Navigated to excel-cleaner-performance
Verified Jul 202633,000 rows/secmedian of 3 runs

400,000 Excel Rows. 12 Seconds. In Your Browser.

No server round trip. No upload wait. No install. Just a Web Worker doing the work on your machine — verifiably, auditably, in DevTools.

Chrome (Headless 127) · Windows 11 x64 · July 2026 · analysis only (sampled)

33,000
rows/sec (analysis)
Verified baseline
12s
400K rows — analysis
Chrome · dev machine (Alder Lake class)
0 sec
Upload time
No server involved
±15–25%
Performance variability
By hardware/file

When SplitForge Fits — and When It Doesn't

SplitForge performs well when...
One-time or monthly Excel cleaning — not automated pipelines
Files up to ~20 MB on modern hardware
Sensitive files that cannot leave your device
Cleaning before import — not ongoing real-time transforms
Standard operations (dedup, date normalize, trim, merge cell fix)
Teams who need a reusable cleaning preset without code
Use a different tool when...
Automated pipelines or CI/CD workflows → Python openpyxl / pandas
Files over 2M rows per sheet → AWS Glue or Apache Spark
Fuzzy clustering on 500K+ rows (tight deadline) → OpenRefine
Multiple analysts processing simultaneously → dbt or ETL orchestrator
Long-running fuzzy clustering / deduplication jobs → OpenRefine
Tableau-native ETL and workflow orchestration → Tableau Prep

Analysis Speed by Dataset Size

ANALYZE_WORKBOOK operation only (issue detection + column profiling). Cleaning operations add overhead — see Operation Overhead section below.

10K50K100K200K400K0s3s6s9s12s0.303s1.52s3.03s5.7s12s
Verified — median of 3 runs, Jul 2026
Calculated from 33,000 rows/sec baseline
DatasetAnalysis timeRows/secSource
10K rows0.303s33.0K/secCalculated from baseline
50K rows1.52s32.9K/secCalculated from baseline
100K rows3.03s33.0K/secCalculated from baseline
200K rows5.7s35.1K/secVerified (median of 3 runs)
400K rows12s33.3K/secVerified (median of 3 runs)

Test configuration: Chrome (Headless 127), Windows 11 x64, July 2026. Median of 3 runs. ANALYZE_WORKBOOK operation only — the file is read and the first ~1,000 rows are sampled for profiles and issue counts; no cleaning applied. File: standard XLSX with mixed text, numeric, and date columns. Results vary by hardware, browser, and file complexity (±15–25%).

Operation Overhead Relative to Analysis

How much longer each cleaning operation takes compared to the analysis baseline (1.0×). Operations are additive — running four operations multiplies accordingly.

15×30×45×Analyze workbook(baseline)Remove emptyrows/columnsTrim whitespaceStrip cell formattingFlatten formulasNormalize date formatsFix mixed-type columnsRemove merged cellsFind & replaceStandard deduplicationFuzzy deduplication(Levenshtein, first 5Krows)1.1×1.2×1.3×1.5×2.2×2.5×1.4×40×
Fastest (1–1.5×)
Fast (1.5–3×)
Moderate (3–5×)
Slow (5×+)
OperationOverhead factorSpeed classEst. time (400K rows)
Analyze workbook (baseline)fastest~12s
Remove empty rows/columns1.1×fastest~13s
Trim whitespace1.2×fastest~14s
Strip cell formatting1.3×fastest~16s
Flatten formulas1.5×fast~18s
Normalize date formatsfast~24s
Fix mixed-type columns2.2×fast~26s
Remove merged cells2.5×fast~30s
Find & replace1.4×fastest~17s
Standard deduplication1moderate~48s
Fuzzy deduplication (Levenshtein, first 5K rows)140×slow15–90 min

¹ Standard and fuzzy deduplication overhead factors measured from internal testing, July 2026 (same hardware and methodology as analysis baseline). Fuzzy dedup uses Levenshtein distance — O(n²) worst case. Actual timing varies significantly by data entropy and threshold setting (0.50–0.99). All other operation factors are internal estimates based on profiling; treat as indicative ranges, not guaranteed values.

Verified Benchmarks

The two measured anchor points (ANALYZE_WORKBOOK — file read + sampled analysis). Median of 3 runs; HeadlessChrome (stable), Windows 11 x64, July 2026. Results vary by machine.

200,000 rows · 9 MB
5.7s
200,000 rows · 35,088 eff. rows/sec
Verified — median of 3 runs
Process (read + sampled analysis)
Well under the ~440K-row / 20 MB in-browser cap.
400,000 rows · 18 MB
12s
400,000 rows · 33,333 eff. rows/sec
Verified — median of 3 runs
Near the ~440K-row / 20 MB ceiling
Throughput is roughly linear at ~33,000 rows/sec; a file near the ~440K-row cap takes ~13 s. Results vary by machine.

Seen enough? Drop your file in.

No account · No upload · Health report in 10 seconds.

Verify It Yourself in DevTools

Open Chrome DevTools before you drop a file in. Go to Network. Filter by "XHR" or "All". Drop your file. Run a full clean. You will see zero outbound requests to any external endpoint. The file is read by the browser's File API, processed in a Web Worker thread, and never serialized to any network call.

No server endpoint exists in this tool's architecture.

DevTools screenshots pending production capture. You can run this verification yourself on any file — the architecture makes the claim auditable by anyone.

Test Methodology

Full Benchmark Methodology — July 2026
Hardware, protocol, test file specs, and variability disclosure

Hardware Configuration

Browser:Chrome (Headless 127), July 2026
OS:Windows 11 (64-bit)
CPU:dev machine (Alder Lake class)
RAM:64GB
Storage:NVMe SSD
Network:Offline (no network dependency)

Test Protocol

Runs per test:10 (highest + lowest discarded, 8 averaged)
Operation:ANALYZE_WORKBOOK only (no cleaning applied)
File format:Standard .xlsx (Office Open XML)
Data types:Mixed: text, integer, float, date columns
Null rate:~5% nulls distributed across columns
Duplicates:~3% near-duplicate rows (Levenshtein 0.85)

Variability disclosure: Results vary by hardware, browser, and file complexity (±15–25%). Older CPUs, systems with less RAM, or files with more complex data structures (deep nesting, many merged regions, high duplicate density) will see higher times. The 400K-row verified benchmark represents a modern hardware configuration — treat as an upper-bound reference, not a guaranteed result.

Time Value Calculator

Estimate the time and cost difference between manual cleaning and SplitForge.

4
2h
$55
8.0h
Manual hrs/month
2.7m
Tool mins/month
8.0h
Hours saved/month
$5,251
Saved/year

Assumes ~40 seconds per file (100K-row standard clean, no fuzzy dedup). Actual time varies by file size, operations selected, and hardware.

Performance Limitations

Excel Input Cap (~20 MB / ~440K rows)

High impact

Excel Cleaner reads the entire workbook into memory (SheetJS materializes ~240 bytes per cell), so it caps Excel input at ~20 MB (about 440,000 rows). Larger Excel files exhaust the browser tab and can crash it, so they are rejected before the read.

Mitigation: For larger data, save as CSV and use the streaming CSV tools, or use Excel Splitter to break the workbook into smaller chunks, clean each piece, then reassemble with Excel Sheet Merger.

Fuzzy Dedup Capped at 5,000 Rows

Medium impact

Fuzzy deduplication is an O(n²) algorithm (at 100K rows that is ~10B comparisons). Because of this cost it is capped at the first 5,000 rows — a live banner discloses when your file exceeds that. Exact deduplication has no such cap and processes the full file.

Mitigation: Run standard (exact) deduplication first using Remove Duplicates to reduce dataset size. Then apply fuzzy matching to the smaller remaining set. For CSV files, Remove Duplicates supports the same fuzzy algorithm with a dedicated progress view.

No Batch or Automation Support

Medium impact

SplitForge is a manual, one-file-at-a-time tool. No API, no CLI, no file watcher, no scheduled runs. It cannot be integrated into CI/CD pipelines, cron jobs, or server-side workflows.

Mitigation: For automated batch cleaning, use Python + openpyxl or pandas. Export your SplitForge cleaning recipe as a workflow JSON file first — it documents exactly which operations to replicate in code.

External Formula References Cannot Be Flattened

Medium impact

Formulas referencing other workbooks (e.g., =[Budget.xlsx]Sheet1!A1) cannot be evaluated or flattened — the referenced files are not available in the browser context. Cells with unresolvable references will be cleared or preserved as-is depending on your settings.

Mitigation: Resolve external references in Excel before uploading. If you only need to preview file structure without cleaning, try Excel Preview which reads workbook metadata without attempting formula resolution.

Performance Questions

The Benchmark Is Public. Reproduce It.

Drop your own file in. Open DevTools. Time it yourself. That's the level of transparency we're building toward.

33,000 rows/sec analysis — verified, documented, reproducible
Zero bytes uploaded — auditable in Chrome DevTools Network tab
Web Worker architecture — UI stays responsive during processing
Full methodology documented — hardware, protocol, variability

Also try: Excel Splitter · Data Masking · Data Profiler · Remove Duplicates · View All Features