Upload CSV.
Get Your Dashboard.

Upload any Speedtest CSV, set a password, and instantly generate a full analytics dashboard with bias-aware statistics, charts, and a password-protected map.

Export from Speedtest by Ookla: Results tab → Export CSV. Or use any CSV with date, download, upload, ping, and location columns. Column names are auto-detected but can be remapped below.
```
1
Upload Speedtest CSV
📂
Drop CSV here or click to browse
Speedtest export, .csv / .tsv · any column order · UTF-8
2
Map Columns (auto-detected — adjust if needed)
3
Set Password (protects Map & Legend tabs)
Anonymize location names with zone codes (e.g. ZONE-A)
Label the most-tested location as HOME
4
Dashboard Title (optional)
All data stays in your browser · nothing is uploaded
```
NETWORK ANALYTICS
Summary
Download by Location · Mean (failures = 0 Mbps)
Key Findings
3-Method Comparison
Same data, three statistical lenses. Failed tests counted as 0 Mbps in all methods.
Mean— inflated by outliers & burst sessions
Median— robust, best for skewed data
Best— capability ceiling only
Download Speed · All 3 Methods (Mbps)
LocationNetworkn (fails)MeanMedianBestSpread
Interpreting the three methods
Mean says…
Heavily-tested locations and burst sessions skew averages up. One great day can inflate the mean for weeks of mediocre service.
Median says…
Resists outliers completely. If you had one 500 Mbps spike and ten 10 Mbps tests, median correctly shows ~10. The honest number.
Best says…
Shows what's physically possible at each location under ideal conditions. Useful for ceiling comparisons, useless for typical experience.
Takeaway
Mean → useful for trending. Median → honest typical experience. Best → capability only. All three together tell the real story.
WiFi vs Cellular
Different network types serve different purposes. Direct comparison can mislead — context is everything.
WiFi Only
Cellular Only
Side-by-Side
MetricWiFiCellular
*Mean may be inflated by high-outlier locations. See Median for honest comparison.
Why direct comparison can mislead
Different scope
WiFi is usually tested at one fixed location. Cellular spans many locations including dead zones. It's not a fair head-to-head.
Mean hides the truth
A few stellar cellular tests make the mean look competitive with WiFi. The median reveals the typical daily experience, often far lower.
Where cellular can win
Near strong 5G towers, cellular can massively outperform any home broadband. The "best" metric reveals this ceiling.
Where WiFi wins
Consistency, symmetric upload, low latency, zero dead zones. For streaming, video calls, and gaming — fixed broadband dominates.
Reliability · Failed Tests = 0 Mbps
% of tests that returned usable speed results. Both including and excluding failures shown to reveal the impact on averages.
Reliability by Location
LocationNetworkTestsFailsReliabilityMean (w/ 0s)Mean (no fails)
Reliability Bars · green = success · red = failure proportion
Charts
Five interactive views. Switch between them using the tabs below.
Download Speed · All Tests
Mean vs Median · Download by Location (Mbps)
Reliability % by Location
All Tests · Newest First
DateNetwork / ServerLocationDownloadUploadPingStatus
🗺️
Map — Password Required
Real geographic locations are shown here. Enter the password you set to unlock.
🔑
Legend — Password Required
This tab maps anonymized zone codes back to real location names from your CSV.