Predictika · Pay-as-you-go location data lookup quotes

An AI sales agent's lookup credit quote, checked before it goes out

An AI sales agent prepares a quote for pay-as-you-go location data lookups, with credit top-ups, lookup volumes, a trial and support terms. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the rate grid and the quote rules and returns a verdict and every reason.

Scope: Pay-as-you-go Loqate lookup quotes: address capture, address verify, email and phone validation, and Store Finder maps, geocoding and distances, bought as $100, $300 or $1,000 credit top-ups. See everything this validator covers

This validator models Loqate pay-as-you-go lookup quotes, not just the scenarios shown below.

  • Lookups and rates — address capture for the US and Canada and for other countries, address verify, email and phone validation, Store Finder maps, geocoding and distances, each at the published rate of its top-up tier.
  • Credit, trial and validity — the number of top-ups, whether the credit covers the expected usage, the 45-day free trial and the 12-month credit validity.
  • Dependencies, stated usage and reviews — distances that need geocoding, the usage stated on the quote against the computed one, and the reviews that service-level agreements, dedicated support, very large credit and credit likely to expire unused need.

Any Loqate pay-as-you-go lookup quote that touches these rules is in scope — not only the scenarios pictured here. Not covered: GBG identity verification, fraud products, data cleansing, bank verification and bespoke plans. Product names, per-lookup rates, the trial and the credit validity follow the public Loqate pricing page; the other rules and thresholds are modeled, not published by GBG.

The AI sales agentBuilds the lookup credit 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
Per-lookup rates that change with the top-up tierOnly with care
Eight lookup types each have three rates, and the tier decides which one applies.		The tier and every lookup volume are in one request, and the usage is computed on every call.
Credit that must cover the expected usageOnly with care
The credit is the number of top-ups times the tier value, and it has to be compared with a usage figure that depends on everything else.		The credit and the usage are computed and compared on every call.
Distances that need geocodingOnly with care
One lookup type depends on another being present for the same locations.		Both volumes are fields of one request and are checked together.
Trial length and credit validityOnly with workarounds
The trial and the validity are separate limits that must be kept in step with what the quote assumes.		The trial days and the validity assumed are checked on every call.
Stated usage and reviews of support terms and creditOnly with workarounds
The stated usage has to be recomputed whenever a volume changes, and each review depends on a different condition.		The stated usage is a field of the quote and is compared with the computed one, and every review comes back with the blocks.
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 quote; approval findings need a person to sign off; everything else is computed.

    
1Expected usage of 10,500 cents is more than the 10,000 cents of credit purchased.
2The quote has no lookups to price.
3A lookup quantity is below zero.
4Expected usage of 2,000 cents is half of the credit or less and unused credit expires after 12 months.
50 top-ups; at least 1 is needed.
6Expected usage of 2,500 cents is more than the 0 cents of credit purchased.
7A trial of 46 days is longer than the 45-day free trial.
8Credit is valid for 12 months; the quote assumes 13 months.
9Store Finder Distances lookups need Geocoding lookups for the same locations.
10The quote states usage of 83,499 cents; the rate grid gives 83,500 cents.
11The quote states usage of 83,501 cents; the rate grid gives 83,500 cents.
12Service-level agreements are part of bespoke pricing; a person should confirm it.
13Dedicated technical support is part of bespoke pricing; a person should confirm it.
14Credit of 500,000 cents is large enough to quote as a bespoke plan.
15Expected usage of 5,000 cents is half of the credit or less and unused credit expires after 12 months.
16A trial of 60 days is longer than the 45-day free trial.
17The quote states usage of 1 cents; the rate grid gives 430,000 cents.
18Credit of 600,000 cents is large enough to quote as a bespoke plan.

  

  

    
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. Credit and trial: A quote for address capture and distance lookups with a long trial and no top-up

    
The agent's request. Please quote pay-as-you-go lookups at the $100 top-up rates with no top-up purchased: 4,000 US and Canada address capture lookups and 15,000 distances lookups, with a 60-day free trial, and credit assumed valid for 13 months. The quote states expected usage of $270.50. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Top-up count. 0 top-ups; at least 1 is needed.
    • 
      
  • Credit does not cover usage. Expected usage of 27,050 cents is more than the 0 cents of credit purchased.
    • 
      
  • Trial longer than 45 days. A trial of 60 days is longer than the 45-day free trial.
    • 
      
  • Credit validity. Credit is valid for 12 months; the quote assumes 13 months.
    • 
      
  • Distances need Geocoding. Store Finder Distances lookups need Geocoding lookups for the same locations.
    • 
    

    
2. Credit and trial: The agent resubmits after fixing the errors

    
The agent's request. Please quote pay-as-you-go lookups at the $300 top-up rates with 2 top-ups purchased: 4,000 US and Canada address capture lookups, 15,000 geocoding lookups and 15,000 distances lookups. The quote states expected usage of $306.00. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Usage summary. 340 hundred lookups use 30,600 cents of 60,000 cents of credit at the selected top-up rates.
    • 
    

    
3. Usage and credit: A quote for 2,100 address capture lookups on a $100 top-up

    
The agent's request. Please quote pay-as-you-go lookups at the $100 top-up rates with 1 top-up purchased: 2,100 US and Canada address capture lookups. The quote states expected usage of $10.00. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Credit does not cover usage. Expected usage of 10,500 cents is more than the 10,000 cents of credit purchased.
    • 
      
  • Stated usage does not match. The quote states usage of 1,000 cents; the rate grid gives 10,500 cents.
    • 
    

    
4. Usage and credit: The agent resubmits after fixing the errors

    
The agent's request. Please quote pay-as-you-go lookups at the $100 top-up rates with 2 top-ups purchased: 2,100 US and Canada address capture lookups. The quote states expected usage of $105.00. It is ready to send.

    
Verdict: Allowed.

    
    
      
  • Usage summary. 21 hundred lookups use 10,500 cents of 20,000 cents of credit at the selected top-up rates.
    • 
    

    
5. Large purchase with support terms: A quote for six $1,000 top-ups with a service-level agreement and dedicated support

    
The agent's request. Please quote pay-as-you-go lookups at the $1,000 top-up rates with 6 top-ups purchased: 100,000 US and Canada address capture lookups, with a service-level agreement and dedicated technical support. The quote states expected usage of $4,000.00. It is ready to send.

    
Verdict: Blocked.

    
    
      
  • Stated usage does not match. The quote states usage of 400,000 cents; the rate grid gives 430,000 cents.
    • 
      
  • SLA needs a bespoke plan. Service-level agreements are part of bespoke pricing; a person should confirm it.
    • 
      
  • Dedicated support needs a bespoke plan. Dedicated technical support is part of bespoke pricing; a person should confirm it.
    • 
      
  • Large purchase. Credit of 600,000 cents is large enough to quote as a bespoke plan.
    • 
    

    
6. Large purchase with support terms: The agent resubmits after fixing the errors

    
The agent's request. Please quote pay-as-you-go lookups at the $1,000 top-up rates with 6 top-ups purchased: 100,000 US and Canada address capture lookups, with a service-level agreement and dedicated technical support. The quote states expected usage of $4,300.00. It is ready to send.

    
Verdict: Needs approval.

    
    
      
  • SLA needs a bespoke plan. Service-level agreements are part of bespoke pricing; a person should confirm it.
    • 
      
  • Dedicated support needs a bespoke plan. Dedicated technical support is part of bespoke pricing; a person should confirm it.
    • 
      
  • Large purchase. Credit of 600,000 cents is large enough to quote as a bespoke plan.
    • 
      
  • Usage summary. 1,000 hundred lookups use 430,000 cents of 600,000 cents of credit at the selected top-up rates.
    • 
    

  

  

    
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 quote for pay-as-you-go location data lookups, with credit top-ups, lookup volumes, a trial and support terms. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the rate grid and the quote rules 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 pay-as-you-go lookups at the $100 top-up rates. The quote states expected usage of $270.50. The rest of the request is on screen.
    4. 
      
  4. Before it goes out, the agent calls the validator.
    5. 
      
  5. Blocked, with five findings, all in one answer. 0 top-ups; at least 1 is needed. Expected usage of 27,050 cents is more than the 0 cents of credit purchased. A trial of 60 days is longer than the 45-day free trial. The other two are on screen.
    6. 
      
  6. The validator returned five 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 credit that must cover the expected usage, a rules-based CPQ is error-prone for this.
    8. 
      
  8. Now the agent resubmits.
    9. 
      
  9. The corrected request. It changes the top-up size to $300, the top-ups purchased to 2 and the stated usage to $306.00 (30,600 cents). It sets the geocoding lookups to 15,000. It removes the free trial and credit validity assumed.
    10. 
      
  10. It calls the validator again.
    11. 
      
  11. This time the validator clears it.
    12. 
      
  12. All five earlier findings are cleared.
    13. 
      
  13. Five 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. Now you can try the other scenarios yourself.
    14.