Predictika · Conversational AI packages

An AI sales agent's Conversational Cloud package order, checked before it goes out

An AI sales agent assembles an order for a Bronze, Silver or Gold Conversational Cloud package: the capabilities the customer wants, the Data Transporter and training-phrase volumes, proactive messaging, and the support, customer success and handling-fee terms it promises. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the published LivePerson pricing page and returns a verdict, every reason and the listed terms it compared against.

Scope: LivePerson Conversational Cloud Bronze, Silver and Gold packages: which capabilities each includes, support and customer success level, listed limits, and Proactive Messaging channels and handling fee. See everything this validator covers

This validator models what the public LivePerson pricing page (www.liveperson.com/pricing, read October 2026) states for the three packages, not just the scenarios shown below.

  • Capabilities by package — eight capabilities are marked from Silver up (Conversation Builder, KnowledgeAI, Conversation Assist, Conversation Orchestrator, AI Annotator, 3rd Party Bot Connectors, LivePerson Analytics Studio, Proactive Messaging) and ten from Gold only (Generative Intent Training, Copilot Summary, Copilot Assist, Copilot Rewrite, the KnowledgeAI Agent and Routing AI Agent, Hallucination Detection, Prompt Library, Bring your own LLM, Generative Insights, which the page marks as early access).
  • Support and customer success — Standard technical support and Community customer success on Bronze; Enhanced support and a Pooled CSM on Silver; Enhanced support and a Designated CSM on Gold. The level the order states is compared with the package.
  • Listed limits — a maximum of 15 Data Transporter file-transfer tasks per site ID and 1,000 training phrases per intent in Generative Intent Training.
  • Proactive Messaging — listed on WhatsApp, SMS and InApp; messaging channels carry a handling fee of 15 percent, which is compared with the percentage the order states.

Any order for these three packages that touches these rules is in scope — not only the scenarios pictured here. Not covered: prices (LivePerson publishes no price list, so there is no price build-up or line items); the entitlements for Analytics Studio, Proactive Messaging and Generative AI tokens, which the page leaves to LivePerson sales; the Implementation row (the page shows it only as marks without a legend); capabilities included in all three packages; SMS gateway and phone-number costs and messaging rates, which the page says are not included. The pricing page can change without notice; the model reflects the page as read.

The AI sales agentBuilds the package 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.

7 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
A capability that exists only from a higher packageOnly with care
Eighteen capabilities each need a rule tying them to the package, and a package change has to re-check all of them, so the rules are easy to leave stale.
Every capability is checked against the package on every call.
Support, customer success and fee stated by the agentOnly with workarounds
The agent states support, customer success and a handling-fee percentage, and each must be compared with a value that depends on the package or channel; a mismatch has to stop the order.
The stated terms are fields in the order, and the validator blocks when one differs from the pricing page.
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.

1Conversation Builder is not included in the Bronze package; the pricing page lists it from Silver.
2KnowledgeAI is not included in the Bronze package; the pricing page lists it from Silver.
3Conversation Assist is not included in the Bronze package; the pricing page lists it from Silver.
4Conversation Orchestrator is not included in the Bronze package; the pricing page lists it from Silver.
5AI Annotator is not included in the Bronze package; the pricing page lists it from Silver.
63rd Party Bot Connectors is not included in the Bronze package; the pricing page lists it from Silver.
7LivePerson Analytics Studio is not included in the Bronze package; the pricing page lists it from Silver.
8Proactive Messaging is not included in the Bronze package; the pricing page lists it from Silver.
9Generative Intent Training is not included in the Bronze package; the pricing page lists it from Gold.
10Copilot Summary is not included in the Bronze package; the pricing page lists it from Gold.
11Copilot Assist is not included in the Bronze package; the pricing page lists it from Gold.
12Copilot Rewrite is not included in the Bronze package; the pricing page lists it from Gold.
13AI Agents: KnowledgeAI Agent is not included in the Bronze package; the pricing page lists it from Gold.
14AI Agents: Routing AI Agent is not included in the Bronze package; the pricing page lists it from Gold.
15Hallucination Detection is not included in the Bronze package; the pricing page lists it from Gold.
16Prompt Library is not included in the Bronze package; the pricing page lists it from Gold.
17Bring your own LLM is not included in the Bronze package; the pricing page lists it from Gold.
18Generative Insights is not included in the Bronze package; the pricing page lists it from Gold.
19Generative Intent Training is not included in the Silver package; the pricing page lists it from Gold.
20Copilot Summary is not included in the Silver package; the pricing page lists it from Gold.
21Copilot Assist is not included in the Silver package; the pricing page lists it from Gold.
22Copilot Rewrite is not included in the Silver package; the pricing page lists it from Gold.
23AI Agents: KnowledgeAI Agent is not included in the Silver package; the pricing page lists it from Gold.
24AI Agents: Routing AI Agent is not included in the Silver package; the pricing page lists it from Gold.
25Hallucination Detection is not included in the Silver package; the pricing page lists it from Gold.
26Prompt Library is not included in the Silver package; the pricing page lists it from Gold.
27Bring your own LLM is not included in the Silver package; the pricing page lists it from Gold.
28Generative Insights is not included in the Silver package; the pricing page lists it from Gold.
29The order asks for 16 Data Transporter file-transfer tasks; the pricing page lists a maximum of 15 tasks per site ID.
30The order asks for 1001 training phrases per intent in Generative Intent Training; the pricing page lists a maximum of 1,000 per intent.
31The order states a 12 percent handling fee on messaging channels; the pricing page lists a handling fee of 15 percent.
32The order states a 20 percent handling fee on messaging channels; the pricing page lists a handling fee of 15 percent.
33The pricing page lists Proactive Messaging on WhatsApp, SMS and InApp; Apple Messaging is not listed.
34The order states Enhanced technical support for the Bronze package; the pricing page lists Standard for Bronze and Enhanced for Silver and Gold.
35The order states Standard technical support for the Silver package; the pricing page lists Standard for Bronze and Enhanced for Silver and Gold.
36The order states Designated CSM customer success for the Silver package; the pricing page lists Community for Bronze, Pooled CSM for Silver and Designated CSM for Gold.
37The order states Pooled CSM customer success for the Gold package; the pricing page lists Community for Bronze, Pooled CSM for Silver and Designated CSM for Gold.
38The order states Pooled CSM customer success for the Bronze package; the pricing page lists Community for Bronze, Pooled CSM for Silver and Designated CSM for Gold.
39The order asks for 20 Data Transporter file-transfer tasks; the pricing page lists a maximum of 15 tasks per site ID.
40The order states Designated CSM customer success for the Bronze package; the pricing page lists Community for Bronze, Pooled CSM for Silver and Designated CSM for Gold.

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. Regional bank Bronze: A Bronze order for a regional bank

The agent's request. Please prepare a Bronze Conversational Cloud package order with Conversation Builder, KnowledgeAI and Copilot Summary. The order states Enhanced technical support and a Designated CSM. It is ready to send.

Verdict: Blocked.

  • Conversation Builder is not in this package. Conversation Builder is not included in the Bronze package; the pricing page lists it from Silver.
  • KnowledgeAI is not in this package. KnowledgeAI is not included in the Bronze package; the pricing page lists it from Silver.
  • Copilot Summary is not in this package. Copilot Summary is not included in the Bronze package; the pricing page lists it from Gold.
  • Stated technical support level differs from the package. The order states Enhanced technical support for the Bronze package; the pricing page lists Standard for Bronze and Enhanced for Silver and Gold.
  • Stated customer success level differs from the package. The order states Designated CSM customer success for the Bronze package; the pricing page lists Community for Bronze, Pooled CSM for Silver and Designated CSM for Gold.

2. Regional bank Bronze: The agent resubmits after fixing the errors

The agent's request. Please prepare a Gold Conversational Cloud package order with Conversation Builder, KnowledgeAI and Copilot Summary. The order states Enhanced technical support and a Designated CSM. It is ready to send.

Verdict: Allowed.

  • Listed terms for this package. Listed for the Gold package: Enhanced technical support and Designated CSM customer success.

3. Retailer Silver: A Silver order for a retailer

The agent's request. Please prepare a Silver Conversational Cloud package order with Conversation Orchestrator, AI Annotator, Hallucination Detection and Prompt Library, with 20 Data Transporter file-transfer tasks. The order states Enhanced technical support and a Pooled CSM. It is ready to send.

Verdict: Blocked.

  • Hallucination Detection is not in this package. Hallucination Detection is not included in the Silver package; the pricing page lists it from Gold.
  • Prompt Library is not in this package. Prompt Library is not included in the Silver package; the pricing page lists it from Gold.
  • Too many Data Transporter tasks. The order asks for 20 Data Transporter file-transfer tasks; the pricing page lists a maximum of 15 tasks per site ID.

4. Retailer Silver: The agent resubmits after fixing the errors

The agent's request. Please prepare a Gold Conversational Cloud package order with Conversation Orchestrator, AI Annotator, Hallucination Detection and Prompt Library, with 20 Data Transporter file-transfer tasks. The order states Enhanced technical support and a Designated CSM. It is ready to send.

Verdict: Blocked.

  • Too many Data Transporter tasks. The order asks for 20 Data Transporter file-transfer tasks; the pricing page lists a maximum of 15 tasks per site ID.

5. Retailer Silver: The agent resubmits after fixing the errors

The agent's request. Please prepare a Gold Conversational Cloud package order with Conversation Orchestrator, AI Annotator, Hallucination Detection and Prompt Library, with 15 Data Transporter file-transfer tasks. The order states Enhanced technical support and a Designated CSM. It is ready to send.

Verdict: Allowed.

  • Listed terms for this package. Listed for the Gold package: Enhanced technical support and Designated CSM customer success.

6. Clinic group proactive messaging: A Silver order with proactive messaging for a clinic group

The agent's request. Please prepare a Silver Conversational Cloud package order with Conversation Builder, KnowledgeAI and Proactive Messaging, sending proactive messages on SMS. The order states Enhanced technical support, a Pooled CSM and a handling fee of 12 percent on messaging channels. It is ready to send.

Verdict: Blocked.

  • Stated handling fee differs from the pricing page. The order states a 12 percent handling fee on messaging channels; the pricing page lists a handling fee of 15 percent.

7. Clinic group proactive messaging: The agent resubmits after fixing the errors

The agent's request. Please prepare a Silver Conversational Cloud package order with Conversation Builder, KnowledgeAI and Proactive Messaging, sending proactive messages on SMS. The order states Enhanced technical support, a Pooled CSM and a handling fee of 15 percent on messaging channels. It is ready to send.

Verdict: Allowed.

  • Listed terms for this package. Listed for the Silver package: Enhanced technical support and Pooled CSM customer success.

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 assembles an order for a Bronze, Silver or Gold Conversational Cloud package: the capabilities the customer wants, the Data Transporter and training-phrase volumes, proactive messaging, and more. Before it goes out, the agent sends it to Predictika, which checks it against a declarative model of the published LivePerson pricing page and returns a verdict, every reason and the listed terms it compared against.
  2. I will start with the first request by the agent.
  3. The agent says. Please prepare a Bronze Conversational Cloud package order with Conversation Builder, KnowledgeAI and Copilot Summary. The order states Enhanced technical support and a Designated CSM. The rest of the request is on screen.
  4. Before it goes out, the agent calls the validator.
  5. Blocked, with five findings, all in one answer. Conversation Builder is not included in the Bronze package; the pricing page lists it from Silver. KnowledgeAI is not included in the Bronze package; the pricing page lists it from Silver. Copilot Summary is not included in the Bronze package; the pricing page lists it from Gold. The other two are on screen.
  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 order are checked against each other.
  7. What this shows: for support, customer success and fee stated by the agent, a rules-based CPQ can do this only with workarounds.
  8. Now the agent resubmits.
  9. The corrected request. It changes the package to gold.
  10. It calls the validator again.
  11. This time the validator clears it.
  12. All five earlier findings are cleared.
  13. Five 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.