Getting in
A lead score nobody can interrogate is a number nobody uses.
Reps do not reject scoring because they dislike prioritisation. They reject it because the first time it is obviously wrong about a lead they know, there is nothing to point at — and after that the column is ignored.
So the reasoning is stored next to the number, in words.
Scoring
Six factors, and the sentence that produced each one.
Where they came from, how senior they are, how big the company is, whether the industry is on your list, how recently anybody spoke to them, and whether you have done business with them before.
Each factor's contribution is written down when the score is calculated, with a human-readable note — spoke twelve days ago, industry not on your target list, never contacted. The screen renders it as a table.
- Where they came fromweight 20Inbound demo request
- Seniorityweight 15Director of Operations
- Company sizeweight 15201–500 people
- Industry fitweight 10Logistics — on your target list
- How recently contactedweight 20Spoke 12 days ago
- Past businessweight 20Two closed deals with a sister company
- Weights are yours; the factors are not —
You set how much each factor counts, and can switch one off entirely without losing its configuration. You cannot invent a seventh — and the settings screen explains why rather than leaving a dead end, which is a small kindness most products skip.
- A factor set to zero weight is off, not broken —
And every other factor's share adjusts, so switching one off does not silently drag every score down.
- Recalculating everything is one action —
And it only rewrites the leads whose score or reasoning actually changed, so the audit does not fill with no-op updates.
Routing
First match wins, and a typo cannot route your whole inbound stream to one person.
Rules in priority order over six fields — where the lead came from, three campaign parameters, the country, the industry and the company size. Two outcomes: a named person, or round-robin over a pool.
The most consequential line in the whole feature is the default for an unrecognised condition. It makes the rule un-matchable, not universal. An administrator's typo turns into "this rule never fires" rather than "every lead goes to Bob" — which is the same one-line decision, made the other way, in most products.
- A blank field never satisfies a rule that tests it —
Absence is not a wildcard. Which is why the web form only passes the three fields it actually collects.
- Round-robin survives two submissions at the same instant —
The position advances and commits before the assignment returns, so two simultaneous form posts cannot both take the first person in the pool.
- An explicit choice always wins —
Routing only runs when the creator did not name an owner. A rep who assigns a lead deliberately is never overridden.
- A pool entry that is no longer valid is skipped, not fatal —
At the moment of assignment rather than at save, so a departed colleague in a pool does not break inbound capture.
›Can a rule test a range, or combine conditions with or?
No. Conditions are equality only, combined with and. No ranges, no negation, no list membership. A rule for "deals over fifty thousand in Germany or Austria" is two rules, and the value comparison is not available at all.
›Can a rule assign to a team rather than a person?
No. The outcome is a named person or a round-robin pool of people.
›Does routing know about capacity or working hours?
No. Round-robin is strictly in turn — it does not skip somebody on holiday or already carrying forty leads.
›Can I edit a rule's conditions after saving it?
A rule can be switched off and deleted. Changing what it tests is currently delete-and-recreate, which is worth knowing if you expect to tune rules frequently.
›Does territory feed into routing?
No. Territories classify accounts and leads; they are not an input to either routing engine. See the territories page.