Six production ML engines · single-tenant · your data stays yours

Turn your CRM into an
Enrollment Intelligence
platform in 60 days.

Six production ML models — lead scoring, counselor routing, retention risk, cancellation risk, enrollment forecasting, and SHAP explainability — running on your D365 / Salesforce / HubSpot data. Your counselors. Your programs. Your decisions.

How It Works
6
Live ML engines
1
Unified dashboard
< 60
Days to live deploy
24h
Demo turnaround
your-school-eip.scriptshub.net
COUNSELOR PERFORMANCE
Illustrative
LP
Luna Park
2,184 active leads
16.4%
conv. rate
AV
Atlas Voss
1,996 active leads
14.8%
conv. rate
NB
Nova Bennett
1,753 active leads
13.2%
conv. rate
KR
Kai Rivers
1,612 active leads
11.9%
conv. rate
Fictional preview · your dashboard will show your real team
What is ScriptsHub EIP?

Not a BI tool. Not a rebadged CRM. A real ML platform.

Most "enrollment analytics" dashboards are just pivot tables in disguise — they tell you what happened. ScriptsHub EIP is a six-engine machine learning platform that tells you what to do next, for which lead, by which counselor, today. Running in production at scale, with counselors acting on recommendations inside D365, Salesforce, and HubSpot.

The insight that makes ML worth it
A small fraction of your pool
drives most of your enrollments

In every enrollment dataset we've analyzed, a small subset of the inbound pool converts at rates many times higher than the rest. If your counselors are spending equal effort on every lead, they're burning most of their day on the wrong cohort. The platform's job is to rank them for you every morning — so the right 10–15% get called first, every time.

The IP — Six ML Models

Every model is yours. Every prediction is explainable.

Each model trains on your historical enrollment data — your counselor outcomes, your program mix, your channel performance. Models are retrained every 90 days and monitored for drift.

1

Lead Scoring

XGBoost · isotonic calibration · SHAP reason codes

Predicts which leads will enroll, so counselors call the right ones first. Every score ships with plain-English SHAP reason codes — counselors trust the model because they can see its reasoning.

Hot / Warm / Cold segmentation · retrained weekly
2

Lead Routing

Counselor-program affinity model

Matches each lead to the best-fit counselor — like assigning specialists. Learns continuously from outcomes to surface over-capacity and under-utilized staff.

Re-learns continuously from outcomes
3

Counselor Grading

Actual vs ML-expected performance

Measures every counselor against what the model predicted their leads would do — separating true performance from "got lucky with easy leads" or "got stuck with a cold cohort." Fair, defensible, quarterly reviews.

ML-adjusted · shown on every counselor profile
4

Enrollment Forecast

Weekly time-series · drift-monitored

Predicts inquiry volume and net new enrollments several weeks ahead, by program, with confidence intervals. Gives your CFO and VP Enrollment a real number to plan staffing and spend against.

Re-fit weekly · confidence intervals included
5

Cancellation Risk

Post-app, pre-term-start probability

Flags enrolled students most likely to cancel before term start — protects revenue already earned. Triggers targeted re-engagement workflows inside Pardot / HubSpot.

Integrates with your existing marketing automation
6

Student Retention

LMS + GPA + financial signals

Scores active students for dropout risk using Canvas / Blackboard engagement, GPA trend, credit-load change, and account balance. Flags the top risk cohort for advisor intervention before the drop happens.

Refreshed weekly · top-N cohort view for advisors
Closed-loop intelligence cycle
Score Leads → Route to Specialists → Grade Performance → Forecast Volume → Flag Cancellations → Protect Retention
Each engine feeds the next. One data layer, one login, one role-based permission model — across enrollment, retention, and marketing.
Three integrated modules

Enrollment. Retention. Marketing. One platform.

The six models power three modules that share one data layer, one login, and one role-based permission model. Buy what you need now, add the rest later.

Enrollment Intelligence

Enrollment Intelligence

Lead scoring + routing + hot-lead alerts + counselor performance + forecast. The daily cockpit for your enrollment team.

  • ✓ Daily hot leads email with routing
  • ✓ Per-counselor conversion + ML performance grades
  • ✓ Program-level enrollment forecast
  • ✓ Lead journey timeline with phone / text / email drill-down
Retention Intelligence

Retention Intelligence

Active-student dropout prediction with LMS, GPA, and financial signals. Advisors see who's at risk this week, not next quarter.

  • ✓ Top-10% risk cohort refreshed weekly
  • ✓ Risk factor attribution per student
  • ✓ Canvas / Blackboard engagement index
  • ✓ Leadership insights for VP Student Success
Marketing Intelligence

Marketing Intelligence

Per-channel and per-campaign lead quality + ROI. Finally answers the question "which Google Ads keywords actually enroll students?"

  • ✓ Lead quality score by source / medium / campaign
  • ✓ Campaign → enrollment → revenue ROI
  • ✓ AI campaign analysis (Claude / GPT-4o)
  • ✓ Channel spend optimization recommendations
Calculate your return

What's a 5-point retention lift worth to your institution?

Every retained student is tuition you don't have to re-recruit. Plug in your own numbers — the calculator updates live.

Your institution
+2 (conservative) +5 (moderate) +8 (aggressive)
Preset scenarios

Industry benchmark range (+2 to +8 points) sourced from EDUCAUSE 2024 analysis of predictive-analytics deployments in higher ed.

Additional annual tuition revenue
$3.0M
250 more students retained each year × $12,000 avg. tuition
4-year lifetime impact
$12.0M
Revenue recovered over the typical program cycle
New retention rate
70%
From 65% today

Want us to walk through these numbers live with your team? 30 minutes, no data handover.

Methodology: Students retained = enrollment × retention lift. Annual revenue = students retained × average annual tuition. Lifetime impact assumes a 4-year average program length. These numbers are directional — the real uplift depends on your student mix, program cost structure, and how many of the six engines you deploy. A full ROI projection with your actual cohort data is covered during the demo walkthrough.
How it works

60 days from handshake to live dashboard.

1

Data connection

Read-only credential into your D365 / Salesforce / HubSpot. We pull inquiries, activities, applications, outcomes. No writes to your CRM until you ask.

2

Model training

Six models fit on your historical data. Validation splits, calibration, feature-importance review with your team. Every model ships with its AUC / precision / recall receipt.

3

Your dashboard

Single-tenant install at yourschool-eip.scriptshub.net with your brand colors, SSO via Microsoft 365, your staff and roles imported from your CRM, and your program catalog.

4

Daily decisions

Hot leads email at 8 AM. Counselors act. Models retrain weekly. Drift monitored. Your VP Enrollment and CFO get one forecast number to steer by.

Why ScriptsHub

Three things that matter.

Single-tenant

Your own VM, your own database, your own scored data. No shared layer with other institutions. Ever.

Fully managed service

The six ML engines are ScriptsHub IP, delivered as a managed service on top of your data. No in-house ML team required — we own training, validation, deployment, and ongoing tuning.

Quarterly retrain cadence

Models re-fit every 90 days on fresh data as part of the retainer. Drift dashboards surface degrading models before they affect decisions.

Commercial model

Predictable cost that scales with your team.

No per-lead surcharges, no usage traps, no "AI credits." Three clean line items — everything else is included.

One-time

Implementation fee

Data integration (D365 / Salesforce / HubSpot / Canvas / Blackboard / Banner / Ellucian), initial model training on your historical data, single-tenant deployment, SSO wiring, counselor onboarding.

60-day timeline · fixed-price · paid at milestone
MOST OF THE SPEND
Per counselor / month

Seat licensing

One license per active counselor, advisor, or admissions-team user. Covers access to all six engines, the full dashboard, hot-leads email alerts, and AI recommendations. Scales up or down with headcount — pay only for the seats you actually use.

Volume tiers available at 25+, 50+, 100+ seats
Monthly retainer

Platform & ML maintenance

Quarterly model retraining on fresh data, drift monitoring, feature engineering updates, platform patches, security updates, uptime SLA, and a named engineer for escalations. The work that keeps the models sharp and the dashboard alive.

Scales with data volume and model count

Actual pricing depends on your institution size, CRM complexity, and how many of the six engines you turn on. One reply and we'll put a quote in front of you within 48 hours.

Start with a 30-minute walkthrough.

No prep, no data, no commitment. We'll screen-share the platform live, show you all six engines, and answer your questions. If it's a fit, the POC on your own data is the natural next step.