Predictika · Data platform subscriptions

An AI sales agent's Morningstar Direct subscription, checked before it goes out

An AI sales agent prepares a subscription quote for Morningstar Direct: the client type, the data scope, the modules, web access, the Excel Add-In seats, index-provider entitlements, the number of users and the term. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the subscription rules and returns a verdict, every reason and the module count it computed.

Scope: Morningstar Direct subscriptions: Research, Performance Reporting, Asset Allocation, Portfolio Management, Portfolio Analysis, Markets, Asset Flows, Presentation Studio and Risk Model Analysis modules, local or global databases, web access, Excel Add-In seats, index-provider entitlements, users and term. See everything this validator covers

This validator models subscriptions of this kind, not just the scenarios shown below.

  • Modules and what they need — Portfolio Analysis needs Portfolio Management, Risk Model Analysis needs Portfolio Management, global databases and web access, Presentation Studio needs Performance Reporting, and Asset Flows needs global databases.
  • Users, term and Excel seats — between 1 and 500 users, a term of 1, 2 or 3 years, Excel Add-In seats only with the add-in and never more than the users, and academic subscriptions without Portfolio Management or index-provider entitlements.
  • Approvals — 100 users or more, index-provider entitlements to set up with a Customer Success Manager, and a bundle of seven or more modules, each for a person to review.

Any subscription that touches these rules is in scope — not only the scenarios pictured here. The module names and the existence of the Excel Add-In, the web version and index-provider entitlements follow public Morningstar information; every dependency, user limit, term, seat limit, academic limit and approval threshold is modeled from comparable data-platform subscriptions and is not published by Morningstar. No prices are modeled. Not covered: prices, other Morningstar products and contract terms.

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

6 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
Portfolio Analysis, Risk Model Analysis, Presentation Studio and Asset Flows each need other modules or a data scope, so every rule has to read the other choices.		The modules, the data scope and web access are checked together on every call.
Seats that depend on an add-in and on usersOnly with care
Excel seats make sense only with the add-in and never exceed the users, so a seat rule must read two other fields.		The add-in, the seats and the users are checked together on every call.
A module count that triggers a reviewOnly with workarounds
The count is one for Research plus every selected module, so it must be computed before it is compared with the review threshold.		The module count is computed and compared with the threshold on every call.
Limits that depend on the client typeOnly with workarounds
Academic subscriptions exclude some modules and entitlements, so the client type changes what the other choices may be.		The client type and every module are checked together on every call.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as 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 offer; approval findings need a person to sign off; everything else is computed.

    
1The Portfolio Analysis module needs the Portfolio Management module.
2The Risk Model Analysis module needs the Portfolio Management module and the Global Databases module.
3Risk Model Analysis runs in the web-based version, so web access must be included.
4The Presentation Studio module needs the Performance Reporting module.
5The Asset Flows module needs the Global Databases module.
65 Excel Add-In seats are quoted without the Excel Add-In.
7The Excel Add-In is quoted with 0 seats; it needs at least 1.
830 Excel Add-In seats exceed the 25 users on the subscription.
9Academic subscriptions do not include Portfolio Management or index-provider entitlements.
10Index-provider data needs entitlement privileges set up with the Customer Success Manager.
11A subscription needs between 1 and 500 users; the quote has 0 users.
12A subscription needs between 1 and 500 users; the quote has 501 users.
13A subscription for 501 users is an enterprise deal, and a person should review it.
14A subscription for 100 users is an enterprise deal, and a person should review it.
15Subscription terms are 1, 2 or 3 years; the quote has 4 years.
16A bundle of 7 modules should be reviewed for fit.
174 Excel Add-In seats are quoted without the Excel Add-In.
18A subscription for 150 users is an enterprise deal, and a person should review it.

  

  

    
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. Dependencies and academic limits: An academic subscription on local databases with Portfolio Analysis, Risk Model Analysis, Asset Flows and Presentation Studio, Excel seats without the add-in and index-provider entitlements

    
The agent's request. Please quote a Morningstar Direct subscription for an academic client with the Research, Portfolio Analysis, Asset Flows, Presentation Studio and Risk Model Analysis modules on local databases, for 40 users and a 2-year term, with no web access, 4 Excel Add-In seats and index-provider entitlements. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Portfolio Analysis needs Portfolio Management. The Portfolio Analysis module needs the Portfolio Management module.
    • 
      
  • Risk Model Analysis needs Portfolio Management and global databases. The Risk Model Analysis module needs the Portfolio Management module and the Global Databases module.
    • 
      
  • Risk Model Analysis needs web access. Risk Model Analysis runs in the web-based version, so web access must be included.
    • 
      
  • Presentation Studio needs Performance Reporting. The Presentation Studio module needs the Performance Reporting module.
    • 
      
  • Asset Flows needs global databases. The Asset Flows module needs the Global Databases module.
    • 
      
  • Excel seats without the add-in. 4 Excel Add-In seats are quoted without the Excel Add-In.
    • 
      
  • Academic subscription limits. Academic subscriptions do not include Portfolio Management or index-provider entitlements.
    • 
      
  • Index-provider entitlements to configure. Index-provider data needs entitlement privileges set up with the Customer Success Manager.
    • 
    

    
2. Dependencies and academic limits: The agent resubmits after fixing the errors

    
The agent's request. Please quote a Morningstar Direct subscription for an asset manager client with the Research, Performance Reporting, Portfolio Management, Portfolio Analysis, Presentation Studio and Risk Model Analysis modules on global databases, for 40 users and a 2-year term, with web access. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Subscription summary. 40 user(s), 6 module(s) including Research, 2 year term.
    • 
    

    
3. Users, term and Excel seats: A 600-user, 4-year subscription with the Excel Add-In and no seats

    
The agent's request. Please quote a Morningstar Direct subscription for an asset manager client with the Research, Performance Reporting and Portfolio Management modules on global databases, for 600 users and a 4-year term, with no web access and the Excel Add-In with 0 seats. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Users out of range. A subscription needs between 1 and 500 users; the quote has 600 users.
    • 
      
  • Term not available. Subscription terms are 1, 2 or 3 years; the quote has 4 years.
    • 
      
  • Excel Add-In without seats. The Excel Add-In is quoted with 0 seats; it needs at least 1.
    • 
      
  • Large user count. A subscription for 600 users is an enterprise deal, and a person should review it.
    • 
    

    
4. Users, term and Excel seats: The agent resubmits after fixing the errors

    
The agent's request. Please quote a Morningstar Direct subscription for an asset manager client with the Research, Performance Reporting and Portfolio Management modules on global databases, for 80 users and a 3-year term, with no web access and the Excel Add-In with 8 seats. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Subscription summary. 80 user(s), 4 module(s) including Research, 3 year term.
    • 
    

    
5. Seats, web access and client type: An academic subscription for 12 users with Portfolio Management, Risk Model Analysis and 20 Excel Add-In seats

    
The agent's request. Please quote a Morningstar Direct subscription for an academic client with the Research, Performance Reporting, Portfolio Management and Risk Model Analysis modules on global databases, for 12 users and a 2-year term, with no web access and the Excel Add-In with 20 seats. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Risk Model Analysis needs web access. Risk Model Analysis runs in the web-based version, so web access must be included.
    • 
      
  • More Excel seats than users. 20 Excel Add-In seats exceed the 12 users on the subscription.
    • 
      
  • Academic subscription limits. Academic subscriptions do not include Portfolio Management or index-provider entitlements.
    • 
    

    
6. Seats, web access and client type: The agent resubmits after fixing the errors

    
The agent's request. Please quote a Morningstar Direct subscription for an institutional investor client with the Research, Performance Reporting, Portfolio Management and Risk Model Analysis modules on global databases, for 25 users and a 2-year term, with web access and the Excel Add-In with 10 seats. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Subscription summary. 25 user(s), 5 module(s) including Research, 2 year term.
    • 
    

  

  

    
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 quote for Morningstar Direct: the client type, the data scope, the modules, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the subscription rules and returns a verdict, every reason and the module count it computed.
    2. 
      
  2. I will start with the first request by the agent.
    3. 
      
  3. The agent says. Please quote a Morningstar Direct subscription for an academic client with the Research, Portfolio Analysis, Asset Flows, Presentation Studio and Risk Model Analysis modules on local databases. It is ready to send. The rest of the request is on screen.
    4. 
      
  4. Before it goes out, the agent calls the validator.
    5. 
      
  5. Blocked, with seven findings and one approval, all in one answer. The Portfolio Analysis module needs the Portfolio Management module. The Risk Model Analysis module needs the Portfolio Management module and the Global Databases module. Risk Model Analysis runs in the web-based version, so web access must be included. The other four are on screen.
    6. 
      
  6. The validator returned seven blocking findings and one approval 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 offer are checked against each other.
    7. 
      
  7. What this shows: for seats that depend on an add-in and on users, a rules-based CPQ is error-prone for this.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the client type to asset manager, the data scope to global databases, the modules to Research, Performance Reporting, Portfolio Management, Portfolio Analysis, Presentation Studio and Risk Model Analysis and the web access to yes. Two more changes are on screen.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. This time the validator clears it.
    12. 
      
  12. All eight earlier findings are cleared.
    13. 
      
  13. Eight problems caught in one call, fixed in one pass, and a clean offer. That is the value of Predictika's validator: catching mistakes before the offer goes out. Now you can try the other scenarios yourself.
    14.