Skip to content
Aplora
EdTech
Aplora Sales

We validated the system inside our own operations first

Lead scoring in the intake funnel: advisors stopped calling everyone in order

An online IT school in Poland. Applications arrived from webinars, quizzes, site forms and ads, landed in the CRM as a single stream and were worked in the order they came.

Published: 2026-08-24 · updated: 2026-08-24

Hands on a laptop keyboard, a list of CRM records on the screen
Illustrative photo of the working context, not a screenshot of the client's system

In short

  • Problem: applications were worked in arrival order with no priority, and some purchase-ready leads went cold in the queue.
  • Solution: scoring by lead actions, a cold / warm / hot status, a priority and a task in the CRM, a script and follow-up matched to the segment.
  • Result: the funnel became manageable; this part has no separately measured effect — it is accounted for in the intake campaign's overall result.
  • Systems: CRM, webinar platform, quizzes, e-mail campaigns, the scoring rules table.

Context

  • An online IT school selling programmes B2C through a consultation call.
  • Team: admissions advisors plus a head of sales.
  • Process volume: an uneven application flow with peaks at the start of each intake.
  • Systems: the CRM as the source of truth, a webinar platform, quizzes, e-mail campaigns, and a table holding the scoring rules with a version log.
  • The constraint became visible as volume grew: the conversion spread between advisors came less from skill than from which leads each of them got.

Baseline

Baseline metrics with their sources. Without them, any later result has nothing to be compared against.
  • Before the work started we fixed: the share of leads worked in priority order, and the time from application to first contact for leads showing readiness.
  • Data source: CRM exports covering a comparable intake period.
  • The values are not published: this part has no separate effect — the same conversion gain is already attributed to conversation review, and claiming it twice would count one result as two.

Diagnosis

Which hypotheses were considered, why this one was chosen, what was assumed, and the condition under which we would have stopped.
  • Three hypotheses were on the table: not enough applications, not enough advisors, or applications distributed wrongly.
  • The third was chosen: volume was sufficient, and reviewing history showed that leads with several consecutive actions — signup, attendance, a quiz answer — converted markedly more often, and those were exactly the ones waiting longest for a call.
  • The assumption: a lead's action signals intent rather than chance. It was tested on historical data by running the rules backwards over closed deals.
  • Stop criterion: if scoring cannot separate closed deals from rejections on historical data, the rules count as unfit and the work stops.

What we implemented

  • Data sources: webinar platform events, quiz answers, opens and clicks in campaigns, pricing page visits, lead records in the CRM.
  • Business rules: the weight of each event and the status thresholds live in a table the head of sales can open. No value is hard-coded.
  • Version log: every weight change carries a “why” comment — otherwise nobody remembers a quarter later where a threshold came from.
  • Integrations: CRM, webinar platform, quizzes, e-mail campaigns.
  • Human checkpoints: the advisor sees which events produced the status and can raise the priority by hand. The decision to call stays with a person.
  • Monitoring: the share of leads in each status and the conversion by status are tracked separately — a gap between them means the thresholds need recalculating.

How the process changed

Before

5 steps
  1. The application lands in the CRM
  2. The advisor takes the next record in order
  3. A call with no idea what preceded the application
  4. A free-form note
  5. Follow-up from the advisor's memory

After

8 steps
  1. The application lands in the CRM
  2. Signals collected: webinar, quiz, e-mails, visits
  3. Scoring against the rules table
  4. Status: cold / warm / hot
  5. Priority and task in the CRM
  6. A call using the segment's script
  7. Segment-matched follow-up
  8. A deal, or back into nurturing

What was stuck

Hot and cold leads sat mixed together. An advisor spent the day on people with neither a reason nor an intent, while leads showing clear readiness went cold in the queue.

  1. 1The application lands in the CRM
  2. 2Signals collected: webinar, quiz, e-mails, visits
  3. 3Scoring against the rules table
  4. 4Status: cold / warm / hot
  5. 5Priority and task in the CRM
  6. 6A call using the segment's script
  7. 7Segment-matched follow-up
  8. 8A deal, or back into nurturing
  9. Measured result

Results

Why there are no numbers here

This part has no separately measured effect. The same conversion gain is already attributed to conversation review in the neighbouring case, and claiming it here again would count one result as two.

  • Share of leads worked in priority order
  • Time from application to first contact on a hot lead

Economic impact

  • The effect is a reallocation of advisor time already paid for, not a reduction of it: the same hours, but spent on leads with intent.
  • There is no separate money figure: scoring's contribution cannot be separated from conversation review — both went into the same loop.
  • Cost of ownership: the rules need regular recalibration — the channel mix shifts, and the weights do not follow on their own.

Adoption

  • Advisors accepted scoring because they can see what it is made of: the status expands into the list of events rather than arriving as a verdict.
  • Priority became the standard way of working rather than a suggestion.
  • The solution owner is the head of sales, who also owns the rules table and signs off changes.
«We used to call everyone in order and burn time on cold leads. Now it is visible who is genuinely close to buying, and the team goes there first.»

Client under NDA — Head of sales, LearnIT

What's next

  • Scaling: the same rules for the second intake and for the upskilling programmes.
  • Next initiative: reactivating leads lost in earlier intakes because of overload.
  • What we decided against: auto-closing cold leads. The threshold is fallible, and a closed record does not reopen.

Have a similar workflow? Let's check whether the hypothesis transfers

Another company's result is not a promise. It does show where to look.

All cases