Meet
Agent Dhara,
the AI analyst built for incentive operations.
Dhara helps teams ask business questions, investigate disputes, trace plan changes, understand performance, and move from insight to action across the Incentivate platform.
Not a chatbot placed on top of incentive data. Dhara is built on the business context, semantic understanding, plan history, payout logic, roster data, territory alignment, transactions, documents, and workflows that compensation teams actually work with every day.
Ask business questions in plain English.
Incentive platforms sit at the intersection of sales, products, territories, quotas, crediting, finance, disputes, rosters, and performance. Many of the questions compensation teams receive are not narrowly compensation questions — they are business questions.
Which is the top state that sells home insurance policies?
Identifies the product category, queries policy transaction data, groups performance by state, ranks the results, and summarizes the answer in business language.
Dhara does not stop at the answer. It suggests the next useful question — whether the same pattern holds for auto policies, producers, branches, or time periods.
Built for the messy, investigative work behind incentive operations.
Incentive questions rarely live in one table. To resolve a dispute or explain a payout, an analyst may need to check transactions, credits, plan rules, rate tables, roster history, territory alignment, account assignments, eligibility rules, manual adjustments, prior disputes, payout outputs, and plan documents. That work is slow because the analyst is not simply looking for data — they are reconstructing a story.
- 1 Open payout report
- 2 Check transaction table
- 3 Check roster
- 4 Check territory alignment
- 5 Check plan eligibility
- 6 Check rate table
- 7 Check prior disputes
- 8 Write explanation manually
- 1 Ask the question
- 2 Dhara checks relevant sources
- 3 Dhara explains the likely cause
- 4 Dhara shows evidence
- 5 Analyst verifies and acts
Dhara does not remove the analyst from the process. It gives the analyst a faster path to evidence, explanation, and action.
Guidance when something needs to change.
Many incentive questions end with an operational action. If a commission rate needs to be changed, a parameter updated, or an eligibility condition corrected, the analyst still needs to know where to go, what to update, and what to check afterward.
- The relevant plan or component
- The parameter or rate table where the value is maintained
- The effective date that should be considered
- Any downstream calculations or reports that may be affected
- Checks to run before closing the change
Trace changes across plan documents and versions.
Plan history is often hard to analyze. Rates change, rules change, product categories change, and sometimes the only way to answer a question is to compare old documents manually. Dhara retrieves historical plan documents, compares the relevant language, identifies changes, and summarizes the timeline.
| Plan version | Effective period | Auto commission rate | Change detected |
|---|---|---|---|
| Producer Plan, January version | Jan – Jun | 3.5% | Baseline |
| Producer Plan, July version | Jul – Dec | 3.5% | No rate change |
| Producer Plan, New Year version | Jan onward | 7.0% | Δ +3.5% |
// sources: producer_plan.v{jan,jul,new_year}.pdf · 3 versions analyzed
Designed for verification, not blind trust.
In incentive management, a fast answer is not enough. Payouts affect compensation, finance, trust, sales motivation, and dispute resolution. Analysts need to understand where an answer came from before relying on it.
The auto commission rate appears to have changed from 3.5% to 7.0%, effective starting in January, based on the most recent Producer Plan document.
The previous two plan versions (January and July) both carried 3.5%. The New Year version introduces the increase.
- doc Producer Plan (July version) — rate table, section 4.2
- doc Producer Plan (January version) — rate table, section 4.2
- doc Auto policy commission language — page 12
- 01 Located the relevant Producer Plan versions across history
- 02 Found the lines where auto plan commission rates are mentioned
- 03 Extracted the rate values and effective dates from each version
- 04 Compared rates across versions to find the change
- 05 Summarized the rate movement and the document it appeared in
AI needs business meaning, not just database access.
Enterprise source-system data is often cryptic. A column may be named
ZCREV13
but the business meaning may be "recognized revenue for eligible home insurance policies." A
generic AI agent cannot reliably infer that. Dhara is supported by a semantic
layer that describes what data means, how fields should be interpreted, how
tables relate, and which business concepts matter.
- ZCREV13
- PRD_TYP_CD
- ALGN_FLG
- CR_ADJ_AMT
- Eligible revenue
- Product type
- Alignment status
- Credit adjustment amount
An agent your team can keep teaching.
Every incentive environment is different. Each company has its own plan language, exception patterns, approval processes, and operational habits. Dhara is built as a skills-based system — teams add or update capabilities by writing Markdown skills that define how Dhara should handle specific questions or tasks.
---
name: dispute-investigation
description: Investigate a filed dispute end-to-end
---
## Steps
1. Retrieve the dispute record and prior history
2. Pull related transactions and roster state
3. Check alignment and eligibility for the period
4. Compare against the applicable plan version
5. Summarize likely cause with linked evidenceDhara becomes more valuable as more institutional knowledge is captured into reusable skills.
When business context changes, Dhara helps analysts investigate faster.
Some of the hardest analyst requests do not arrive as clean tickets. They start as business questions from leadership — a market disruption, a regional event, an unusual sales decline. Dhara helps structure the investigation, surface comparisons, and prepare a defensible recommendation.
- Step 01Identify affected territories
- Step 02Review performance before & after
- Step 03Compare with unaffected peers
- Step 04Check product-level impact
- Step 05Temporary or recurring?
- Step 06Prepare summary for review
Where external research or live news context is required, Dhara helps structure and support the investigation alongside the analyst — it does not replace external context.
Enterprise AI with governance built in.
Incentive management is a high-trust function. AI here must respect permissions, preserve auditability, and support human review. Dhara is designed to support governed operations — helping users investigate, explain, and act with better context while the right humans stay in control of decisions and changes.
What teams can ask Agent Dhara.
Dhara meets each role where they work — from the analyst chasing down a dispute, to the leader looking at this quarter's payout movement.
See what AI can do when it understands incentives.
Agent Dhara brings business-aware AI into the daily work of incentive management — answering questions, tracing changes, investigating disputes, guiding analysts, and surfacing insights across sales performance and payout operations.
