Predictika · Professional cleaning products orders

An AI sales agent's bulk cleaning products order, checked before it goes out

An AI sales agent prepares a bulk order of professional cleaning products: a product, a pack size, a number of units, a dispensing system, a discount, payment terms and a stated shipping weight. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the P&G Professional line and returns a verdict and every reason.

Scope: P&G Professional foodservice and janitorial orders: Dawn, Tide, Downy, Comet, Mr. Clean and Spic and Span, in 38 oz cases, 1 gal cases and 5 gal pails, with dispensing system, minimum order, truckload weight, discount and payment terms. See everything this validator covers

This validator models P&G Professional bulk orders, not just the scenarios shown below.

  • Product and pack — which products come in which pack sizes, and the dispensing system that closed-loop packs need, new or already at the customer.
  • Volume and weight — a four-unit minimum, the weight of each pack and the 40,000 lb truckload limit, and the stated shipping weight.
  • Discount and terms — discounts above 8 percent need approval and above 15 percent are not allowed, and payment terms over 45 days need credit approval.

Any P&G Professional bulk order that touches these rules is in scope — not only the scenarios pictured here. Not covered: consumer lines, other P&G brands and prices. The product names and common pack sizes are public; the pack availability for each product, the dispenser rule, weights, the truckload and minimum limits, and the discount and payment thresholds are modeled and are not published by P&G (see the note at the foot of the page). The validator models no prices.

The AI sales agentBuilds the bulk 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.

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
Packs offered for only some productsOnly with care
Each product is made in its own list of pack sizes, so a rule keyed to the product has to hold the list for every one.		The product and pack are fields of one order and are checked together on every call.
A dispenser the pack depends onOnly with care
Closed-loop packs need a dispensing system, either new on the order or already at the customer, and the two cases are handled differently.		The pack, the new dispenser and the existing dispenser are checked together on every call.
A stated shipping weightOnly with workarounds
The agent states a weight, and a rules engine has to rebuild it from the units and the pack weights to catch a mismatch.		The weight is computed from the ordered units and compared with the stated figure 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 order; approval findings need a person to sign off; everything else is computed.

    
1The chosen product is not offered in the chosen pack size in this model.
2Closed-loop packs need a dispensing system and the order has none, new or existing.
3The order relies on a dispensing system the customer already has; a person should confirm it fits the closed-loop pack.
4The order has 3 units; at least 4 are required.
5The order weighs 40,040 lb; a single truckload is limited to 40,000 lb.
6The order states 3,800 lb; the ordered units weigh 4,000 lb.
7The discount is 12 percent; discounts above 8 percent need account manager approval.
8The discount is 16 percent; discounts above 15 percent are not allowed.
9Payment terms of 60 days need credit approval; terms over 45 days are not standard.

  

  

    
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. Product, pack, dispenser and discount: A Tide Professional order of 3 one-gallon cases with no dispenser and a 16 percent discount

    
The agent's request. Please order 3 cases (1 gal) of Tide Professional liquid detergent, with no dispensing system, at a 16 percent discount and 30 day payment terms, with a total weight of 120 lb. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Pack size not offered for the product. The chosen product is not offered in the chosen pack size in this model.
    • 
      
  • Closed-loop product needs a dispenser. Closed-loop packs need a dispensing system and the order has none, new or existing.
    • 
      
  • Below minimum order. The order has 3 units; at least 4 are required.
    • 
      
  • Discount above the ceiling. The discount is 16 percent; discounts above 15 percent are not allowed.
    • 
    

    
2. Product, pack, dispenser and discount: The agent resubmits after fixing the errors

    
The agent's request. Please order 100 pails (5 gal) of Tide Professional liquid detergent, with a new dispensing system on the order, at an 8 percent discount and 30 day payment terms, with a total weight of 4,500 lb. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 100 unit(s) weighing 4,500 lb in total.
    • 
    

    
3. Pack and truckload weight: A Comet Professional disinfectant order of 1,000 five-gallon pails

    
The agent's request. Please order 1,000 pails (5 gal) of Comet Professional disinfectant, with a new dispensing system on the order, at a 5 percent discount and 30 day payment terms, with a total weight of 45,000 lb. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Pack size not offered for the product. The chosen product is not offered in the chosen pack size in this model.
    • 
      
  • Over truckload weight. The order weighs 45,000 lb; a single truckload is limited to 40,000 lb.
    • 
    

    
4. Pack and truckload weight: The agent resubmits after fixing the errors

    
The agent's request. Please order 900 cases (1 gal) of Comet Professional disinfectant, with a new dispensing system on the order, at a 5 percent discount and 30 day payment terms, with a total weight of 36,000 lb. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 900 unit(s) weighing 36,000 lb in total.
    • 
    

    
5. Existing dispenser, discount, terms and weight: A Dawn Professional order relying on the customer's own dispenser, with a 12 percent discount, 60 day terms and a stated weight that is off

    
The agent's request. Please order 100 cases (1 gal) of Dawn Professional pot and pan detergent, for use with the dispensing system the customer already has, at a 12 percent discount and 60 day payment terms, with a total weight of 3,800 lb. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Stated weight does not match. The order states 3,800 lb; the ordered units weigh 4,000 lb.
    • 
      
  • Discount needs approval. The discount is 12 percent; discounts above 8 percent need account manager approval.
    • 
      
  • Long payment terms. Payment terms of 60 days need credit approval; terms over 45 days are not standard.
    • 
      
  • Existing dispenser to confirm. The order relies on a dispensing system the customer already has; a person should confirm it fits the closed-loop pack.
    • 
    

    
6. Existing dispenser, discount, terms and weight: The agent resubmits after fixing the errors

    
The agent's request. Please order 100 cases (1 gal) of Dawn Professional pot and pan detergent, with a new dispensing system on the order, at an 8 percent discount and 45 day payment terms, with a total weight of 4,000 lb. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Order summary. 100 unit(s) weighing 4,000 lb in total.
    • 
    

  

  

    
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 bulk order of professional cleaning products: a product, a pack size, a number of units, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the P&G Professional line 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 order 3 cases (1 gal) of Tide Professional liquid detergent, with no dispensing system, at a 16 percent discount and 30 day payment terms. 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 four findings, all in one answer. The chosen product is not offered in the chosen pack size in this model. Closed-loop packs need a dispensing system and the order has none, new or existing. The order has 3 units; at least 4 are required. 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 packs offered for only some products, a rules-based CPQ is error-prone for this.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the pack to 5 gal pail, the units to 100, the new dispenser on order to yes, the discount to 8 percent and the stated weight to 4,500 lb.
    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.