19 Functions · SQL HAVING Filters · 10M Rows in 34 Seconds · 100% Private

The Excel Pivot Table Alternative for Large Files. SQL Power. No Database. No Uploads.

GROUP BY up to 10 dimensions. Apply 19 aggregation functions including MEDIAN, percentiles, and STDDEV. Filter groups with SQL HAVING logic. A browser-native aggregation engine for CSV — the analytical power of OLAP without infrastructure, uploads, or code.

10M rows in 34 seconds19 aggregation functionsSQL HAVING filtersFiles never uploadedSupports HIPAA/GDPR-sensitive workflows

Benchmarked February 2026 · Chrome 132 · Intel i7-12700K · 32GB RAM · Full methodology →

The Excel Pivot Table Problem

Freezes at 300K–500K rows

Excel Pivot Tables become unusable on large datasets. You sit watching the spinning cursor while your deadline passes.

No HAVING filters

"Show only regions where revenue > $50K" requires copying pivot results to a new sheet, applying manual filters, and repeating every time data changes.

No MEDIAN, no percentiles

Excel Pivot Tables don't support true MEDIAN or P90/P95 percentiles. AVERAGE is misleading when your data has outliers — which it always does.

Everything Excel Pivot Tables Can't Do

SQL HAVING Filters

Filter groups after aggregation using the full SQL HAVING logic. AND/OR combinations. Show only the groups that meet your threshold — not the full table.

19 Aggregation Functions

COUNT, SUM, AVG, MIN, MAX, MEDIAN, MODE, STDDEV, VARIANCE, P25, P50, P75, P90, P95, FIRST, LAST, RANGE, COUNT_DISTINCT, CONCAT_UNIQUE.

Multi-Level GROUP BY

Group by up to 10 dimensions with drag-to-reorder. Hierarchical subtotals and grand total row included. Region → Country → State → City — all in one pass.

Top-N with Correct Other Rollup

Show only the top 100 SKUs by revenue. All remaining groups become a single 'Other (47,832 SKUs)' row. Critically: AVG for Other is recalculated as a true weighted average — not average of averages, which would be mathematically wrong.

% of Total Columns

Automatically add '% of Total' columns for every metric. SUM(Revenue) and SUM(Revenue) % of Total side by side — no manual formula required.

Zero Uploads — Supports HIPAA/GDPR Workflows

All processing runs in your browser using Web Workers. No file contents are transmitted to our servers — not during upload, not during processing, not at any point. Designed for HIPAA/GDPR-sensitive workflows by architecture.

What Does Slow Aggregation Cost?

Excel Pivot on large files: ~20 min setup + wait. SplitForge: ~2 min.

Hours saved per year (4 analyses/week)
57.6h
Annual value at $60/hr
$3,456

Baseline: 20 min derived from internal timing of Excel Pivot Table setup and processing on 500K–1M row exports. Results vary by file size and workflow. Adjust assumptions with the full calculator →

SplitForge vs Excel Pivot Tables — Feature Comparison

See exactly where Excel Pivot Tables stop working — and where SplitForge picks up.

FeatureSplitForgeExcel PivotPower BISQLPython/Pandas
Row Limit10M+ rows tested300K–500K (struggles)5M (dataset limits)Database-dependent10M+ (memory-bound)
Aggregation Functions19 functions11 basic functionsMany (DAX)Full SQLUnlimited (code)
MEDIAN / True PercentilesYesNoVia DAX MEDIAN()PERCENTILE_CONT()Yes (code)
HAVING FiltersYes — visual GUINoVia DAX (complex)Yes (HAVING clause)Via .query() (code)
Multi-Level GROUP BYUp to 10 dimensions2–3 (degrades)Many dimensionsUnlimitedUnlimited (code)
Data PrivacyNever uploadedLocalCloud uploadsDepends on hostLocal
No Code RequiredYesYesMostly (DAX needed)NoNo
CostNo subscriptionIncluded with Excel$10/user/monthDB infrastructureOpen source (code required)
Setup Time< 30 seconds2–5 minutesHours (data model)Hours (DB pipeline)5–30 min (script)

* Power BI dataset row limits depend on licensing tier and whether Premium capacity is enabled. Prices and limits based on publicly listed plans as of February 2026.

Which Tool Fits How You Think?

Use SplitForge if you…

Need results in 2 minutes, not 20
Have a file too large for Excel Pivot to handle
Can't upload sensitive data (healthcare, finance, HR)
Need MEDIAN, percentiles, or STDDEV — not just AVERAGE
Want SQL HAVING logic without learning SQL
Work with business users who can't run Python scripts
Need hierarchical subtotals in a downloadable CSV

Stick with alternatives if you…

Need scheduled / automated aggregation pipelines → Python pandas or dbt
Are already running data in AWS → AWS Glue or Athena
Have 50M+ row files → database or Spark
Need interactive drill-down dashboards → Power BI or Tableau
Require version-controlled aggregation logic → dbt models
Need pivot table wide-format output → Excel Pivot Table is fine if file is under 300K rows

Real-World Aggregation Workflows

Six workflows where Excel Pivot Tables break down — and SplitForge picks up.

Sales Performance Reports

The Problem

500K transaction rows. Finance needs regional performance (Region → State → Rep) with revenue, order count, and average deal size — by 9am Monday.

With SplitForge

GROUP BY Region/State/Rep. SUM(Revenue), COUNT(Orders), AVG(Deal), MEDIAN(Deal). HAVING Revenue > $100K. Hierarchical subtotals. Download. 23 seconds total.

E-commerce SKU Analysis

The Problem

2M order lines. Need SKU-level metrics — units sold, revenue, unique customer count — across 50K SKUs. Excel freezes at 500K rows.

With SplitForge

GROUP BY SKU/Category. SUM(Units), SUM(Revenue), COUNT_DISTINCT(CustomerID), P50/P90(Price). HAVING Units >= 100. Top 500 SKUs + Other rollup. Under 90 seconds.

Healthcare Outcomes Analysis

The Problem

1M patient records. Need outcomes by Facility → Department → Procedure. HIPAA requires no uploads to third-party servers.

With SplitForge

GROUP BY Facility/Dept/Procedure. AVG(StayDays), STDDEV(StayDays), COUNT(Readmissions), VARIANCE(Cost). HAVING AvgStay > 5 OR Readmissions > 20. 100% client-side. 45 seconds.

Expense Reporting for Finance

The Problem

800K expense rows. CFO needs Department → Category → Vendor breakdown with totals, frequency, and P95 outlier detection.

With SplitForge

GROUP BY Dept/Category/Vendor. SUM(Amount), COUNT(Txns), AVG(Amount), P95(Amount). HAVING SUM(Amount) > $10K. % of Total columns for share-of-budget view. 18 seconds.

Marketing Attribution

The Problem

3M campaign events. Need Channel → Campaign → Creative performance with conversions, CPA, and ROI — before the morning standup.

With SplitForge

GROUP BY Channel/Campaign/Creative. SUM(Conversions), SUM(Spend), AVG(CPA), COUNT_DISTINCT(Users), MEDIAN(CPA). HAVING Conversions > 50 AND AvgCPA < $25. Top 100 campaigns. 52 seconds.

Inventory Turnover Optimization

The Problem

1.5M inventory movements across 200 warehouses. Need turnover rates, stock-out frequency, and reorder patterns by Warehouse → SKU → Vendor.

With SplitForge

GROUP BY Warehouse/SKU/Vendor. FIRST(DateIn), LAST(DateOut), RANGE(Days), AVG(TurnoverDays), COUNT(StockOuts). HAVING TurnoverDays > 90 OR StockOuts > 3. 38 seconds.

Edge Cases & Technical Depth

How SplitForge handles the aggregation problems most tools quietly get wrong.

MEDIAN vs AVERAGE — When It Actually Matters

When your data has outliers, AVERAGE lies. MEDIAN tells the truth.

HAVING vs WHERE — Why You Need Post-Aggregation Filtering

Filtering before aggregation and after aggregation are completely different operations.

Multi-Level GROUP BY at Scale — 10 Dimensions, 10M Rows

Excel Pivot Tables often become slow and memory-constrained above 3 dimensions depending on system RAM. SplitForge handles 10.

Mixed Data Types in GROUP BY Columns

What happens when the same column contains '123', 123, and '123.0'?

COUNT_DISTINCT at 10M Rows — Memory vs Speed Trade-off

Counting unique values across millions of rows requires storing a Set per group.

Top-N 'Other' Rollup — When Your Last-Place Groups Still Matter

Top-100 SKUs plus one 'Other (47,832 SKUs)' row — full picture without full detail.

Who SplitForge Is — and Isn't — Built For

Great fit for

Data analysts and ops teams aggregating CSVs manually each week
Finance teams generating executive reports from 500K+ transaction exports
Healthcare teams analyzing patient data that cannot be uploaded (HIPAA)
Marketing teams measuring campaign performance across millions of events
Operations teams reporting on inventory, logistics, or supply chain data
Any workflow where Excel Pivot Tables freeze, crash, or hit row limits

Not the right fit if you need

Scheduled / automated aggregation pipelines
→ Use Python pandas + cron or dbt
50M+ row files
→ Use a database or Spark
Interactive drill-down dashboards
→ Use Power BI or Tableau
Pivot table wide-format output
→ Excel Pivot is fine if under 300K rows
Version-controlled aggregation logic
→ Use dbt models or SQL views
Cross-file join before aggregation
→ Use VLOOKUP/Join tool first, then aggregate

10 Million Rows in ~34 Seconds

Full advanced mode — 19 functions including MEDIAN, P90, STDDEV — on a 1.2GB CSV. All in your browser, zero uploads.

File Size
~1.2 GB
Total Rows
10M
Advanced Mode
~34 sec
Simple Mode
~26 sec
Test config: Chrome 132 (stable), Windows 11, Intel i7-12700K, 32GB RAM, February 2026
Advanced mode: 19 functions including MEDIAN, P90, P95, STDDEV, COUNT_DISTINCT
Method: 10 runs, highest/lowest discarded, 8 averaged
Variance: Results vary by hardware, browser, and configuration complexity (±15–20%)

Frequently Asked Questions

Complete Your Data Workflow

Stop Fighting Excel's Limits.

19 aggregation functions. SQL HAVING filters. Multi-level subtotals. 10M rows in under a minute. Browser-native OLAP for CSV — analytical power without infrastructure, uploads, or code.

10M+ rows — no row limits
Files processed entirely in your browser. No contents transmitted to our servers.
SQL HAVING filters Excel cannot do
MEDIAN, percentiles, STDDEV included

Also try: Pivot/Unpivot · VLOOKUP/Join · Data Cleaner · Data Profiler