Predictika · GE HealthCare-style CT system quotes

An AI sales agent's CT scanner quote, checked before it goes out

An AI sales agent prepares a quote for a Revolution CT system, in US dollars. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the clinical-fit, site, schedule, discount and pricing rules, compares the total the quote states with the computed total, and returns every finding together with the line items. This is an illustrative model built from public product information with fictional prices; it is not GE HealthCare's internal rule set.

Scope: GE HealthCare-style Revolution CT system quotes (five scanners, three software options, service plans, trade-in and renovation), checked against clinical-fit, site, schedule, discount and pricing rules. See everything this validator covers

This validator checks a CT scanner system quote of the kind GE HealthCare sells: five Revolution scanners, the cardiac software suite, spectral imaging and deep-learning reconstruction, a service plan, a trade-in credit and a room renovation. It handles any quote built from these items, not only the scenarios shown on this page. It does not cover other imaging modalities, financing, installation drawings or other manufacturers' scanners.

  • Covered: clinical fit — coverage and rotation for a cardiac program, coverage for whole-brain stroke perfusion, bore for radiation-therapy simulation, and the pediatric dose sign-off when deep-learning reconstruction is missing.
  • Covered: site — patient table weight limit, minimum room size, site power, the physicist shielding survey, and the trauma-room uptime guarantee.
  • Covered: software options — which options each scanner offers, and the cardiac suite required for a cardiac program.
  • Covered: schedule — delivery and install weeks plus any room renovation against the date the first patient is needed.
  • Covered: price — list price, options, discount against rep authority and the customer-type ceiling, trade-in credit and multi-year service, as line items that add up to the total; the total the agent states is compared with the computed one.
  • Not covered — other scanner families, real GE HealthCare prices, financing and the actual order submission. Other products can be added.
  • Labeled as assumptions — every price, limit, lead time and discount tier is an illustrative estimate from public product information, not GE HealthCare data.
The AI sales agentBuilds the CT system 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.

8 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
Scanner capability against the clinical programOnly with workarounds
Whether a scanner can run a cardiac, stroke-perfusion or radiation-therapy program depends on its coverage, rotation time and bore, and the patient table and room on its size and weight limits. A picklist cannot hold a decision table that spans five scanners and six programs; it ends up in custom code.
Every capability limit is evaluated on every call, and each finding names the limit and the scanner value it was compared with.
Software options that exist only on some scannersOnly with care
The cardiac suite and spectral imaging are offered on some scanners only, and a cardiac program needs the suite on the quote. Option-to-model compatibility can be written as rules, but it multiplies with every scanner and package.
Options are checked against the scanner and the clinical program together, so the same option passes on one scanner and is blocked on another.
Discount ceilings by customer type with an approval bandOnly with workarounds
A rep may give up to 10 percent, a manager can approve up to the ceiling for the customer type, and above that the quote is blocked. Three outcomes from one number and one customer type is more than an approval flag.
The discount is read against both limits, so 12 percent needs approval, 22 percent is blocked for a community hospital, and 25 percent is approvable for a health network.
Lead time: delivery, install and renovation against the first-patient dateOnly with care
The weeks add up from the scanner, the room renovation and the install, and the quote fails only if the sum passes the date. This is a calculation across lines that a price book handles only with scripting.
Lead time is computed on every call and compared with the date the customer needs.
Stated total that must match the computed oneOnly with workarounds
A validation rule can compare a stated figure with a calculated one only if the stated figure is a field on the quote, which it rarely is.
The total the agent states travels with the quote and is compared with the computed total on every call; a difference blocks the quote.

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.

118 percent discount is above the 10 percent rep authority; approvable up to 25 percent.
226 percent discount exceeds the 25 percent ceiling for this customer type.
3A cardiac program requires the cardiac software suite (CardIQ / SnapShot) on the quote.
4A pediatric program without TrueFidelity deep-learning reconstruction needs clinical physics sign-off on dose.
5A trauma room on Standard service has no uptime guarantee; service leadership must approve or move to Premium.
6Delivery plus install takes 18 weeks but first patient is needed in 12 weeks.
7Gemstone Spectral Imaging is not available on Revolution EVO.
8Patients up to 280 kg exceed the 227 kg table limit of Revolution Ascend.
9Radiation-therapy simulation needs an 80 cm bore; Revolution Maxima has 70 cm.
10Revolution EVO has 4 cm coverage and 350 ms rotation; a cardiac program needs 8 cm or more or 280 ms or faster.
11Room of 340 sq ft is below the 360 sq ft minimum for Revolution Ascend.
12Site power 125 kVA is below the 150 kVA Revolution Apex requires.
13Stated total does not match: the agent stated $744,000.00; the validator computes $669,000.00.
14The cardiac software suite is not offered on Revolution EVO.
15The physicist shielding survey is not complete; site readiness must sign off.
16Whole-brain stroke perfusion needs 8 cm or more coverage; Revolution Ascend has 4 cm.

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. Clinical fit: Cardiac program on a 4 cm scanner

The agent's request. Quote for a health network: one Revolution EVO for a cardiac program, with the cardiac software suite and deep-learning reconstruction, 8 percent discount, Premium service for 3 years. Total $1,525,000. It is ready to send.

Verdict: Blocked.

  • Scanner too slow or narrow for a cardiac program. Revolution EVO has 4 cm coverage and 350 ms rotation; a cardiac program needs 8 cm or more or 280 ms or faster.
  • Cardiac suite not offered on this scanner. The cardiac software suite is not offered on Revolution EVO.

2. Clinical fit: Revolution CT for the same program

The agent's request. Quote for a health network: one Revolution CT for a cardiac program, with the cardiac software suite and deep-learning reconstruction, 8 percent discount, Premium service for 3 years. Total $2,712,500. It is ready to send.

Verdict: Allowed.

  • Quote summary. Quote total $2,712,500: net equipment $1,996,400 (list $2,170,000, options $270,000, after discount and trade-in), plus service $238,700 per year for 3 years; delivery and install take 18 weeks.

3. Site and schedule: Site power and the first-patient date

The agent's request. Quote for a health network: one Revolution Apex with a room renovation, a trade-in of the existing scanner and Standard service for 3 years, on a 125 kVA service. First patient is needed in 24 weeks. Total $2,901,000. It is ready to send.

Verdict: Blocked.

  • Site power below the scanner requirement. Site power 125 kVA is below the 150 kVA Revolution Apex requires.
  • Delivery later than the first-patient date. Delivery plus install takes 28 weeks but first patient is needed in 24 weeks.

4. Site and schedule: Power raised, no renovation

The agent's request. Quote for a health network: one Revolution Apex with a trade-in of the existing scanner and Standard service for 3 years; the site power was raised to 160 kVA so no renovation is needed. First patient is needed in 24 weeks. Total $2,901,000. It is ready to send.

Verdict: Allowed.

  • Quote summary. Quote total $2,901,000: net equipment $2,400,000 (list $2,400,000, options $0, after discount and trade-in), plus service $192,000 per year for 3 years; delivery and install take 20 weeks.

5. Approvals: Pediatric program, 12 percent discount

The agent's request. Quote for an academic medical center: one Revolution CT for a pediatric program with Premium service for 5 years and a 12 percent discount. Total $2,717,000. It is ready to send.

Verdict: Needs approval.

  • Pediatric dose sign-off needed. A pediatric program without TrueFidelity deep-learning reconstruction needs clinical physics sign-off on dose.
  • Discount above rep authority. 12 percent discount is above the 10 percent rep authority; approvable up to 20 percent.
  • Quote summary. Quote total $2,717,000: net equipment $1,672,000 (list $1,900,000, options $0, after discount and trade-in), plus service $209,000 per year for 5 years; delivery and install take 18 weeks.

6. Approvals: Reconstruction added, 10 percent discount

The agent's request. Quote for an academic medical center: one Revolution CT for a pediatric program with deep-learning reconstruction, Premium service for 5 years and a 10 percent discount. Total $2,885,500. It is ready to send.

Verdict: Allowed.

  • Quote summary. Quote total $2,885,500: net equipment $1,791,000 (list $1,990,000, options $90,000, after discount and trade-in), plus service $218,900 per year for 5 years; delivery and install take 18 weeks.

7. Stated total: Stated total is off

The agent's request. Quote for a community hospital: one Revolution Maxima with a trade-in of the existing scanner and Standard service for 3 years. Total $744,000. It is ready to send.

Verdict: Blocked.

  • Stated total does not match. Stated total does not match: the agent stated $744,000.00; the validator computes $669,000.00.

8. Stated total: Total matches the validator

The agent's request. Quote for a community hospital: one Revolution Maxima with a trade-in of the existing scanner and Standard service for 3 years. Total $669,000. It is ready to send.

Verdict: Allowed.

  • Quote summary. Quote total $669,000: net equipment $600,000 (list $600,000, options $0, after discount and trade-in), plus service $48,000 per year for 3 years; delivery and install take 10 weeks.

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 a Revolution CT system, in US dollars. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the clinical-fit, site, schedule, discount and pricing rules, compares the total the quote states with the computed total, and returns every finding together with the line items. This is an illustrative model built from public product information with fictional prices; it is not GE HealthCare's internal rule set.
  2. I will start with the first request by the agent.
  3. The agent says. Quote for a health network: one Revolution EVO for a cardiac program, with the cardiac software suite and deep-learning reconstruction, 8 percent discount, Premium service for 3 years. Total $1,525,000. The rest of the request is on screen.
  4. Before it goes out, the agent calls the validator.
  5. Blocked, with two findings, all in one answer. Revolution EVO has 4 cm coverage and 350 ms rotation; a cardiac program needs 8 cm or more or 280 ms or faster. The cardiac software suite is not offered on Revolution EVO.
  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. What this shows: for scanner capability against the clinical program, a rules-based CPQ can do this only with workarounds.
  8. Now the agent resubmits.
  9. The corrected request. It changes the scanner to Revolution CT and the stated total to $2,712,500.
  10. It calls the validator again.
  11. This time the validator clears it.
  12. The validator also returns the line-item quote. It starts with Revolution CT and Cardiac software suite (CardIQ / SnapShot); the other three lines are on screen.
  13. Both earlier findings are cleared, and the validator's total, $2,712,500, matches the quote.
  14. Two problems caught in one call, fixed in one pass, and a clean $2,712,500 quote. That is the value of Predictika's validator: catching mistakes before they reach the customer. Now you can try the other scenarios yourself.