Skip to content
Download
FOR DATA ANALYSTS & BI

Workbook Connect for data analysts and BI teams

The numbers add up, but the category is wrong. Fix the dimension behind the report in Excel, validated against an allowed set, and commit it straight to the warehouse. No data request, no ticket, no waiting for the next load.

Download free

Fix the data behind the dashboard

Correct a Category attribute in DimProduct and commit. The dimension your model reads from is right at the source, so the next refresh shows it.

EXCELDATA WAREHOUSE

The totals are right. The labels are wrong.

You know this bug because it never shows up as a wrong number. The grand total ties out, but the chart has both "Footwear" and "Shoes" as separate bars, a discontinued SKU still appears in the active range, or a product rolls up under the wrong category so a manager's filter quietly drops half their sales. The measure is fine. The model is fine. The defect is one stale attribute in a dimension table, and it is making the whole report look untrustworthy to the people who read it.

Where the defect actually lives

Your report groups FactSales by attributes that live in DimProduct, joined on ProductKey. A wrong Category or a duplicate spelling in the dimension is what splits a bar or misroutes a rollup. That dimension is exactly the table Workbook Connect lets you edit.

∞1ProductKeyDimProductProductKeyPKSKUProductNameCategorySubcategoryIsActiveFactSalesSalesKeyPKProductKeyFKDateKeyQuantityNetAmount

An illustrative data model, not your exact schema.

Fix the dimension yourself, in the grid you already use

Instead of filing a request and waiting for someone with write access to run an UPDATE, you open DimProduct in Excel through the ribbon, filter to the SKUs in question, and retype the attribute. Pick a Category and a foreign-key lookup offers only the real values, so you collapse "Shoes" back into the canonical "Footwear" without inventing a third spelling. Re-parent a Subcategory, flip an IsActive flag, or correct a description the same way. Because the table configuration (lookups, filters, primary-key mapping, and the hidden technical columns) is set once and shared to the team on connect, every analyst opens the same guard-railed sheet rather than their own private version of the truth.

Cleaning up DimProduct as you edit it

Standardise the Category values that are splitting your report, watch the per-row status track every change, and let validation catch the one typo that would create a phantom group.

DimProduct.xlsx (product cleanup)
StatusSKUProductNameCategorySubcategoryIsActive
ChangedSKU-4471Trail Runner GTXFootwearRunningY
ChangedSKU-4472Trail Runner GTX WmnsFootwearRunningY
SKU-5108Alpine Down JacketApparelOuterwearY
Needs fixSKU-5109Alpine Down Jacket (2024)ApparellOuterwearN
NewSKU-6630Summit Trekking PolesEquipmentHikingY
ChangedSKU-6204Merino Base LayerApparelBaselayersY
Commit Validated, then written to dbo.DimProduct
  • Row 1: moved from "Shoes" to the canonical "Footwear" value, so it stops splitting the Footwear total in the sales report.
  • Row 4: "Apparell" is not one of the allowed Category values (Footwear, Apparel, Accessories, Equipment), so the commit is blocked until this row is fixed. Nothing is written until every row passes, then they all commit together.

An illustrative example of the editing workflow, not a product screenshot.

Validated before anything is written

A direct edit to a dimension is exactly where reporting goes wrong, so Workbook Connect validates the whole sheet in Excel first and nothing is written until every row passes. A validation rule constrains Category to its allowed set (Footwear, Apparel, Accessories, Equipment), the foreign keys keep every ProductKey resolvable, and the per-row status column marks new, changed, and flagged-for-deletion rows before you commit. The "Apparell" typo above does not slip through and quietly create a new group in tomorrow's chart. If any row fails, the commit is blocked and nothing is written; the failing rows stay in the sheet with their reasons, so you fix them and commit again, and then every row is written to DimProduct together in one transaction, under the database permissions you already have, with conflicting edits detected if a teammate touched the same row first. Authoring those rules is the Validation Designer (a Pro feature), but the rules and roles themselves are enforced for everyone, including the free plan.

What analysts use it for

Works with your warehouse

FAQ

Analysts and BI, answered

You stay in Excel, with no UPDATE or INSERT statements to write and no risk of an unbounded UPDATE. You browse the dimension table, type new values into the grid, and a per-row status marks each row as new, changed, or flagged for deletion before anything is written. On commit the changed rows go back in one transaction, and optimistic concurrency catches the case where someone else touched the same row since you loaded it.

The commit writes to the source dimension table your model reads from, for example DimProduct in the warehouse. Refresh your semantic model or dataflow and the corrected Category or Subcategory flows through to every visual that groups or filters on it. Workbook Connect changes table data, not your model, so the refresh cadence is whatever you already have.

Two ways. A foreign-key lookup configured on the table offers only the real values for a column (so ProductKey or a Subcategory code can only reference rows that exist), and an authored validation rule can constrain a free-text column to an allowed set, for example Category IN (Footwear, Apparel, Accessories, Equipment). If any row fails, the commit is blocked and nothing is written; the failing rows stay in the sheet with the reason, so a typo like "Apparell" never reaches the warehouse and never silently spawns a phantom row in your report.

Yes. A self-referencing dimension (a Subcategory pointing at its parent Category row, or a cost-center rollup) is just another table with a foreign key back to itself. You edit the parent reference in the grid and the lookup only lets you point the parent reference at a row that exists, so the link stays valid.

No more than you have now. You sign in with your existing authentication (Entra ID with MFA, Entra ID username and password, Service Principal, Managed Identity, Windows Integrated, SQL auth, and Default Azure Credential) and every read and write honours the permissions already on the database. Workbook Connect makes no server-side schema changes, so it cannot grant you anything you were not already granted.

Yes. It connects live to Microsoft Fabric Warehouse, Fabric SQL Database, Azure SQL Database, and SQL Server today, over TLS 1.2 or higher. PostgreSQL, Databricks Lakebase, and Databricks SQL Warehouse are coming soon with early access.

Correct the dimension once and every visual follows

Download Workbook Connect free, point it at your warehouse, and maintain the dimensions and mappings your dashboards depend on without leaving Excel. The next refresh reads the corrected dimension, so every visual that groups on Category shows the fixed value.

Download free