Skip to content
SentientOne

Use case

Lead qualification

Add AI to your sales pipeline: every enquiry from your website form read the moment it arrives, scored against your criteria, and routed to sales with the details filled in.

Start freeSee the setup

The problem

What we're solving

Who has it

Sales teams that get enquiries through a website form and answer them in order of arrival.

What it costs today

Good leads wait behind the rest until someone reads them, and every enquiry gets retyped into the CRM or a chat message.

What solved looks like

Each enquiry is read immediately, its company size, need and timeline pulled out, scored hot, warm or cold, and the hot ones posted to sales with a reply already drafted.

  • Hot enquiries in your sales channel seconds after they arrive.
  • Consistent fields on every lead, ready for your CRM.
  • A drafted first reply for the leads that matter most.

How it runs

Lead qualification step by step

Each stage is a real step in SentientOne. Timings are medians from our latest benchmark; your model and network change them.

Lead qualification flow, a published workflow. Your website form posts each enquiry to the workflow's webhook URL. 1: the Start block checks the fields. 2: Extract pulls out company, team size, need and timeline. 3: Classify scores the lead hot, warm or cold against your description. 4: a Condition checks for hot. 5: an Agent block drafts a first reply. 6: a Slack message posts the lead and the draft to your sales channel; warm and cold leads go to an HTTP request into your CRM.

Set it up

The walkthrough, in 7 steps

  1. 1

    Add your LLM key

    Agents run on your own model provider key; there's no shared one, so nothing answers until you add it. Create a key with OpenAI, Anthropic, Google Gemini or Groq, then open Configure → LLM keys → Add key. It's encrypted at rest with AES-256-GCM, used only to call your provider, and never shown again in full. If a reply fails with INTERNAL_ERROR straight away, the agent has no key; with LLM_AUTH_ERROR, the provider rejected it.

    LLM keys
    Provider
    Anthropic
    Key
    sk-ant-api03-•••• (shown masked after saving)
    Help: Add your LLM key
  2. 2

    Create the drafting agent

    Agents → New agent → Start from: Custom. Its prompt describes your product, your ideal customer and how sales replies should sound.

    System prompt
    You draft first replies to sales enquiries for Acme Software.Two or three sentences, warm and specific to what they asked.Offer a short call. Never quote prices or promise features.
    Help: Create an agent
  3. 3

    Build the workflow

    Workflows → New workflow. Start fields: name, email, company, message. Extract team_size (number), need (text) and timeline (text). Classify hot, warm, cold with a sentence each on what they mean for you. A Condition on hot; true goes to Agent then Slack message, false to an HTTP request to your CRM.

    Workflow designer
    Start
    name, email, company, message
    Extract
    team_size, need, timeline
    Classify
    hot · warm · cold
    Condition
    label equals hot
    Agent → Slack message
    true
    HTTP request (CRM)
    false
    Help: Workflow blocks
  4. 4

    Write the Slack message

    Fill the message from earlier steps with template values so sales sees everything at a glance.

    Slack message block
    Hot lead: {{input.company}} ({{steps.extract_1.output.team_size}} people)Need: {{steps.extract_1.output.need}} · Timeline: {{steps.extract_1.output.timeline}}Contact: {{input.name}} <{{input.email}}>Draft reply: {{steps.agent_1.output}}
    Help: Templates and variables
  5. 5

    Test, then publish

    Test run with real past enquiries and check the scores. Adjust the Classify descriptions until they match your judgement, then Publish.

    Help: Build a workflow
  6. 6

    Connect your form

    In the Start block's Webhook trigger, Create token and copy the Incoming webhook URL. Have your form (or form tool) POST each submission there with the X-Workflow-Token header.

    Help: Run a workflow
  7. 7

    Watch runs and credits

    Check the runs list for failed steps (for example a CRM that rejected a request) and turn on auto top-up under Billing.

    Help: Auto top-up

Help

Need a hand? We'll set it up with you.

Tell us about your setup and we'll help you build lead qualification on SentientOne: the agent, the knowledge, the tools and the deployment.

Book a setup callStart free

Or email support@sentientone.ai. 200 free credits when you sign up, no card needed.

Your first week

  1. Day 1Write down what hot, warm and cold mean for you, then build the workflow.
  2. Days 2–3Test run it with 30 past enquiries and compare the scores with your sales team's.
  3. Day 4Publish, post to a test Slack channel and connect a test copy of your form.
  4. Day 5+Switch the live form over; review the drafts and misses with sales each week.

More use cases