Does Excel have a QUERY function?

No. Excel does not have a QUERY function as of July 2026, and Microsoft has not announced plans to add one. The practical replacement is a combination of FILTER (WHERE), SORT or SORTBY (ORDER BY), UNIQUE (SELECT DISTINCT), and SUMIFS with spill references (GROUP BY). All of these require Excel 365 or Excel 2021 - they are not available in Excel 2019 or earlier.

How do I replicate Google Sheets QUERY GROUP BY in Excel?

Use UNIQUE to generate the dimension list in one column, then SUMIFS with a spill reference to aggregate in the next. Put `=SORT(UNIQUE(dimension_column))` in cell A2, then `=SUMIFS(value_column, dimension_column, A2#)` in B2. The `A2#` spill reference makes SUMIFS calculate a result for every unique value in the spill range automatically - equivalent to `SELECT dim, SUM(val) GROUP BY dim ORDER BY dim` in 2 formulas instead of 1.

Why does my SORTBY formula return a #VALUE! error when combined with FILTER?

The sort array and the output array must be exactly the same number of rows. When you combine SORTBY with FILTER, you need to apply the identical filter condition to both the output range and the sort key separately. A single extra row in either array - from a slightly different condition or range boundary - causes the mismatch. Check that both FILTER conditions are character-for-character identical and reference the same row bounds.

Can FILTER combine data from multiple sheets in Excel?

FILTER operates on a single contiguous range. To combine data from multiple tabs before filtering, use VSTACK (available in Excel 365 since 2022): `=FILTER(VSTACK('Q1 Actuals'!A2:F500, 'Q2 Actuals'!A2:F500, 'Q3 Actuals'!A2:F500), condition)`. VSTACK stacks the ranges vertically into a single array that FILTER evaluates normally. The column structure across all stacked ranges must be identical.

How do I select specific non-contiguous columns from FILTER output?

Wrap the source range in `CHOOSE({1,2,3}, col_A, col_C, col_F)` before passing it to FILTER: `=FILTER(CHOOSE({1,2,3}, A2:A1000, C2:C1000, F2:F1000), condition)`. The CHOOSE array constant `{1,2,3}` tells it to return 3 columns in that order. This is the equivalent of `SELECT A, C, F WHERE...` in QUERY - you can reorder columns or skip any column from the source range.

Excel QUERY Function for Financial Models (2026)

The 4 QUERY patterns FP&A analysts use most often are WHERE, ORDER BY, SELECT DISTINCT, and GROUP BY. Excel handles 3 of them cleanly and one with a workaround. Here's what each replacement looks like when the data is real.

QUERY Syntax vs. Excel Equivalent: Quick Reference

QUERY clause	Excel equivalent	Notes
`SELECT A, B, C`	`CHOOSE({1,2,3}, A:A, B:B, C:C)`	Wrap around FILTER for column selection
`WHERE x = 'val'`	`FILTER(range, condition)`	Use `*` for AND, `+` for OR
`ORDER BY D DESC`	`SORTBY(range, sort_col, -1)`	Sort key can be outside output range
`SELECT DISTINCT A`	`UNIQUE(A:A)`	Composes cleanly with FILTER and SORT
`GROUP BY A, SUM(B)`	`UNIQUE()` + `SUMIFS(…, A2#)`	Spill reference is the key
`LIMIT n`	`TAKE(result, n)`	Available Excel 365 (2022+)

According to Google's Visualization API Query Language documentation, QUERY uses "a query language similar to SQL" with 12 supported clauses. Excel's dynamic array stack covers 8 of them reliably as of July 2026.

WHERE → FILTER (With Column Selection)

FILTER returns entire rows. QUERY lets you specify columns inline - SELECT B, D, F WHERE.... FILTER doesn't.

The fix is CHOOSE with an array constant to pick specific columns before or after FILTER evaluates. For a revenue detail pull filtering ARR contracts above $500K:

=FILTER(
  CHOOSE({1,2,3},
    'Revenue Detail'!B2:B5000,   -- Account name
    'Revenue Detail'!D2:D5000,   -- Revenue type
    'Revenue Detail'!F2:F5000),  -- Contract value
  ('Revenue Detail'!C2:C5000 = "ARR") *
  ('Revenue Detail'!F2:F5000 >= 500000)
)

The * operator is AND logic. + is OR. For multi-condition filters reaching across tabs into your assumptions sheet:

=FILTER(
  CHOOSE({1,2,3,4},
    'GL Detail'!A2:A10000,
    'GL Detail'!C2:C10000,
    'GL Detail'!D2:D10000,
    'GL Detail'!F2:F10000),
  ('GL Detail'!B2:B10000 = Assumptions!$B$4) *
  ('GL Detail'!E2:E10000 >= Assumptions!$C$3) *
  ('GL Detail'!E2:E10000 <= Assumptions!$D$3)
)

Entity filter anchored to Assumptions!$B$4, date range from $C$3:$D$3. This is the formula structure behind a quarterly actuals pull for a board pack - the kind where the CFO asks you to re-run it for a different entity 20 minutes before the meeting.

ORDER BY → SORTBY

SORT handles single-column ascending/descending. SORTBY is what you need when the sort key isn't in your output, which happens constantly in contribution margin tables, deal rankings, and SKU performance summaries.

Sorting active SKUs by gross margin dollars descending, displaying only name and revenue:

=SORTBY(
  FILTER(
    CHOOSE({1,2}, 'SKU Detail'!A2:A500, 'SKU Detail'!C2:C500),
    'SKU Detail'!E2:E500 = "Active"
  ),
  FILTER('SKU Detail'!D2:D500, 'SKU Detail'!E2:E500 = "Active"),
  -1
)

The sort array must match the output array row-for-row. The inner FILTER on the sort key has to use the identical condition as the outer one, or you get a #VALUE! error. That's the main trip wire here.

SELECT DISTINCT → UNIQUE

QUERY's SELECT DISTINCT is UNIQUE in Excel, and it composes more cleanly.

A dynamic dimension list for a cost center variance report, sorted alphabetically:

=SORT(UNIQUE('GL Detail'!D2:D10000))

Two functions. In QUERY you'd write SELECT DISTINCT D ORDER BY D and still need to handle the header row separately.

Where UNIQUE earns its place in a real model: generating row labels for a summary table that auto-expands as new cost centers appear in the GL. No pivot table refresh, no hardcoded list to maintain, no broken reference when finance adds a new department mid-year.

GROUP BY - The One That Requires a Workaround

This is where the QUERY analogy breaks. QUERY's GROUP BY with aggregation is a single formula. Excel has no direct equivalent - you need 2 formulas working together.

The pattern: UNIQUE generates the dimension list, SUMIFS with a spill reference does the aggregation.

In column A (dimension labels, spills down automatically):

=SORT(UNIQUE('GL Detail'!B2:B3000))

In column B (aggregated EBITDA, one formula that covers the entire column):

=SUMIFS('GL Detail'!F2:F3000, 'GL Detail'!B2:B3000, A2#)

The A2# spill reference is the key. SUMIFS evaluates against every value in the spill range and returns a matching array, one result per business unit. This replicates SELECT B, SUM(F) GROUP BY B ORDER BY B in 2 formulas instead of 1.

For a multi-condition group-by - say, EBITDA by entity and cost category, pulling actuals from a P&L tab with 5,000+ rows:

=SUMIFS(
  'P&L'!D2:D5000,                                   -- Values to sum
  'P&L'!B2:B5000, Dimensions!A2#,                   -- Entity (spill ref)
  'P&L'!C2:C5000, ">=" & Assumptions!$C$3,          -- Period start
  'P&L'!C2:C5000, "<=" & Assumptions!$D$3,          -- Period end
  'P&L'!E2:E5000, "Operating Expense"               -- Line item type
)

This is more verbose than QUERY but more auditable. Every condition is visible. Your reviewer can trace each argument without running the formula in a test environment.

The Version Constraint Nobody Mentions

FILTER, SORT, SORTBY, and UNIQUE require Excel 365 or Excel 2021. They're not available in Excel 2019 or earlier. According to Microsoft's Excel documentation, these dynamic array functions were introduced in Excel 365 starting in 2019, with Excel 2021 perpetual licenses adding them in October of that year.

If your model needs to open on legacy Excel - common in banking where IT policy can run 3+ years behind current releases - every dynamic array formula breaks. Your FILTER becomes #NAME?. You're back to SUMIFS for aggregation, INDEX/MATCH for filtering, and pivot tables for anything GROUP BY-adjacent.

That's a real constraint in bank syndicate DCF work where the counterparty opens your file in a controlled environment running Excel 2016 or 2019. QUERY has the same portability problem in Sheets (it can't reach external files), but at least it works consistently within the platform.

What QUERY Does That Excel Still Can't Match Cleanly

Two patterns don't translate well:

Inline calculated columns. QUERY allows SELECT A, B*C AS gross_profit WHERE... and returns the calculated column in the output. FILTER returns stored values only. You need a helper column, or BYROW/LAMBDA, which works in Excel 365 but isn't standard practice in most FP&A models as of mid-2026.

The PIVOT clause. QUERY can transpose row values into column headers dynamically. There's no Excel equivalent short of VBA or Power Query. If your QUERY was doing cross-tab pivoting inline, that's the use case that genuinely has no clean replacement in standard Excel formulas.

Rebuilding QUERY Formulas at Scale

If you're migrating a Sheets model with 15-20 QUERY formulas to Excel, the translation work is mechanical but tedious. Each formula has to be decomposed into component operations, column selections rewritten with CHOOSE arrays, and GROUP BY logic split into UNIQUE + SUMIFS pairs.

ModelMonkey handles this inside the spreadsheet - describe the QUERY you had in Sheets, and it writes the FILTER/SORT/UNIQUE equivalent matched to your actual data layout and tab structure. For a full model migration that would take a senior analyst 2-3 hours to do cleanly by hand, it runs in minutes. Try ModelMonkey free for 14 days - it works in both Google Sheets and Excel.