Predictika · luminaires, drivers, controls, outdoor ratings, utility rebates and discount authority

An AI sales agent's lighting project quote, checked before it goes out

An AI sales agent prepares a lighting project quote for a fictional customer: a luminaire family, a driver, a control system, an indoor or outdoor location, a region, an optional utility rebate, a fixture count and a discount. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the compatibility, rebate and approval rules and returns a verdict, every reason, and the price it computed.

Scope: Lighting project quotes in the style of Signify: luminaire, driver and control compatibility, outdoor ratings, utility rebates and discount authority. An illustrative model; prices and limits are fictional. See everything this validator covers

This validator models lighting project quotes in the style of Signify, with four luminaire families, four drivers and four control levels — not just the scenarios shown below.

  • Driver and control compatibility — The dimming level of the driver and of the luminaire against what the chosen control needs, and the luminaire power against the driver limit.
  • Location — An outdoor project needs a luminaire rated IP65 or higher.
  • Utility rebates — The luminaire must be eligible, the control must be networked, and the project must be in North America.
  • Quantity, discount and size — One to 5000 fixtures per quote, a volume discount from 200 fixtures, an approval band above 15 percent total discount and a limit at 25 percent, a lead time check from 1000 fixtures, and a deal desk review above USD 250,000.

Any project quote for these luminaires, drivers, controls and regions is in scope — not only the scenarios pictured here. It does not cover other Signify product lines, lighting design or photometric calculations, or real list prices.

The AI sales agentBuilds the lighting project 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 5 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 a rules-based CPQ gets this wrong

Click to expand

Our assessment of what a typical rules-based CPQ (product rules, price rules and validation rules, as in Salesforce CPQ or Conga) 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
Driver and control compatibilityOnly with workarounds
A product rule per driver, control and luminaire combination; each new driver multiplies the rules.
Each requirement is a catalog attribute compared with the quote.
Power limit of the driverOnly with care
A validation rule per driver family comparing wattage; easy to forget when a new luminaire is added.
The luminaire power is compared with the driver limit on every call.
Outdoor rating by locationOnly with care
A rule tied to a location field and a luminaire attribute; usually added late.
The location and the IP rating are compared directly.
Utility rebate with three conditionsOnly with workarounds
Eligibility lives in a spreadsheet or separate system, so the quote is built before the answer is known.
Luminaire, control level and region are three separate findings returned together.
Volume discount stacked with an extra discountOnly with workarounds
Stacked schedules and approval limits; double counting is common.
The total discount is one calculation; the approval band and the limit are separate findings.
Every problem returned in one answerOnly with workarounds
Rules tend to fire as the quote is saved, 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.

1Total discount of 17 percent is above the 15 percent a rep can give; sales director must approve.
2Total discount of 16 percent is above the 15 percent a rep can give; sales director must approve.
31200 fixtures is at or above 1000; supply chain must confirm the lead time.
4Project value of USD 349695.0 is above USD 250000 and needs a deal desk review.
5The Fixed output driver supports dimming level 0 but the DALI networked control needs level 2.
6The Recessed LED downlight is rated IP44; outdoor use needs IP65 or higher.
7The Warehouse LED high bay supports dimming level 2 but the Connected sensor node needs level 3.
8The 150 watt Warehouse LED high bay exceeds the 100 watt limit of the D4i sensor ready driver.
9The Recessed LED downlight is not eligible for a utility rebate.
10A utility rebate needs networked controls of level 2 or higher; the 0-10V wall dimmer is level 1.
11Utility rebates are only offered for projects in North America; the project region is EU.
126000 fixtures are outside the 1 to 5000 fixtures of one quote.
136000 fixtures is at or above 1000; supply chain must confirm the lead time.
14Project value of USD 783840.0 is above USD 250000 and needs a deal desk review.
15Total discount of 27 percent exceeds the 25 percent limit.
16The Fixed output driver supports dimming level 0 but the Connected sensor node needs level 3.
17The 0-10V dimming driver supports dimming level 1 but the Connected sensor node needs level 3.
18The quoted net total of USD 37637.1 does not match the USD 40470.0 the validator computed.

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. Driver and control compatibility: Fixed output driver with DALI control

The agent's request. Recessed LED panels with fixed output drivers and DALI networked control, 300 fixtures. Ready to send.

Verdict: Blocked.

  • Driver Dimming. the Fixed output driver supports dimming level 0 but the DALI networked control needs level 2.

2. Driver and control compatibility: The agent resubmits after fixing the errors

The agent's request. Recessed LED panels with DALI drivers and DALI networked control, 300 fixtures for an indoor project in North America. Ready to send.

Verdict: Allowed.

  • Price estimate. PRICE (estimated list prices in USD): USD 40470.0 total for 300 fixtures; USD 142.0 per fixture with driver and control; USD 42600.0 before discount; 5 percent total discount; USD 0.0 estimated utility rebate.

3. Sensor control and driver limits: High bay with a sensor ready driver and sensor node

The agent's request. Warehouse high bays with D4i drivers and connected sensor nodes, 50 fixtures. Ready to send.

Verdict: Blocked.

  • Fixture Dimming. the Warehouse LED high bay supports dimming level 2 but the Connected sensor node needs level 3.
  • Driver Power. the 150 watt Warehouse LED high bay exceeds the 100 watt limit of the D4i sensor ready driver.

4. Driver, sensor node and rebate region: Fixed output driver with sensor node and a rebate in Europe

The agent's request. Outdoor floodlights with fixed output drivers, connected sensor nodes and the utility rebate for a European site. Ready to send.

Verdict: Blocked.

  • Driver Dimming. the Fixed output driver supports dimming level 0 but the Connected sensor node needs level 3.
  • Rebate Region. utility rebates are only offered for projects in North America; the project region is EU.

5. Discount approval: 12 percent extra discount on 300 fixtures

The agent's request. The same panel project with another 12 percent off to win the account. Ready to send.

Verdict: Needs approval.

  • Discount Approval. total discount of 17 percent is above the 15 percent a rep can give; sales director must approve.
  • Price estimate. PRICE (estimated list prices in USD): USD 35358.0 total for 300 fixtures; USD 142.0 per fixture with driver and control; USD 42600.0 before discount; 17 percent total discount; USD 0.0 estimated utility rebate.

6. Utility rebate: High bays with DALI control and a utility rebate, 150 fixtures

The agent's request. Warehouse high bays with DALI control for a North American project, with the utility rebate applied. Ready to send.

Verdict: Allowed.

  • Price estimate. PRICE (estimated list prices in USD): USD 41850.0 total for 150 fixtures; USD 279.0 per fixture with driver and control; USD 41850.0 before discount; 0 percent total discount; USD 3750.0 estimated utility rebate.

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 lighting project quote for a fictional customer: a luminaire family, a driver, a control system, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the compatibility, rebate and approval rules and returns a verdict, every reason, and the price it computed.
  2. I will start with the first request by the agent.
  3. The agent says. Recessed LED panels with fixed output drivers and DALI networked control, 300 fixtures. Ready to send.
  4. Before it goes out, the agent calls the validator.
  5. Blocked, with one finding, all in one answer. the fixed output driver supports dimming level 0 but the DALI networked control needs level 2.
  6. The validator returned one blocking finding 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 driver and control compatibility, a rules-based CPQ can do this only with workarounds.
  8. Now the agent resubmits.
  9. The corrected request. It changes the driver to DALI dimming driver.
  10. It calls the validator again.
  11. This time the validator clears it.
  12. The earlier finding is cleared.
  13. One problem 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.