Ask a question that changes the next step
Start from a decision your team already makes. Do you serve the visitor location? Does the request fit a service you offer? Is a consultation required before a quote? A useful qualification question helps resolve that decision without turning chat into a long questionnaire.
Ask one question at a time and explain why a detail is helpful. Budget and timing may matter for a custom project, but they may be unnecessary for a simple policy question. Let the visitor request a person when the standard flow does not describe their situation.
Understand how the score is calculated
The application uses configured weights for available lead fields and signals. It sums the relevant weights, caps the score, and assigns a priority category. This is a rules-based prioritisation aid in the current product, not a machine-learning probability that a lead will close.
Different businesses need different weights. A clinic appointment request, an agency project brief, and an ecommerce support question are not comparable by one universal score. Review the fields your business uses and check whether the resulting categories reflect your own follow-up priorities.
Separate a missing field from a negative answer
A visitor may decline to share a budget or may not know a timeline. That absence is not the same as an unsuitable request. Read the conversation before rejecting a lead solely because some fields are blank. Your team may need to ask a more relevant question directly.
Likewise, a pricing question is not necessarily evidence of strong buying intent. Some visitors are gathering information or comparing options. Use detected signals alongside the request and your service requirements rather than converting every signal into a sales claim.
Suggested qualification questions
- What are you trying to achieve with this service?
- Which service or product are you asking about?
- Is there a date or timing requirement the team should know?
- Which area should the team check for service availability?
- How would you prefer the team to contact you, if you want follow-up?
Review the weights with real team outcomes
During a trial, have staff review a small group of enquiries and note why each needs attention. Compare those judgements with the configured score. If a high score often reflects irrelevant information, adjust the questions and weights rather than promoting the score as an objective measure.
Do not add sensitive questions simply to improve a priority category. Keep qualification focused on business needs and use the relevant privacy policy. A person should review requests involving medical, legal, financial, or other consequential decisions.
Connect qualification with action
Once your team understands the requirement, decide whether to answer, arrange a discussion, or explain that the service is not suitable. A qualification process is unfinished if the lead remains in a list without a responsible person and a next action.
Lead records, tags, tasks, and configured automation can support follow-up. Check the workflow in your own workspace and plan limits before relying on it. Native external CRM connectors are not confirmed; signed webhooks still need a developer-managed receiving system.
Test a high-score enquiry that is still unsuitable
A visitor can provide every field and still ask for a service your business does not offer. Build a test like that with synthetic details and check whether staff can recognise the mismatch from the conversation. A high score should direct attention, not override your service policy.
Then test a short, suitable request with fewer fields. Review whether your team would want to answer it despite a lower category. This comparison helps reveal weights that reward completeness instead of relevance. Adjust the process thoughtfully rather than setting every weight high to make the dashboard look positive.
If your team uses budget or timing, explain their purpose and allow a person to clarify an uncertain answer. Avoid rejecting a request simply because the visitor phrases it differently from your examples. Review changes to questions and scoring together, then test how they affect the next action rather than only the displayed score.
A controlled trial to adapt
Example to adapt: a visitor asks for a service outside your published area but provides a detailed brief and contact details. The record may contain strong scoring signals. Staff should still explain the service-area limit before requesting more information. Compare it with a brief enquiry from someone in the supported area. Record which one needs action and why, then review whether your questions and weights help the team reach that judgement. These are synthetic tests, not customer evidence or an instruction to reject every enquiry outside a fixed category.