Predictika · Automotive build and price

An AI sales agent's Atlas order, checked before it goes out

An AI sales agent prepares a Volkswagen Atlas order for a fictional customer. Before it goes out, the agent sends it to Predictika’s validator, which checks it against the product and pricing rules, compares the net price the agent states with the one it computes, and returns a verdict with every reason.

Scope: 2026 Atlas build-and-price orders. See everything this validator covers

This validator models Atlas orders in the style of Volkswagen of America’s 2026 model year. It handles any build made of these trims, packages and options, not just the scenarios shown below.

  • Trim and drivetrain — which drivetrains, packages, paint, interiors and dealer accessories each trim can have, including when 4Motion is standard
  • Packages — prerequisites and exclusions, such as ventilated seats needing the leather package
  • Towing and payload — the factory hitch, the tow rating with and without the towing package, and payload
  • Seating — second-row captain chairs and what they change
  • Incentives — eligibility by customer type and trim, and which incentives can stack
  • Pricing and authority — employee and fleet pricing, and the discount a salesperson, manager or deal type allows
  • Order bank and lead time — closed order banks, allocation and the requested build week
  • The stated net price — the net price the agent states against the one the validator computes

It does not cover financing, trade-ins, taxes or fees, or other models.

The AI sales agentBuilds the quote from the conversation and says it is ready.
Predictika validatorChecks the quote against the product and contract rules and returns every finding it has.

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) does with the same cases. It is a judgment about effort and reliability, not a claim that nothing can be built.

    
What the validator catchesTypical rules-based CPQPredictika
Towing needs the hitch and the right packageOnly with care
The tow rating depends on the package, and the hitch and trailer weight interact, so several rules must agree.		The hitch, the rating and the trailer weight are checked together.
Package prerequisites and exclusionsOnly with care
Each prerequisite and exclusion is its own rule, and trims change which apply.		They are properties of the items and are checked on every call.
Incentive eligibility and stackingOnly with workarounds
Eligibility by customer and trim plus stacking limits multiply the rules.		Each incentive is checked against the customer, the trim and the others selected.
Discount authority by role and deal typeOnly with care
Approval thresholds usually live in a separate approval process after the quote is saved.		The salesperson, manager and maximum limits are evaluated with the quote.
The net price the agent statesOnly with workarounds
Comparing a stated price with a computed one needs a separate rule per layout, and price rules rarely run on stated figures.		The stated net price is a field in the request and is checked against the computed price.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as the quote is saved, so a rep fixes one error and meets the next.		All findings come back together, whatever order the choices arrived in.

  

  

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

    
RuleEffectWhat it checks
Trim availabilityblockDrivetrain, package, paint, interior and accessory must be offered on the chosen trim.
4Motion requirementsblockTrims where 4Motion is standard or must be selected.
Package prerequisites and exclusionsblockA package needs its prerequisite and cannot be combined with an excluded one.
Paint and interior pairingblockInterior colors that do not go with the chosen paint.
Hitch and tow ratingblockA planned trailer needs the factory hitch and must be within the build’s tow rating.
PayloadblockThe load must be within the payload of the build.
Incentive eligibility and stackingblockCustomer type, trim and whether incentives can be combined.
Employee and fleet pricingblockWho can receive these prices.
Discount authorityneeds approval or blockAbove the salesperson’s limit needs a manager; above the deal-type maximum is blocked.
Order bank and lead timeblockClosed order banks and a requested build week that is too soon.
Stated net priceblockThe net price the agent states must be within $1 of the computed price.

  

  

    
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. Towing: Atlas SEL, 4,500 lb trailer

    
The agent's request. The Atlas SEL in Pure White with captain chairs and roof cross bars is ready for Dana Whitfield. She tows a 4,500 lb camper. Net price $50,121.80, Owner Loyalty applied. Ready to release to the factory.

    
Verdict: Blocked.

    
    
      
  • Trailer hitch missing. A 4500 lb trailer is planned but no factory Class III trailer hitch is on the build.
    • 
      
  • Tow rating exceeded. The planned 4500 lb trailer exceeds the 2000 lb tow rating of this build (2.0T 269 hp 8-speed automatic with 4Motion all-wheel drive (standard)).
    • 
    

    
2. Towing: Same build, towing package added

    
The agent's request. I added the Towing Package with the Class III hitch, which raises the rating to 5,000 lb. Same Atlas SEL, net price $50,660.80. Ready to release.

    
Verdict: Allowed.

    
3. Discount authority: 8 percent discount

    
The agent's request. Same Atlas SEL, Pure White, captain chairs and cross bars. To close with Marcus Reyes today I included an 8 percent dealer discount. Net price $47,082.20.

    
Verdict: Blocked.

    
    
      
  • Discount above the maximum. 8 percent discount exceeds the 6 percent maximum for a Retail deal.
    • 
    

    
4. Discount authority: 5 percent discount

    
The agent's request. I lowered the discount to 5 percent. Net price $48,602.00.

    
Verdict: Needs approval.

    
    
      
  • Discount needs manager approval. 5 percent discount is above the 2 percent a salesperson can give; needs manager approval (max 6 percent).
    • 
    

    
5. Discount authority: 2 percent discount

    
The agent's request. I set the discount to 2 percent, which is within my authority. Net price $50,121.80.

    
Verdict: Allowed.

    
6. Stated net price: Atlas SE with Technology

    
The agent's request. The Atlas SE with Technology with 4Motion in Mountain Lake Blue, with leather and ventilated front seats and the SE Bonus Cash, is ready for Priya Nair. Net price $49,450.00.

    
Verdict: Blocked.

    
    
      
  • Stated net price does not match. The quoted net price of $49450.0 does not match the $48450.0 the validator computed.
    • 
    

    
7. Stated net price: Same build, price restated

    
The agent's request. I restated the net price with the SE Bonus Cash applied: $48,450.00.

    
Verdict: Allowed.

  

  

    
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 Volkswagen Atlas order for a fictional customer. Before it goes out, the agent sends it to Predictika’s validator, which checks it against the product and pricing rules, compares the net price the agent states with the one it computes, and returns a verdict with every reason.
    2. 
      
  2. I will start with the first request by the agent.
    3. 
      
  3. The agent says. The Atlas SEL in Pure White with captain chairs and roof cross bars is ready for Dana Whitfield. She tows a 4,500 lb camper. The rest of the request is on screen. Net price $50,121.80, Owner Loyalty applied.
    4. 
      
  4. Before it goes out, the agent calls the validator.
    5. 
      
  5. Blocked, with two findings, all in one answer. A 4500 lb trailer is planned but no factory Class III trailer hitch is on the build. The planned 4500 lb trailer exceeds the 2000 lb tow rating of this build (2.0T 269 hp 8-speed automatic with 4Motion all-wheel drive (standard)).
    6. 
      
  6. The validator returned two 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 quote are checked against each other.
    7. 
      
  7. What this shows: for towing needs the hitch and the right package, a rules-based CPQ is error-prone for this.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the pkg1 to Towing Package with Class III hitch and the stated net as to 50,660.8.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. This time the validator clears it.
    12. 
      
  12. The validator also returns the line-item quote. It starts with Atlas SEL and Towing Package with Class III hitch; the other three lines are on screen.
    13. 
      
  13. Both earlier findings are cleared.
    14. 
      
  14. Two problems caught in one call, fixed in one pass, and a clean quote. That is the value of Predictika's validator: catching mistakes before they reach the customer.
    15.