Make · Help

Make Google Sheets Search Rows returns only one bundle?

Several spreadsheet rows but one result? Compare spreadsheet ID, sheet, range and the actual module output before downstream filters.

Source date: 5 November 2026Editorial team: KAN KAI WORKSEditorially updated:

Quick answer

A higher Search Rows limit does not establish that Make selected several rows from the same Google spreadsheet. First compare spreadsheet ID, sheet and range with the source open in your browser. Inspect every output bundle from that module in the current run, then read the same range using the same identity. This separates source differences, search conditions and later router filtering.

Source basis: Q1 · Q2

Applies to
The original case describes three visible test rows and one bundle without a confirmed cause. Filenames, operation indicators and post-router output cannot substitute for inspecting the search module.
How to verify the result
Each labelled test ID appears where the intended search condition requires it; exclusions are explained and subsequent updates use the corresponding row reference rather than a fixed row number.
Your first check

Which exact spreadsheet ID and range did the failing run read?

Google Sheets visibly contains several matching rows, yet Make Search Rows returns only one result despite a higher configured limit. First establish whether the scenario reads the same file and range you are viewing in the browser.

A limit does not prove that several rows matched

An original report describes three visible test rows, no intended filter and only one output bundle. Confusion between similarly named files was discussed, but not established as a universal cause. Q1 The Google Sheets API supports an unambiguous read using spreadsheet ID and a specific range. Q2

These references are more useful diagnostically than a filename. Also distinguish the search module's actual output from bundles later discarded downstream. The final scenario result does not automatically explain what the search returned.

Follow three test records through the flow

  1. Identify the source precisely. Compare the spreadsheet ID in the browser with the ID used by the module. Check the sheet and configured range. Treat similarly named copies and other tabs separately.
  2. Mark test rows. Use three harmless unique IDs in a bounded test range. Record their row numbers and expected values without restructuring a production sheet.
  3. Inspect the current execution. Examine inputs and every output bundle from that specific Search Rows module. Do not substitute an overall operations indicator, an earlier test or the result after a router.
  4. Perform a read-only comparison. Read the same range using an appropriate range-reading operation and the same identity. If that also sees fewer rows, investigate source and range first. If it sees all three, compare search parameters, header settings and filter values after evaluation.
  5. Check downstream processing afterwards. If search returns several bundles, then inspect router filters and mappings. Where multiple rows feed an update, its row reference must correspond to each bundle; a fixed row number can distort the visible final result.

Establish a verifiable expectation

Each test ID should appear exactly where the intended search condition requires it. Account explicitly for records excluded by a filter. Increase the limit only after the existing selection is understood: a larger limit does not correct a wrong source or unsuitable condition.

Retain the three-row example with its actual module outputs. It provides a small comparison that can be repeated after mapping or module changes without exposing customer data or repeatedly updating real rows.

Frequently asked questions

Should I immediately add an Iterator?

Only if the actual output contains an array that needs splitting. Several existing bundles and an array inside one bundle are different data shapes.

What belongs in a support request?

Provide an anonymised spreadsheet identity, sheet and range, module configuration, and input/output from a current run. Three neutral test rows are usually sufficient to make the investigation reproducible.

Help with this case

We can help isolate the affected process and prepare a controlled next step. Discuss case A0399 with KAN KAI WORKS.

Support with this issue

We investigate Make using your error message, versions and a narrowly scoped example.

Describe your case →Related service: AI & automation

Sources

Sources checked on 5 November 2026. Individual reports do not establish general prevalence.