Predictika · Higher-education ERP subscription orders

An AI sales agent's Banner or Colleague order, checked before it goes out

An AI sales agent prepares a subscription order for a college or university: Banner or Colleague, the modules, the deployment, the term and a discount. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the size bands, module dependencies, terms and discount limits and returns a verdict and every reason.

Scope: Ellucian Banner and Colleague orders with the Student, Financial Aid, Finance, Human Resources, Advancement, Degree Works, CRM Recruit, Experience and Analytics modules, SaaS, hosted or on-premise deployment, size band, term, discount and campuses. See everything this validator covers

This validator models Banner and Colleague subscription orders, not just the scenarios shown below.

  • Platform and deployment — Banner or Colleague, delivered as SaaS, hosted or on-premise, with the modules that are cloud-only and the SaaS minimum term.
  • Size bands — enrollment (FTE students) and employee counts checked against the size band chosen for the order.
  • Module dependencies — Financial Aid, Degree Works and CRM Recruit need Student; Advancement needs Finance and Banner.
  • Term, discount and approvals — terms up to seven years, a discount that needs sign-off above one level and is not allowed above another, and review of very large institutions and multi-campus orders.
  • Stated module count — the module count the order states against the modules it lists.

Any Banner or Colleague order that touches these rules is in scope — not only the scenarios pictured here. Not covered: Ethos, Journey, TouchNet, Talent Management, services and prices; this validator models no prices. Size bands, dependencies, terms and discount thresholds are modeled, not published by Ellucian.

The AI sales agentBuilds the subscription order from the conversation and says it is ready.
Predictika validatorChecks every rule at once and returns the verdict, each finding, and the computed figures.

7 scenarios in 3 groups · each group runs from the agent’s first attempt to the outcome · click one to switch

What the agent proposes

The AI sales agent says

Illustrative agent statement. The validator receives the structured request below.

Proposing agent: AI sales agent (scripted)

Before sending, the agent must call Recorded run

Predictika validator


      

        
        
        
      

      
Waiting for the call.

      
    

  

  
  

    
Where rules-based CPQ gets it wrong
Click to expand
    
Our assessment of what a typical rules-based CPQ (product rules, price rules and validation rules, as in Salesforce CPQ) does with the same cases. It is a judgment about effort and reliability, not a claim that nothing can be built.

    
What the validator caughtTypical rules-based CPQPredictika
Modules that depend on other modulesOnly with care
Each dependency is its own rule, and the rules multiply with the platforms the module is offered on.		The module list, the platform and the deployment are in one request and checked together on every call.
Size bands that limit the orderOnly with care
Enrollment and employee counts each need a rule against the chosen band, and the band limits sit beside the approval for very large institutions.		The institution's size and the chosen band are fields of the request and are compared on every call.
Terms that depend on the deploymentOnly with workarounds
The SaaS minimum term and the cloud-only modules both depend on the deployment, so each deployment needs its own set of term and module rules.		Deployment, term and modules are fields of the request and are checked against each other on every call.
Discount tiersOnly with workarounds
A discount above one level needs sign-off and above another is not allowed, so each tier is a separate rule that has to be kept in step.		The discount lands in the right tier on every call, whatever else is on the order.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as the order is saved or a step is completed, so a rep often fixes one error and meets the next.		All findings come back together, whatever order the choices arrived in.

  

  

    
What this validator catches: findings from its test cases
Click to expand
    
Every rule is checked on every call, in any order the choices arrive; one problem never hides another. Blocking findings stop the order; approval findings need a person to sign off; everything else is computed.

    
1The order has no modules.
2The institution has 8,000 FTE students; the selected size band allows 7,500.
3The institution has 3,500 employees on payroll; the selected size band allows 3,000.
4The Financial Aid module needs the Student module on the same order.
5Degree Works needs the Student module on the same order.
6CRM Recruit needs the Student module on the same order.
7The Advancement module posts gifts to the general ledger and needs the Finance module on the same order.
8In this model the Advancement module is offered on Banner only, not Colleague.
9Ellucian Experience and CRM Recruit are cloud-delivered; they cannot be ordered with an on-premise deployment.
10A SaaS subscription in this model has a minimum term of 3 years; the order has 2.
11Terms in this model are at most 7 years; the order has 8.
12The discount is 40 percent; no order in this model goes above 35 percent.
13A discount of 20 percent is above 15 percent and needs sales management sign-off.
14The order states 5 modules; the order lists 4.
1530,000 FTE students is a very large institution; a person should review sizing and implementation scope.
165 campuses on one order; a person should confirm the multi-campus setup.

  

  

    
Scenarios and results
Click to expand
    
The recorded result of the validator for each scenario on this page: the request the AI agent sent, the verdict, and every finding, in the order they appear above.

    
1. Platform, deployment and size: A Colleague on-premise order with Advancement and Experience

    
The agent's request. Please quote Ellucian Colleague as an on-premise deployment in the B2 size band for 8,000 FTE students and 2,000 employees, on a 5-year term, with Student, Finance, Human Resources, Advancement and Experience. The order has 6 modules. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Enrollment over the band. The institution has 8,000 FTE students; the selected size band allows 7,500.
    • 
      
  • Advancement only on Banner. In this model the Advancement module is offered on Banner only, not Colleague.
    • 
      
  • Experience and Recruit need SaaS or Hosted. Ellucian Experience and CRM Recruit are cloud-delivered; they cannot be ordered with an on-premise deployment.
    • 
      
  • Stated module count does not match. The order states 6 modules; the order lists 5.
    • 
    

    
2. Platform, deployment and size: The agent resubmits after fixing the errors

    
The agent's request. Please quote Ellucian Banner as a hosted deployment in the B3 size band for 8,000 FTE students and 2,000 employees, on a 5-year term, with Student, Finance, Human Resources, Advancement and Experience. The order has 5 modules. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 5 module(s) on a 5 year term; size band allows 20,000 FTE students and 8,000 employees.
    • 
    

    
3. Module dependencies and term: An order for Financial Aid, Degree Works and CRM Recruit on a short SaaS term with a deep discount

    
The agent's request. Please quote Ellucian Colleague as a SaaS deployment in the B2 size band for 6,000 FTE students and 1,500 employees, on a 2-year term, with Financial Aid, Degree Works and CRM Recruit, and a 40 percent discount. The order has 3 modules. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Financial Aid needs Student. The Financial Aid module needs the Student module on the same order.
    • 
      
  • Degree Works needs Student. Degree Works needs the Student module on the same order.
    • 
      
  • CRM Recruit needs Student. CRM Recruit needs the Student module on the same order.
    • 
      
  • SaaS term too short. A SaaS subscription in this model has a minimum term of 3 years; the order has 2.
    • 
      
  • Discount over the hard limit. The discount is 40 percent; no order in this model goes above 35 percent.
    • 
    

    
4. Module dependencies and term: The agent resubmits after fixing the errors

    
The agent's request. Please quote Ellucian Colleague as a SaaS deployment in the B2 size band for 6,000 FTE students and 1,500 employees, on a 3-year term, with Student, Financial Aid, Degree Works and CRM Recruit, and a 15 percent discount. The order has 4 modules. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 4 module(s) on a 3 year term; size band allows 7,500 FTE students and 3,000 employees.
    • 
    

    
5. Discount tiers: A Banner SaaS order with a deep discount

    
The agent's request. Please quote Ellucian Banner as a SaaS deployment in the B3 size band for 15,000 FTE students and 5,000 employees, on a 4-year term, with Student, Financial Aid, Finance, Human Resources, Advancement and Analytics, and a 40 percent discount. The order has 6 modules. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Discount over the hard limit. The discount is 40 percent; no order in this model goes above 35 percent.
    • 
    

    
6. Discount tiers: The agent resubmits after fixing the errors

    
The agent's request. Please quote Ellucian Banner as a SaaS deployment in the B3 size band for 15,000 FTE students and 5,000 employees, on a 4-year term, with Student, Financial Aid, Finance, Human Resources, Advancement and Analytics, and a 25 percent discount. The order has 6 modules. It is ready to send.

    
Verdict: Needs approval.

    
    
      
  • Discount between 15 and 35 percent. A discount of 25 percent is above 15 percent and needs sales management sign-off.
    • 
      
  • Order summary. 6 module(s) on a 4 year term; size band allows 20,000 FTE students and 8,000 employees.
    • 
    

    
7. Discount tiers: The agent resubmits after fixing the errors

    
The agent's request. Please quote Ellucian Banner as a SaaS deployment in the B3 size band for 15,000 FTE students and 5,000 employees, on a 4-year term, with Student, Financial Aid, Finance, Human Resources, Advancement and Analytics, and a 12 percent discount. The order has 6 modules. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 6 module(s) on a 4 year term; size band allows 20,000 FTE students and 8,000 employees.
    • 
    

  

  

    
Narration transcript
Click to expand
    
The text of the narrated walkthrough, step by step.

    
    
      
  1. Welcome to a demo of Predictika's Validator. An AI sales agent prepares a subscription order for a college or university: Banner or Colleague, the modules, the deployment, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the size bands, module dependencies, terms and discount limits and returns a verdict and every reason.
    2. 
      
  2. I will start with the first request by the agent.
    3. 
      
  3. The agent says. Please quote Ellucian Colleague as an on-premise deployment in the B2 size band for 8,000 FTE students and 2,000 employees. The order has 6 modules. The rest of the request is on screen.
    4. 
      
  4. Before it goes out, the agent calls the validator.
    5. 
      
  5. Blocked, with four findings, all in one answer. The institution has 8,000 FTE students; the selected size band allows 7,500. In this model the Advancement module is offered on Banner only, not Colleague. Ellucian Experience and CRM Recruit are cloud-delivered; they cannot be ordered with an on-premise deployment. The other one is on screen.
    6. 
      
  6. The validator returned four blocking findings in one answer. A check that reads one field at a time would pass most of this; it shows only when the choices in the order are checked against each other.
    7. 
      
  7. What this shows: for modules that depend on other modules, a rules-based CPQ is error-prone for this.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the platform to banner, the deployment to hosted, the size band to B3 and the module count stated on the order to 5.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. This time the validator clears it.
    12. 
      
  12. All four earlier findings are cleared.
    13. 
      
  13. Four problems caught in one call, fixed in one pass, and a clean order. That is the value of Predictika's validator: catching mistakes before they reach the customer. Now you can try the other scenarios yourself.
    14.