Ask most teams how they qualify leads from a form, and the honest answer is: they don't, not really. The form collects answers, everything lands in a CRM or a spreadsheet, and a person — usually the least available person on the team — manually decides who's worth a callback. That's not a qualification system. That's a queue with extra steps.
A real qualification system scores the lead the moment they submit, before a human ever sees it. That requires two things most form builders don't actually have: a way to compute a score from multiple answers, and a way to route based on that computed score rather than any single raw answer.
The scoring model: points, not guesses
The simplest version that actually works is a weighted points system. Every answer that correlates with lead quality gets a point value; the form adds them up as the respondent answers; the total determines what happens next. No AI, no black box — just arithmetic anyone on the team can audit and adjust.
| Question | Answer | Points |
|---|---|---|
| Company size | 50+ employees | +15 |
| Company size | 10–49 employees | +8 |
| Company size | Under 10 employees | +0 |
| Budget | Over $10,000 | +20 |
| Budget | $2,000–$10,000 | +10 |
| Timeline | Ready to start now | +15 |
| Timeline | Just researching | −10 |
| Decision authority | Final decision-maker | +10 |
Notice "just researching" is a negative weight, not zero. Zero says "this answer doesn't tell us anything." Negative says "this answer actively signals low intent right now" — and it should pull a borderline lead below the threshold rather than leaving it neutral. Most scoring models people build by hand forget this, and end up over-qualifying window-shoppers.
Turning points into a computed variable
Each row in that table is a point value tied to one answer. The score itself is a computed variable — a value the form calculates by adding together every point value that applies, updated live as the respondent answers. It isn't a question anyone answers directly; it's the sum the form derives from what they did answer.
This matters because branching logic can then use the computed score exactly like it would use a real question — "if computed_score is greater than or equal to 45, do X" is a normal condition, just built on a number the form itself produced.
From score to action: the routing tiers
A score by itself doesn't do anything until it's tied to an outcome. Three tiers is usually enough to start — more than that and the distinctions stop being meaningful.
- Hot (45+ points): Route directly to a calendar booking link on the outcome screen, and fire a webhook to notify sales immediately. No queue — this lead gets a same-day response.
- Warm (20–44 points): Route to a standard follow-up sequence. Show an outcome screen that sets expectations ("we'll be in touch within 2 business days") rather than an immediate booking link, which would overpromise on response time.
- Cold (under 20 points): No sales handoff at all. Route to a resource — a guide, a case study, a newsletter signup — and let them self-educate until they're ready to re-engage.
The outcome screen respondents actually see should be genuinely different per tier, not just an internal routing tag. A cold lead who gets shown "Thanks, a rep will call you shortly" and then never gets called is a worse experience than one who's told upfront they're getting a resource guide instead.
Getting the score to your CRM, not just your form
The scoring logic only matters if the score actually reaches wherever sales works. Every submission — along with the computed score and whichever tier it landed in — should push out via webhook the moment the form is submitted, not on a schedule. If the receiving CRM or Slack channel doesn't come back with a 200 immediately (a brief outage, a deploy in progress), the delivery needs to retry automatically rather than silently drop a hot lead because the receiving end blinked for ten seconds.
The mistake to avoid: scoring what's easy to ask, not what predicts outcome
The most common failure isn't a missing point value — it's assigning points to questions that are easy to ask but don't actually correlate with who becomes a customer. Company size is a common offender: it feels like a solid proxy for budget, but plenty of small teams have real budget and plenty of large ones are just kicking tires.
Before finalizing point values, pull your last 20–30 actual closed deals and check which answers those leads gave. If "company size" doesn't actually differentiate your closed-won list from your closed-lost list, don't weight it — weight whatever field does. The scoring model should be built from your own conversion history, not copied from a generic template.
A scoring model with the wrong weights isn't neutral — it actively misroutes leads with more confidence than having no scoring at all, because everyone trusts the number.
Building this in MarketKloud Forms
Computed variables, priority-ordered branching on the resulting score, and per-tier outcome screens are all native — no external scoring service, no Zapier chain adding latency between submission and routing. Webhooks fire the moment the form is submitted, with automatic retries on failed deliveries, so a hot lead's score reaches your CRM in real time, not on the next scheduled sync.