A detailed enquiry can contain useful buying signals that never reach the CRM: the requested service, a delivery constraint, or an unanswered question about scope. Norml builds AI lead qualification around your written fit criteria. The workflow reads an eligible enquiry, checks the approved CRM context, and prepares a qualification record with evidence for each criterion. Your team can see why a lead was flagged and correct the assessment before it affects follow-up.
The scope covers interpreting an enquiry and preparing a reviewable CRM update. Your team defines what counts as a suitable opportunity.
We turn your qualification rules into named criteria, such as service fit, location or required capability. Each assessment carries the relevant enquiry text or approved CRM fact. A missing answer stays unknown; an enthusiastic tone does not become evidence of budget or buying authority.
The workflow looks up the contact using agreed identifiers and checks whether the enquiry has already been processed. We map approved assessment fields and preserve human-owned lifecycle stages, deal stages and notes. A repeated enquiry from the same person remains a separate event where your process requires it.
Ambiguous requests, conflicting facts and failed CRM updates go to an owner. The model cannot invent new qualification rules or follow instructions embedded in an enquiry. We keep a processing record so an interrupted run can be reconciled without silently overwriting a salesperson’s correction.
Supply examples of suitable, unsuitable and unclear enquiries, with the reasons your team would record. Include which criteria are mandatory and which merely inform a conversation.
We need the enquiry source, record identifiers, HubSpot property definitions and permitted access. Decide which existing context may be sent to the model and which data must remain outside the workflow.
Choose who reviews ambiguous assessments, approves CRM changes and maintains the criteria. Define whether a completed review creates a follow-up task, updates fields or only records a recommendation.
We test representative enquiries against decisions made by your sales team. Disagreements reveal unclear criteria, missing inputs or cases that should remain manual.
We build intake, contact matching, assessment storage and the agreed HubSpot update. Failure handling covers missing contacts, duplicate events, denied access and an update interrupted after submission.
Your team checks evidence, field protection and uncertain cases. We document how to change a criterion, pause processing and review failed runs without submitting duplicate updates.