Build a team of AI agents that run your business and get paid, no code.Start building

AI operations · Role guide

The AI lead qualification agent: every inquiry read, explained and routed in minutes.

Leads go cold in inboxes. The lead qualification agent reads each inquiry as it arrives, applies the fit criteria your sales team already uses, writes down why it scored the lead the way it did, and routes it to the person who should call back. Missing information becomes a follow-up question, never an invented score.

A laptop open on a desk showing a sales and traffic analytics dashboard
The KPIMedian time from inquiry to routed owner
Photo via Unsplash

What is an AI lead qualification agent?

An AI lead qualification agent reads inbound inquiries from forms, email and chat, applies a written set of fit criteria, records the reason for each result, and routes the lead to the right owner with a suggested next step. Suede AI builds and runs it; your team works the qualified queue.

Reads
Forms, inbound email, chat
Produces
Scored leads with stated reasons
Hands back
Ambiguous fits, large deals
KPI
Median time from inquiry to routed owner

What the agent does

The lead qualification agent's working day

  1. Catch every inquiry

    Watches the web forms, the shared sales inbox and chat transcripts, and logs each new inquiry with its source and arrival time.

  2. Enrich from what you already have

    Checks the CRM for an existing account, open deals or past conversations so a returning customer is never treated as a cold lead.

  3. Apply the fit criteria

    Scores service fit, scope, timing and decision role against your written rules, and quotes the inquiry text behind each answer.

  4. Route with a next step

    Assigns the lead to the owner your routing rules name, books or suggests the next action, and confirms the owner received it.

  5. Ask for what is missing

    When a deciding fact is absent, drafts a short follow-up question for approval instead of guessing.

The guide

How to set up AI lead qualification, step by step

Qualification is only as good as the rulebook behind it. Most of the work happens before the agent runs, with your sales owner, and it pairs naturally with AI integration when the CRM needs connecting first.

  1. Step 1

    Write the fit criteria with the sales owner

    Suede reviews the last fifty inquiries with the person who owns sales: which became customers, which wasted a call, which were close. The answers become a short written rulebook.

    • Four to eight criteria, each with a defined meaning
    • A rule for missing information on every criterion
    • The difference between a clear mismatch and an incomplete inquiry
  2. Step 2

    Map the routing

    Who gets which lead, by service, region, size or language, and who covers when that person is out. Every path ends at a named person or queue.

    • Owner per route plus a backup
    • A review queue for ambiguous fits
    • Business hours and after-hours handling
  3. Step 3

    Connect intake and the CRM

    The agent reads from your forms and inbox and writes to your CRM. Suede scopes the access to exactly the objects and fields the rulebook needs.

    • Form and inbox sources listed
    • CRM fields for score, reason and owner
    • Consent captured for any text or email follow-up
  4. Step 4

    Back-test on last month's leads

    The agent qualifies last month's inquiries and the sales owner compares its results with what actually happened. Rules that misfire get rewritten before live leads flow.

    • Side-by-side results for every back-tested lead
    • Sales owner sign-off on the rulebook
    • Baseline for the time-to-owner KPI
  5. Step 5

    Go live and watch the queue

    Live in Agent Studio, every lead shows the criteria applied and the route chosen. Suede tunes the rules weekly from the review queue and the reps' feedback.

    • Reps can see the reason on every lead
    • Weekly review of the ambiguous queue
    • Monthly rulebook update with the sales owner
A sales team arranging sticky notes on a glass wall while planning
The fit criteria start as the team's own judgement, written down. Photo via Unsplash

Guardrails

Written down before the agent runs a single task

Every role ships with the same four lists, agreed with the person who owns the outcome today. They live in the flow itself in Agent Studio, so the rules and the runs sit side by side.

May

  • Read inquiries from the agreed sources
  • Write score, reason and owner to the CRM
  • Notify the assigned owner

Never

  • Discount, quote or promise delivery
  • Message a lead without approved wording and consent
  • Delete or merge CRM records

Hands back to a person

  • Ambiguous fits to the review queue
  • Deals above the agreed size
  • Existing customers with an open issue

Stops on

  • The routed owner is unavailable with no backup
  • The CRM write fails
  • An inquiry that looks like spam or abuse

The one number

Median time from inquiry to routed owner

Definition
For every qualified inquiry in a week, the time between arrival and the assigned owner receiving it, reported as the median.
How it is measured
Arrival time from the form or inbox, owner receipt from the CRM assignment, read from the Agent Studio run history.
Read alongside
  • Share of leads routed with a stated reason
  • Leads sent to the review queue
  • Reps' agreement with the score on spot checks

In practice

What a typical cycle looks like

  1. Monday

    The weekend's inquiries are already scored and routed when the team logs in.

  2. Tuesday

    Reps work the queue with the reason for each score on the record.

  3. Wednesday

    The sales owner clears the ambiguous queue and notes any rule that misfired.

  4. Thursday

    Suede updates the rulebook in Agent Studio from those notes.

  5. Friday

    The weekly KPI note shows median time to owner and the review-queue count.

Systems it connects to

Built on the tools you already run

The agent reads and writes through each tool's own API, with access scoped to what the guardrails allow. Each link opens the developer documentation Suede builds against.

Guides and links

Further reading for this role

Jason Colapietro, founder and CEO of Suede AI, smiling in sunglasses and a white tee

Who runs it

An operator's method, built by an engineer.

Before Suede, Jason Colapietro built, ran and sold call centers, where call review, lead handling, script discipline and the reporting rhythm were run by hand. The agents encode that method. He has 54 pull requests merged across 46 open-source repositories and scopes every engagement himself.

Read the founder record

Questions buyers ask

Questions about the AI lead qualification agent

How does AI lead scoring differ from our CRM's lead score?

A CRM score usually adds points for page views and form fields. The qualification agent reads what the person actually wrote, applies your written fit rules, and records the reason in plain language, so a rep can see why the lead was routed to them.

What happens when an inquiry is missing key information?

The agent marks the missing fact and drafts a short follow-up question for approval. It never fills the gap with a guess, and the lead stays visible in the queue while the answer is pending.

Can it book meetings automatically?

Yes, when you approve it. Qualified leads that meet the booking rule can receive a scheduling link with approved wording; everything else is routed to a person first.

Which CRMs does it work with?

HubSpot, Salesforce and Pipedrive are the common ones, and any CRM with an API can be connected. Suede scopes access to the objects and fields the rulebook needs.

How long does setup take?

The rulebook and routing map take the most time, because they come from your sales owner's judgement. Suede quotes the timeline in the written scope after the first call.

Who owns the qualification rules?

Your sales owner. Suede writes and tunes them, and every change is signed off before it reaches live leads.

The other roles

Four more lanes from the AI operations map

Each role follows the same method: map the lane, write the guardrails, agree one KPI, then build. See how they fit together on the AI operations page.

Next step

Tell Jason where this work piles up.

Thirty minutes is enough to map the lane and agree the number it should move. You leave with a written scope, and every run is open to watch in Suede AI Agent Studio.

Slots run Monday to Friday until 6 p.m. Eastern.