Stark Informatics
Home · Accelerators · Finance Data Agent
AI accelerator

Finance Data Agent

A pre-configured Fabric Data Agent sitting on a Finance semantic model — with the curated instructions, example queries, and evaluation harness that make it answer questions the CFO would actually accept.

StackData Agent + Semantic Model
LanguageDAX + Markdown + JSON
Deploy time3–5 days
Tested questions50-question eval set
LicenseCommercial

The hard part of a Data Agent isn't standing one up — it's getting it to answer correctly. The difference between a demo-ready agent and a production-trusted one is 50 hours of curated instructions, example pairs, glossary entries, and an evaluation harness.

This accelerator gives you all of that, scoped to a Finance domain. Drop in your own semantic model, swap the glossary entries to your terminology, and you have a CFO-grade agent in a week instead of a quarter.

Who this is for

  • Finance + data teams who want to give controllers and FP&A analysts conversational access to actuals, forecasts, and variances — without writing reports for every question.
  • Consultancies who want to drop a working Data Agent into client engagements and customize from there.
  • Teams already on Fabric who experimented with a Data Agent, got mediocre answers, and want to see what a tuned one looks like.

What's in the box

FileTypeWhat it does
semantic-model/finance.bimTMDLFinance star schema: GL accounts, cost centers, entities, scenarios, time intelligence. ~80 documented measures.
semantic-model/glossary.mdMDBusiness glossary: revenue, OpEx, EBITDA, accruals, FTE — defined in your auditor's language.
agent/instructions.mdMD15,000-character system prompt: routing rules, terminology, refusal guidance, formatting standards.
agent/examples.jsonJSON25 curated question→DAX pairs covering YoY, variance, drill-down, period close, ratios.
eval/eval_set.jsonJSON50 questions with expected answers + acceptance criteria. Re-run any time you change instructions.
eval/run_eval.ipynbPythonNotebook that runs the eval set against your agent and produces a pass-rate scorecard.
integration/teams_app.jsonJSONTeams app manifest: pin the agent in any channel.
integration/copilot_studio.jsonJSONCopilot Studio custom skill definition.
rls/dynamic_rls.daxDAXEntity + cost-center RLS using USERPRINCIPALNAME() + bridge table — tested across 5 user personas.
data/seed_data.csvCSVSynthetic finance data so you can run the agent end-to-end before plugging in your real data.
docs/CFO_GRADE_CHECKLIST.mdMDThe 14-point checklist we use to certify an agent for executive consumption.

The CFO-grade checklist

An agent is "demo-grade" when it answers most questions. It's "CFO-grade" when:

  1. It refuses confidently when the data can't support a precise answer.
  2. It always cites the underlying measure and time period.
  3. It applies the right time-intelligence pattern (calendar vs fiscal, period-end vs running).
  4. It honors RLS — a regional VP doesn't see other regions.
  5. It uses your terminology, not Microsoft's defaults ("OpEx" not "operating expenses").
  6. It handles "show me" in Teams the same as "what's" — both work.
  7. It produces numbers that tie to the close, not approximations.

Every one of these is hard-won, embedded in the included instructions and example set.

Pricing

Individual
$699
1 developer · personal projects
Buy individual
Site
$7,999
Unlimited developers · lifetime updates
Buy site

Frequently asked questions

What about Azure OpenAI token costs?
Those are billed separately to your tenant — typically $0.01–$0.10 per question depending on context length. With a tuned agent on a curated model, most production agents cost ~$50–$500/month in tokens.
My finance data doesn't look like a star schema — what then?
The semantic model template is generic enough that most teams can map their warehouse tables in 1–2 days. If you're starting from a Snowflake/Synapse extract, our Medallion Starter pairs naturally.
How accurate is the agent really?
On the 50-question eval set against our reference data, it scores 92% exact-match and 96% directionally-correct. Your numbers will depend on your data and how well you customize the glossary.
Does it work with regional finance terminology?
The glossary is editable text. Swap "EBITDA" for "EBIT", "fiscal year" for "FY", whatever your team uses. The agent picks it up.
Can I deploy multiple agents (one per region or function)?
Yes. The Team license covers unlimited agent instances within your org.
Refund policy?
Full refund within 14 days.

The CFO asks "what was OpEx variance last quarter?" — and gets the answer in Teams

That's the promise. The work between zero and that moment is what this accelerator covers.

Buy team license — $2,499