Bring your brand into every Microsoft Fabric pipeline. Enrich any data object with your mission, vision, values, personas and competitors, then rank records by OnBrand fit and persona match, all from your notebooks, dataflows and pipelines.
Attach the full brand object to any row or table
Match data to personas and target audiences
Score and rank records by brand fit
Compliance checks inside your pipeline
Call BrandHalo from anywhere in Fabric to add brand intelligence to your own data, analysis and queries.
Pull mission, vision, values, tagline, positioning, messaging and competitors, and join them to any table in your lakehouse.
Work out whether a customer or record matches a specific persona or target audience, then tag and segment on that fit.
Generate OnBrand scores, persona-fit scores and competitor-overlap flags to rank and prioritise records as you process them.
Score any text against your brand guidelines and return a compliance score with specific recommendations.
Search products, services and locations from your Brand Data graph and follow their relationships to personas and content.
Combine BrandHalo signals with your own Fabric data, models and semantic layers to power downstream analytics.
BrandHalo works with the Fabric tools you already use. Store a BrandHalo API key as a secret, call our MCP server or REST connector, and write the enriched columns back to your lakehouse.
Call the MCP server from PySpark or Python for row-level enrichment, persona matching and OnBrand scoring.
Use the Web API connector in Power Query to pull brand data and compliance scores with a simple bearer header.
Use a Web activity to fetch brand dimensions once per run and land them as tables for joins and analysis.
import requests, json
BH_URL = "https://www.brandhalo.io/mcp"
BH_KEY = notebookutils.credentials.getSecret(
"https://<vault>.vault.azure.net/", "brandhalo-api-key")
HEADERS = {"Authorization": f"Bearer {BH_KEY}", "Content-Type": "application/json"}
def mcp_call(tool, arguments=None):
body = {"jsonrpc": "2.0", "id": 1, "method": "tools/call",
"params": {"name": tool, "arguments": arguments or {}}}
r = requests.post(BH_URL, headers=HEADERS, json=body, timeout=30)
r.raise_for_status()
return json.loads(r.json()["result"]["content"][0]["text"])
profile = mcp_call("get_brand_profile") # mission, vision, values, tagline
personas = mcp_call("get_brand_personas") # target personas
score = mcp_call("check_brand_compliance",
{"textContent": row_text[:10000]}) # OnBrand ranking signal
spark.createDataFrame([profile]).write.mode("overwrite").saveAsTable("dim_brand_profile")let
ApiKey = "bh_your_api_key",
Url = "https://www.brandhalo.io/api/copilot/connector?action=get_brand_profile",
Source = Json.Document(
Web.Contents(Url, [ Headers = [ #"Authorization" = "Bearer " & ApiKey ] ])
)
in
Source[result]Log in to BrandHalo, create an API key on the Connectors page, then follow the step-by-step Fabric guide.
Generate an API key in your BrandHalo dashboard, then use the Microsoft Fabric guide for the full walkthrough, including notebook, Dataflow Gen2 and pipeline examples for enriching and ranking your data.
The Fabric integration uses the same secure, read-only connector that powers our enterprise AI integrations.
Store a BrandHalo API key as a Fabric or Azure Key Vault secret.
Pipelines can read brand data and request scores, never modify your brand.
Connector calls are logged so you can track integration activity.
Disable a key anytime from the BrandHalo dashboard.
Our team can help you design your Fabric enrichment and ranking pipelines. Contact us for onboarding support.