Profile settings

The Settings tab of a scoring profile holds everything except the criteria themselves. Changes here are saved and published together with the criteria.

Name and description

  • Profile name — shown in the profile list and in the source settings.
  • What a good lead looks like — a few sentences about your business and your best customers: what you sell, to whom, where, what a great job looks like. The AI reads it with every message; it helps it understand your enquiries and write useful summaries. For example: "Catering company in Melbourne. Best jobs are weddings and corporate events of 100+ guests within 50 km of Collingwood, booked months ahead."

The general description of the business (shared by all profiles) is in Workspace settings.

How criteria combine

How the answers to the criteria become the score from 0 to 100:

OptionHow the score is computed
Points: each answer adds points (recommended)Each answer of a criterion has its own points; the score is the sum of the chosen answers, capped at 100. See How the score is calculated.
Grades: the score is their averageEach answer is a grade from 1 to the best grade (usually 5). The score is the average grade of the criteria that could be answered, as a share of the best grade: grades 5, 3 and 4 → average 4 of 5 → 80. Criteria that couldn't be answered are left out of the average.
Points per criterion: the score is their share of the maximumEach criterion has a fixed number of points; the score is the points earned as a share of all positive points in the profile.

With Points, the numbers you see are the numbers you get, and you can make one criterion matter more than another just by giving it more points. Switching an existing profile to Points gives every answer points from its grade (best grade = 20, the next 16, and so on) — adjust them afterwards and check with the Test panel.

Best grade appears only for the grades option: the top of the grade scale, 5 by default.

Tiers

Tiers turn the score into a label your team can act on — for example:

TierFrom score
critical60
high40
normal20
low0

A lead gets the first tier, from the top, whose minimum it reaches: a score of 45 is high. List tiers from the highest to the lowest, and let the last one start at 0. You can have two to six tiers and name them as you like (hot / warm / cold works too).

Some tiers are set by Leadveyor itself: disqualified (a disqualify criterion fired) and unscored (no criterion could be answered). An exceptional criterion puts the lead into the top tier, or the tier it names.

How to choose the minimums: test a few real messages you would call great, average and poor, and set the boundaries between their scores. With Points, scores depend on how much each message tells — a short form with just a name and a date scores lower than a detailed email.

Fields to extract

Contact details (name, email, phone, company), a summary of the request, the service and the address are always extracted. Add the fields that matter for your business; they are shown on every lead, sent in webhooks and can be used in rules as fields.<key>.

Each field has:

  • Key — its identifier: lowercase letters, digits and _, starting with a letter, e.g. guest_count.
  • Type — what kind of value it is:
TypeForExample
stringa short text"Buffet"
texta longer textdietary requirements
numbera count or quantity120; "100–150" keeps both ends
moneyan amount with currency15 000 AUD; without a currency, the workspace currency is used
datea day"next Saturday" → the date; "in spring" → a range of days
datetimea day and time2027-05-15 18:00
booleanyes / no"needs staff": yes
enumone of a listwedding / corporate / birthday
multi_enumseveral of a listcanapés, mains, dessert
email, phone, urlcontact details beyond the main onesa second email
  • Description — what the field means, in plain words. The AI reads it: "Number of guests expected", "Date of the event".
  • Values — for enum and multi_enum: the allowed values, separated by commas.

Dates always keep the customer's own wording too ("the Saturday after Easter"). Up to 30 fields per profile.

Dimensions

The groups used to break a lead's score down — for example Who, What, When, Where, Why. Each lead then shows, next to its score, how each group did: "What 80 · When 0 · Where 60" means the request and the location are good but the date is missing.

Here you only set the list of groups: add, rename or remove them. Which criterion belongs to which group is chosen in the criterion itself, in its Dimension field. Removing a group also removes it from the criteria that used it. Dimensions never change the score; leave the list empty if you don't need the breakdown.

Junk filter

  • Filter junk — when on (recommended), ads, spam, newsletters, automatic replies and other mail that isn't a customer enquiry are set aside and not scored. You still see them in the Inbox and can send a message back for scoring with Not junk — process again.
  • Extra rules — one per line, in plain words, for junk specific to your business: "Messages from suppliers about deliveries are not leads", "Job applications are not leads".
  • Always junk (domains) — email domains whose messages are always junk, one per line: cheap-seo.example.
  • Never junk (domains) — domains whose messages are never treated as junk, for example a lead platform that forwards enquiries: bookings.example.com.

Was this page helpful?