Predictika · Technology financing quotes

An AI sales agent's technology financing quote, checked before it goes out

An AI sales agent prepares a financing quote for a customer buying IT hardware, software and services: a financing program, a lease or loan structure, a term, a credit tier and a monthly payment. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the financing programs and returns a verdict, every reason and the monthly payment it computed.

Scope: ePlus financing programs (Technology Financing, Vendor Financing, PO Financing, subscription services financing, As a Service and technology refresh) on lease, loan and usage-based structures, for commercial, federal and state and local customers. See everything this validator covers

This validator models technology financing quotes of this kind, not just the scenarios shown below.

  • Programs and structures — which structure each program allows: fair market value lease, dollar out lease, installment loan or usage-based, with As a Service on usage-based only and technology refresh on a fair market value lease for 24, 36 or 48 months.
  • Amount, term and mix — the financed amount range, the PO Financing minimum, the terms offered (12, 24, 36, 48 and 60 months), subscription financing for software and services only, and the limits on software and services in a lease.
  • Customer and credit — the non-appropriation clause a federal or state and local customer needs on a lease, the credit tier surcharge, and the approvals for an unrated customer or a large transaction.
  • Payment — the monthly payment the quote states against the payment computed from the modeled factors.

Any technology financing quote that touches these rules is in scope — not only the scenarios pictured here. Not covered: taxes, fees, residual values and equipment-specific terms. The validator shows a computed monthly payment but no line-item table, and all payments are illustrative estimates.

The AI sales agentBuilds the financing 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
Programs that allow only certain structures and termsOnly with care
Technology refresh allows only a fair market value lease for 24 to 48 months, and usage-based payments go with As a Service only, so each program needs its own set of rules on structure and term.
The program, structure and term are fields of one request and are checked together on every call.
Limits on software and services that depend on the structureOnly with care
A fair market value lease and a dollar out lease allow different shares of software and services, so each structure needs its own percentage rule.
The amounts and the structure are in one request and the shares are compared on every call.
A clause that depends on the customer and the structureOnly with care
A federal or state and local customer needs the non-appropriation clause on leases but not on a loan, so a rule must look at the sector and the structure together.
The sector, the structure and the clause are fields of the request and are checked together.
A stated payment that must match the computed oneOnly with workarounds
The payment comes from a factor that depends on the structure, the term and the credit tier, so a separate rule is needed to recompute it and compare it with the stated figure.
The stated payment is a field of the request and is compared with the computed payment on every call.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as the quote 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.

1PO financing starts at a modeled 25,000 dollars, but the total financed amount is 20,000 dollars.
2Subscription services financing covers software and services only, but the quote includes 80,000 dollars of hardware.
3Usage-based payments go with the As a Service program only, and the As a Service program is quoted on the usage-based structure only.
4A technology refresh program is quoted as a fair market value lease for 24, 36 or 48 months.
5The total financed amount of 4,900 dollars is below the modeled 5,000 dollar minimum.
6A total financed amount of 500,100 dollars is above 500,000 dollars and needs credit committee approval.
7A total financed amount of 5,000,000 dollars is above 500,000 dollars and needs credit committee approval.
8The total financed amount of 5,000,100 dollars is above 5,000,000 dollars; larger transactions are structured individually, so no modeled factor applies.
9A total financed amount of 5,000,100 dollars is above 500,000 dollars and needs credit committee approval.
10A term of 18 months is not offered; the modeled terms are 12, 24, 36, 48 and 60 months.
11A term of 72 months is not offered; the modeled terms are 12, 24, 36, 48 and 60 months.
12The customer has no credit rating on file, so credit review must approve the quote.
13Software and services are between 50 and 60 percent of a fair market value lease, which needs approval.
14Software and services are more than 60 percent of the financed amount, above the modeled limit for a fair market value lease.
15Software and services are more than 35 percent of the financed amount, above the modeled limit for a dollar out lease.
16A federal or state and local customer needs the non-appropriation clause on a lease or usage-based structure, and the quote does not include it.
17The quote states a monthly payment of 306,900 cents; the modeled factor gives 307,000 cents.
18The quote states a monthly payment of 307,001 cents; the modeled factor gives 307,000 cents.
19The quote states a monthly payment of 307,000 cents; the modeled factor gives 327,000 cents.
20The total financed amount of 3,000 dollars is below the modeled 5,000 dollar minimum.
21A total financed amount of 700,000 dollars is above 500,000 dollars and needs credit committee approval.

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. Technology refresh for a federal customer: A technology refresh quote for a federal customer on a dollar out lease

The agent's request. Please quote a technology refresh program for a federal customer. The structure is a dollar out lease over 36 months, with 60,000 dollars of hardware, 25,000 dollars of software and 15,000 dollars of services, credit Tier 1, and the non-appropriation clause not included. The quote states a monthly payment of 307,000 cents, which is 3,070 dollars. It is ready to send.

Verdict: Blocked.

  • Technology refresh rules. A technology refresh program is quoted as a fair market value lease for 24, 36 or 48 months.
  • Soft costs too high for a dollar out lease. Software and services are more than 35 percent of the financed amount, above the modeled limit for a dollar out lease.
  • Non-appropriation clause missing. A federal or state and local customer needs the non-appropriation clause on a lease or usage-based structure, and the quote does not include it.
  • Stated payment does not match. The quote states a monthly payment of 307,000 cents; the modeled factor gives 319,000 cents.

2. Technology refresh for a federal customer: The agent resubmits after fixing the errors

The agent's request. Please quote a technology refresh program for a federal customer. The structure is a fair market value lease over 36 months, with 80,000 dollars of hardware, 10,000 dollars of software and 10,000 dollars of services, credit Tier 1, and the non-appropriation clause included. The quote states a monthly payment of 307,000 cents, which is 3,070 dollars. It is ready to send.

Verdict: Allowed.

  • Financing summary. Monthly payment of 307,000 cents on 100,000 dollars financed over 36 months, total of payments 11,052,000 cents (modeled illustrative estimate before credit review and taxes).

3. Subscription financing: A subscription financing quote for a state and local customer with hardware in it

The agent's request. Please quote a subscription services financing program for a state and local customer. The structure is a fair market value lease over 36 months, with 30,000 dollars of hardware, 120,000 dollars of software and 50,000 dollars of services, credit Tier 3, and the non-appropriation clause not included. The quote states a monthly payment of 614,000 cents, which is 6,140 dollars. It is ready to send.

Verdict: Blocked.

  • Subscription financing has hardware. Subscription services financing covers software and services only, but the quote includes 30,000 dollars of hardware.
  • Soft costs too high for a fair market value lease. Software and services are more than 60 percent of the financed amount, above the modeled limit for a fair market value lease.
  • Non-appropriation clause missing. A federal or state and local customer needs the non-appropriation clause on a lease or usage-based structure, and the quote does not include it.
  • Stated payment does not match. The quote states a monthly payment of 614,000 cents; the modeled factor gives 654,000 cents.

4. Subscription financing: The agent resubmits after fixing the errors

The agent's request. Please quote a subscription services financing program for a state and local customer. The structure is a installment loan over 36 months, with 0 dollars of hardware, 120,000 dollars of software and 50,000 dollars of services, credit Tier 3, and the non-appropriation clause not included. The quote states a monthly payment of 566,100 cents, which is 5,661 dollars. It is ready to send.

Verdict: Allowed.

  • Financing summary. Monthly payment of 566,100 cents on 170,000 dollars financed over 36 months, total of payments 20,379,600 cents (modeled illustrative estimate before credit review and taxes).

5. PO financing: A PO financing quote for 20,000 dollars over 72 months

The agent's request. Please quote a PO Financing program for a commercial customer. The structure is a fair market value lease over 72 months, with 20,000 dollars of hardware, 0 dollars of software and 0 dollars of services, credit Tier 1, and the non-appropriation clause not included. It is ready to send.

Verdict: Blocked.

  • Term not offered. A term of 72 months is not offered; the modeled terms are 12, 24, 36, 48 and 60 months.
  • PO financing below its minimum. PO financing starts at a modeled 25,000 dollars, but the total financed amount is 20,000 dollars.

6. PO financing: The agent resubmits after fixing the errors

The agent's request. Please quote a PO Financing program for a commercial customer. The structure is a fair market value lease over 48 months, with 25,000 dollars of hardware, 0 dollars of software and 0 dollars of services, credit Tier 1, and the non-appropriation clause not included. The quote states a monthly payment of 59,500 cents, which is 595 dollars. It is ready to send.

Verdict: Allowed.

  • Financing summary. Monthly payment of 59,500 cents on 25,000 dollars financed over 48 months, total of payments 2,856,000 cents (modeled illustrative estimate before credit review and taxes).

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 financing quote for a customer buying IT hardware, software and services: a financing program, a lease or loan structure, a term, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the financing programs and returns a verdict, every reason and the monthly payment it computed.
  2. I will start with the first request by the agent.
  3. The agent says. Please quote a technology refresh program for a federal customer. The structure is a dollar out lease over 36 months, with 60,000 dollars of hardware, 25,000 dollars of software and 15,000 dollars of services. The rest of the request is on screen. The quote states a monthly payment of 307,000 cents, which is 3,070 dollars.
  4. Before it goes out, the agent calls the validator.
  5. Blocked, with four findings, all in one answer. A technology refresh program is quoted as a fair market value lease for 24, 36 or 48 months. Software and services are more than 35 percent of the financed amount, above the modeled limit for a dollar out lease. A federal or state and local customer needs the non-appropriation clause on a lease or usage-based structure, and the quote does not include it. The other one is on screen.
  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 quote are checked against each other.
  7. What this shows: for a stated payment that must match the computed one, a rules-based CPQ can do this only with workarounds.
  8. Now the agent resubmits.
  9. The corrected request. It changes the structure to fair market value lease, the hardware to 80,000 dollars, the software to 10,000 dollars, the services to 10,000 dollars and the non-appropriation clause to included.
  10. It calls the validator again.
  11. This time the validator clears it.
  12. All four earlier findings are cleared.
  13. Four 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.