Combine Queries: Merge, Append & Parameters

Real ETL is more than cleaning one file

Earlier you cleaned a single CSV in Power Query. On a real job you usually have to combine data first — stacking this year’s file under last year’s, or pulling a customer’s city onto each sale. The word for all of this prep work is ETL: Extract (get the data), Transform (clean and combine it), Load (send it to the model). This lesson covers the two combining moves every analyst needs — append and merge — plus two time-savers, parameters and reference queries.

Both combining moves live on the Home ribbon of the Power Query Editor, under Combine. Let us define the two clearly before doing each one.

A simple memory hook: append makes a table taller (more rows), merge makes it wider (more columns). Pick the one that matches the shape you need.

Append — stack same-shaped tables

Imagine two files with the exact same columns: Sales2025 and Sales2026. To analyse them together you want them as one table. Append puts the 2026 rows directly under the 2025 rows. Here is exactly what you click:

1. In Power Query, click Home → Append Queries → Append Queries as New.
2. Choose the first table (Sales2025) and the second (Sales2026).
3. Click OK — a new query appears with every row from both.

Sales2025  (120 rows)        Sales2026  (140 rows)
  Date Product Amount           Date Product Amount

            --- Append ---->

AllSales  (260 rows)   = 2025 rows, then 2026 rows underneath

Merge — join columns by a matching key

Merge is different: it brings columns from another query and lines them up by a matching key (a shared column), just like a database join. Say your Sales table has a CustomerID but no city, and a Customers table has CustomerID and City. Merge pulls City onto each sale.

1. Click Home → Merge Queries.
2. Pick the left table (Sales) and click its CustomerID column to select the key.
3. Pick the right table (Customers) and click its CustomerID column.
4. Choose a join kind (start with Left Outer — keep all Sales rows), then OK.
5. A new column appears; click its expand (↔) icon and tick City to pull it in.

Sales                       Customers
  CustomerID Amount            CustomerID City
  7          1200              7          Delhi
  12         450               12         Pune

      --- Merge on CustomerID (Left Outer) --->

Sales (now with City)
  CustomerID Amount  City
  7          1200    Delhi
  12         450     Pune

Reference vs Duplicate — reuse without re-cleaning

Sometimes you want a second version of a query — say a summary built from a cleaned table. Right-clicking a query offers two options that sound alike but differ:

• Duplicate: makes an independent copy with its own copy of all the steps. Change the original and the copy does not follow.
• Reference: makes a new query that points back to the original and starts from its result. Fix a step in the original and every reference updates too.

Use Reference when several queries should share the same cleaning — clean once in a base query, then reference it for each summary. Use Duplicate when you want a true fork that can drift apart.

Parameters — one value, controlled in one place

A parameter is a named, reusable value you define once and plug into many places — a file path, a year, a server name. Change the parameter and everything using it updates. To make one: Home → Manage Parameters → New Parameter, give it a name (e.g. SourceYear) and a value (e.g. 2026).

Parameter:  SourceYear = 2026

Used in a filter step:   Keep rows where  Year = SourceYear
Change the parameter to 2025  ->  every query using it now reads 2025