Predictika · E-bike orders

An AI sales agent's e-bike order, checked before it goes out

An AI sales agent prepares an order for Gazelle e-bikes: a model, frame type and size, battery, racks, a child seat, a care plan and a quantity for one rider or a fleet. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of how the bikes and options fit together and returns a verdict, every reason and the totals it computed.

Scope: Gazelle e-bikes (Medeo T9, Arroyo C5 Elite, Ultimate C8, C380 and T10): frame type and size against rider height, battery, total load, racks, child seat, second battery, care plan, fleet size and discount. See everything this validator covers

This validator models Gazelle e-bike orders, not just the scenarios shown below.

  • Model, frame and battery — the five models, the frame types each is modeled with, and the batteries each model is modeled with.
  • Fit and load — frame size against rider height, and the rider, rack cargo and child together against the most the model allows.
  • Racks and child seat — a rack limit, a rear rack for the child seat, and a child weight limit.
  • Second battery and care plan — a second battery only with the largest battery, and the care plan options.
  • Quantity, discount and total — fleet orders and discounts above what a sales agent may give go to a person, and the total the order states is compared with the computed total.

Any Gazelle e-bike order that touches these rules is in scope — not only the scenarios pictured here. Not covered: other Gazelle models, accessories beyond those listed, dealer terms and shipping. Which frame types and batteries each model has, the size ranges, the load limits and all prices are modeled from comparable e-bikes, not published by Gazelle.

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

6 scenarios in 4 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
A frame size that must fit the riderOnly with care
Each size covers a height range, so a lookup of the range per size and a rule comparing it with the rider are needed.		The rider height and the size are in one request and are compared directly on every call.
A total load across rider, cargo and childOnly with workarounds
The load is the sum of several fields, so a separate calculation is needed before it can be compared with each model's limit.		The weights are fields of the request and are summed and compared with the model's limit in the same pass.
Options that depend on other optionsOnly with care
A child seat needs the rear rack and a second battery needs the largest battery, so each is its own rule that reads another line.		The model, battery, racks and seat are checked together on every call.
A fleet order with a discountOnly with care
The fleet size and the discount each trigger their own approval, so rules have to be written for each and kept in step as limits change.		The quantity and the discount are fields of the request and each is checked on every call.
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 quote; approval findings need a person to sign off; everything else is computed.

    
1The order has 0 bikes; at least 1 is needed.
2The chosen battery is not offered on this model.
3This model is not offered with a mid-step frame.
4The Arroyo C5 Elite is not offered with a low-step frame.
5A rider of 174 cm does not fit this frame size; it fits riders from 175 to 193 cm.
6The total load of 161 kg (rider, rack cargo and child) is above the 160 kg this model allows.
7Rack cargo of 28 kg is above the 27 kg rack limit.
8Rack cargo of 10 kg is listed but no rear rack is ordered.
9The child seat mounts on the rear rack; order the rear rack too.
10A child seat needs the safety notice acknowledged by the customer.
11A child of 23 kg is above the 22 kg child seat limit.
12A child of 15 kg is listed but no child seat is ordered.
13The second battery is offered only with the 625 Wh battery.
14A discount of 11 percent is above the 10 percent a sales agent may give.
15A discount of 26 percent is outside the 0 to 25 percent range.
16An order of 10 bikes is a fleet order and needs dealer program review.
17The quote states a total of 4,899 dollars; the modeled price list gives 4,999 dollars.

  

  

    
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. Fleet order, frame and discount: A fleet of Medeo T9 bikes with a mid-step frame and a large discount

    
The agent's request. Please quote 12 Medeo T9 bikes with a mid-step frame in size 49 and the 545 Wh battery, for a rider of 160 cm and 65 kg, with a front rack, and a 15 percent discount, for a total of 30,480 dollars. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Mid-step frame. This model is not offered with a mid-step frame.
    • 
      
  • Discount above 10 percent. A discount of 15 percent is above the 10 percent a sales agent may give.
    • 
      
  • Fleet order. An order of 12 bikes is a fleet order and needs dealer program review.
    • 
    

    
2. Fleet order, frame and discount: The agent resubmits after fixing the errors

    
The agent's request. Please quote 12 Medeo T9 bikes with a low-step frame in size 49 and the 545 Wh battery, for a rider of 160 cm and 65 kg, with a front rack, and a 10 percent discount, for a total of 32,280 dollars. It is ready to send.

    
Verdict: Needs approval.

    
    
      
  • Fleet order. An order of 12 bikes is a fleet order and needs dealer program review.
    • 
      
  • Order summary. 12 bike(s) at 2,690 dollars each after discount, estimated total of 32,280 dollars at modeled prices, total load 65 kg of 140 kg allowed.
    • 
    

    
3. Frame size and rider: An Arroyo C5 Elite for a tall rider

    
The agent's request. Please quote 1 Arroyo C5 Elite bike with a high-step frame in size 53 and the 500 Wh battery, for a rider of 190 cm and 75 kg, with the 4-year care plan, for a total of 4,348 dollars. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Frame size does not fit rider. A rider of 190 cm does not fit this frame size; it fits riders from 163 to 182 cm.
    • 
    

    
4. Frame size and rider: The agent resubmits after fixing the errors

    
The agent's request. Please quote 1 Arroyo C5 Elite bike with a high-step frame in size 57 and the 500 Wh battery, for a rider of 190 cm and 75 kg, with the 4-year care plan, for a total of 4,348 dollars. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 1 bike(s) at 4,348 dollars each after discount, estimated total of 4,348 dollars at modeled prices, total load 75 kg of 150 kg allowed.
    • 
    

    
5. Load and child seat: A family order with rack cargo and a child seat on an Ultimate C380

    
The agent's request. Please quote 1 Ultimate C380 bike with a high-step frame in size 57 and the 625 Wh battery, for a rider of 185 cm and 125 kg, with a rear rack carrying 25 kg, a child seat for a child of 18 kg, for a total of 5,299 dollars. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Total load above limit. The total load of 168 kg (rider, rack cargo and child) is above the 160 kg this model allows.
    • 
      
  • Child seat. A child seat needs the safety notice acknowledged by the customer.
    • 
    

    
6. Battery: An Ultimate C8 with a 400 Wh battery

    
The agent's request. Please quote 1 Ultimate C8 bike with a low-step frame in size 53 and the 400 Wh battery, for a rider of 170 cm and 75 kg, with the 4-year care plan, for a total of 4,348 dollars. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Battery not offered. The chosen battery is not offered on this model.
    • 
    

  

  

    
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 an order for Gazelle e-bikes: a model, frame type and size, battery, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of how the bikes and options fit together and returns a verdict, every reason and the totals it computed.
    2. 
      
  2. I will start with the first request by the agent.
    3. 
      
  3. The agent says. Please quote 12 Medeo T9 bikes with a mid-step frame in size 49 and the 545 Wh battery, for a rider of 160 cm and 65 kg. 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 one finding and two approvals, all in one answer. This model is not offered with a mid-step frame.
    6. 
      
  6. The validator returned one blocking finding and two approvals 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 a fleet order with a discount, a rules-based CPQ is error-prone for this.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the frame type to low step, the discount to 10 percent and the total stated on the order to 32,280 dollars.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. Not blocked, but held for a person.
    12. 
      
  12. The one blocking finding is cleared; one approval remains for a person: fleet order.
    13. 
      
  13. Three problems caught in one call, fixed in one pass, and the corrected quote now goes to a person for approval. That is the value of Predictika's validator: catching mistakes before they reach the customer. Now you can try the other scenarios yourself.
    14.