Dispatches
Customer story

Semantic model generation

A data lead typed three sentences; Athena inspected the Snowflake catalog and drafted a semantic model—the shared definitions and calculations that ensure every dashboard and agent sees the same numbers.

ThousandsTables in a the raw warehouse catalogNo agreed definition of a net saleThe old wayNowExplore catalogMap tableDraft dimension schemaDefine every measureDocument joins & logicWeeks to build a semantic layerType three sentencesReview draft modelReady in minutesBefore: manual schema archaeologySESSIONWe have stores, products, and regions.Each sale links a store, a part, and aregion.Inspecting catalog… found STORES,PARTS, VEHICLES, SALES. Draftingsemantic model.Snowflake catalogThousands of tablesData lead via Athena · 9:14A data lead types three sentencessemantic_model_draft.xlsxElementSourceDefinitionReasoningStatusDimension: StoreSTORESstore_id, namePrimary keyDraftDimension: PartPARTSpart_idPrimary keyDraftDimensionVEHICLESvehicle_idPrimary keyDraftMeasureSALESamountInferredDraftJoin: SALESSALES.store_idForeign keyPlease confirmNeeds reviewJoin: SALESSALES.part_idForeign keyPlease confirmNeeds reviewWHO CHANGED WHATAthenaInspected catalog lineageAthenaInferred tablerelationshipsAthenaDrafted measures &dimensionsAthenaFlagged missing foreignkeysAthena drafts dimensions, measures, joinsAthena wants toPublish semantic model with two join keysflagged for reviewSALES → STORES join via store_id (no FK constraint)SALES → PARTS join via part_id (no FK constraint)Net Sale = amount - discount - returnsWAITING FOR A PERSONData lead confirms keys; model publishedLegacy tables without foreign keys need confirmationHuman confirms join keys for legacy tablesNet Sale measure fromsemantic modelDashboard tileUses model definitionAgent queryUses model definition=ATHENAQUERY() cellUses model definitionWarehouse permissionsEnforced for every query×Agents and humanssee the samenumbersEvery query goes through the modelCatalog in. Measures out.athenaintel.com
Fig. 02The story in 23 seconds, from the old way to “Catalog in. Measures out.”
Before · what we did · after
  1. Before

    A raw warehouse catalog with thousands of tables and no agreed definition of a net sale.

  2. What we did

    A data lead typed three sentences about their business. Athena inspected the Snowflake catalog, inferred table relationships and lineage from column names and structure, and drafted dimensions, measures, and calculations.

  3. After

    A draft semantic model with dimensions, measures, and join logic, ready to review in minutes. Every dashboard, agent, and =ATHENAQUERY cell now goes through it with the user's own warehouse permissions. Agents and humans see the same numbers.

STORESPARTSVEHICLESSALESONE DEFINITIONNet SaleSALES.amount -SALES.discount - SALESDashboard measureDefinedAgent queryConsistentsame number, both places
Fig. 01Athena inferred lineage and drafted a net sale measure from the catalog
How long it takes · what to plan for

Legacy tables without foreign keys needed a human to confirm join keys. The draft showed its reasoning for each.

Catalog in. Measures out.

Example workflow

Catalog to semantic model in the time it takes to read it

SURFACES INWHAT COMES INTHE PLATFORMWHAT GOES OUTDELIVERED BACKWebMobile · voiceText (SMS)EmailSlack · TeamsChromeOffice add-inMeeting botMeetingsWordPowerPointFiles · PDFsDataAgentsWorkflows (AOPs)AppsLibrarySessionsSpacesStudiosMeetingsLakehouseDatabaseDashboardsWordPowerPointSheetsData · APIsApps · portalsWebMobile · voiceText (SMS)EmailSlack · TeamsChromeOffice add-inMeeting botSheets1Semantic models2Sheets3Dashboards41 INData lead typed threesentences about stores,products, and regions in asheet2 BUILDAthena inspected theSnowflake catalog, inferredtables and lineage, drafteddimensions, measures, andcalculations3 REVIEWHuman confirmed join keys forlegacy tables without foreignkeys; Athena showed itsreasoning for each4 ANSWEREvery dashboard, agent, and=ATHENAQUERY() cell now goesthrough the semantic modelwith user warehousepermissions
Fig. 03The route on the system map: Sheets → Semantic models → Dashboards. Connectors: Snowflake catalog → Semantic models → =ATHENAQUERY() cells
What it runs on
What customers say
“Lets us get a lot more out of our data than [...] We can sit on top of all of the data in [our data warehouse] as opposed to just, like, our little ingested, you know, three gigabytes semantic models. I think we have a lot of power here.”
Analytics lead · A Fortune 500 retailer
“Gave Athena the spreadsheet, explained which column I was trying to figure out, and then ten secs it told me... exactly how the calculation was based on the other data.”
Analyst · A global manufacturer
“One workspace. All different, let's say, avatars of their output data. Right? Suddenly, it's dashboard. Excel is here. PowerPoint is when it's all together. It's very impressive.”
Global Sustainability lead · A global manufacturer
“Digital worker, it gets spun up, it gets given access to the toolkits that it needs. It gets given access to the data that it needs. Just exists for the period it needs to exist to perform the task. And then it gets torn back down again afterwards.”
Audit lead · A global professional services firm
“I like that direction, that concept of the agent is the agent. It becomes an entity in its own right with its own ownership and its own missioning, and you get rid of the gray area and the blurry lines.”
Audit lead · A global professional services firm
“We have a variety of tools today that we have trails for, but not many of them can quickly and easily revert back to the previous versions.”
Partner · A global professional services firm
“Some of the traceability stuff that you guys have within your tool that doesn't exist anywhere else”
IT team · A global manufacturer
“Obviously, you guys have really good security. We trust you guys with your security of what we do.”
Innovation lead · An AmLaw 100 firm
“It's a touchless, you know, seamless. Like, it just gives them whatever they have access to.”
Analytics lead · A Fortune 500 retailer
“This is where the puck's going and this is where the moat is for an enterprise.”
AI Council lead · A Fortune 50 retailer

Verbatim from customer calls. Customers anonymized.

Related

Related products and stories

See this on your own work

Bring one recurring task, a sample input, and an example of work you would accept.