questions from your own content
  • qualifies them with the right follow-up questions or live data
  • handles the obvious objection with proof
  • captures the contact and books the meeting
  • runs live in the AI client your buyers actually use

Build It Step By Step

We will use a running example: a buyer asking ChatGPT "what's the best [your category] for a fast-growing team?" Your app's job is to win that answer and turn it into a booked call.

Frame the app around one buyer and one decision

Start with a single buyer and the one decision you want them to make. Good framings:

  • "Help an evaluating buyer pick the right plan and book a demo"
  • "Turn a pricing question into a qualified trial"
  • "Answer a migration question and book an implementation call"

Give the app a name your team will immediately understand. If the app tries to serve every buyer at once, narrow it.

Configure tools that answer and qualify

A tool is one thing the assistant can do well. A converting app needs two kinds:

  • Answer tools that respond from your approved content, so the buyer gets your facts, not a guess.
  • Qualify tools that pull live data or ask the right questions, so you learn who the buyer is and what they need.

For the example, that might be an answer tool grounded in your product and pricing content, plus a qualify tool that asks team size and use case. Keep each tool narrow. If a tool feels like "do everything," split it.

See Plan assistant actions.

Ground every answer in a reliable source

Connect each tool to the cleanest source you have: your content, product data, or a stable API. Reliability matters more than coverage. A narrow app that is always right beats a broad one that guesses.

This is what makes the app trustworthy enough to recommend a buyer toward a decision.

Choose widgets that move the buyer toward booking

Match the response format to the buyer's next step:

  • comparisons and proof cards when they are weighing options
  • tables when they are checking fit or pricing
  • a form when it is time to capture contact details and book

Every response should make the next step toward the meeting obvious. If a response ends the conversation instead of advancing it, rework it.

See Choose the right response format.

Brand, test, and connect

Apply your logo, colors, and copy so the app feels like you inside the chat. Then test with the real questions your buyers ask, not idealized demo prompts. Walk the whole spine yourself: ask, get an answer, get qualified, hit an objection, and book.

When it holds up, copy the MCP endpoint and connect it in the AI client your buyers use.

See Match your brand and Connect to AI clients.

Success Checklist

  • the app answers the top three buying questions from your own content
  • it asks at least one qualifying question before pushing to book
  • it meets the most common objection with proof, not deflection
  • the booking step is reachable the moment the buyer is ready
  • the live endpoint works in your target AI client

When To Bring In A Technical Teammate

You can configure most of this yourself. Bring in a technical reviewer when:

  • the app needs per-user access to systems like Salesforce, HubSpot, or internal tools to qualify or book
  • the upstream API needs non-trivial auth or request shaping
  • the team wants direct management API automation
  • security, logging, or deployment questions need sign-off

Next Steps