Nav0 vs Chrome: Which Browser Actually Respects Your Data?
March 14, 2026 · 10 min read
Last week we published a performance benchmark showing that nav0 uses roughly half the memory and a fraction of the CPU compared to Chrome. But performance is only part of the story.
This time, we measured something different: how much data each browser actually transfers when you browse the web. Every HTTP request, every tracker pixel, every background ping — we captured it all using the Chrome DevTools Protocol's network interception, then compared the two browsers side by side across 15 real websites.
The short version: Chrome transferred 44.02 MB. Nav0 transferred 36.22 MB. That's 17.7% less data. But the raw totals don't tell the full story. Where Chrome and nav0 really diverge is in what that data is and who it's for.
The Numbers
We opened the same 15 websites in both browsers and measured every network request — bytes sent, bytes received, request category, and resource type. Here's what we found.
Per-Page Data Consumption
| Site | Chrome | Nav0 | Difference | Winner |
|---|---|---|---|---|
| news.ycombinator.com | 11.33 KB | 11.36 KB | +0.3% | Chrome |
| text.npr.org | 2.79 KB | 2.79 KB | +0.0% | Nav0 |
| en.wikipedia.org | 857.71 KB | 830.86 KB | -3.1% | Nav0 |
| craigslist.org | 465.36 KB | 775.62 KB | +66.7% | Chrome |
| lite.cnn.com | 83.41 KB | 38.74 KB | -53.6% | Nav0 |
| developer.mozilla.org | 708.96 KB | 524.51 KB | -26.0% | Nav0 |
| docs.github.com | 441.27 KB | 435.34 KB | -1.3% | Nav0 |
| stackoverflow.com | 2.09 MB | 1.46 MB | -30.2% | Nav0 |
| npmjs.com | 1.12 MB | 1.11 MB | -0.2% | Nav0 |
| github.com | 1.14 MB | 1.49 MB | +30.8% | Chrome |
| reddit.com | 1.51 MB | 1.27 MB | -15.3% | Nav0 |
| youtube.com | 460.81 KB | 476.91 KB | +3.5% | Chrome |
| bbc.com | 4.41 MB | 2.08 MB | -52.8% | Nav0 |
| edition.cnn.com | 26.45 MB | 22.42 MB | -15.3% | Nav0 |
| twitch.tv | 3.78 MB | 3.01 MB | -20.5% | Nav0 |
Nav0 used less data on 11 of 15 pages. Chrome won on 4 — all by small margins except Craigslist, which we'll address in the methodology section.
The biggest wins were on heavy, ad-laden pages. BBC News: nav0 cut data consumption by 52.8%. CNN Lite: 53.6% less. Stack Overflow: 30.2% less. These are exactly the pages where third-party scripts and tracker payloads dominate the page weight.
On lightweight pages like Hacker News and NPR Text, both browsers transferred nearly identical amounts. That's expected — when there's no junk to block, both browsers do the same work. The gap only appears when sites start loading trackers, analytics, and ad scripts.
The Totals
| Metric | Chrome | Nav0 | Difference |
|---|---|---|---|
| Total received | 43.46 MB | 35.87 MB | -17.5% |
| Total sent | 575.30 KB | 359.13 KB | -37.6% |
| Total requests | 1,168 | 828 | -340 (29.1% fewer) |
Nav0 made 340 fewer HTTP requests across the same 15 pages. That's 340 fewer connections opened, 340 fewer DNS lookups, 340 fewer TLS handshakes. Each one costs time, battery, and bandwidth.
The sent data difference is even more telling. Nav0 sent 37.6% less data upstream — meaning it's not just downloading less junk, it's also sending less information about you back to third-party servers.
Where the Data Goes
The raw byte counts are interesting, but the breakdown by category is where it gets uncomfortable for Chrome.
Request Categories
| Category | Chrome Requests | Chrome Bytes | Nav0 Requests | Nav0 Bytes |
|---|---|---|---|---|
| First-party | 211 | 3.87 MB | 202 | 4.38 MB |
| Third-party | 935 | 39.59 MB | 612 | 31.49 MB |
| Tracker | 104 | 2.63 MB | 14 | 137.64 KB |
| Blocked | 0 | 0 B | 21 | 0 B |
| Blocked tracker | 0 | 0 B | 25 | 0 B |
Read that tracker row again. Chrome loaded 104 tracker requests totaling 2.63 MB. Nav0 loaded 14 tracker requests totaling 137.64 KB — and blocked 25 more outright.
Chrome blocked zero requests. Zero trackers. It loaded every single one.
Nav0's built-in tracker blocking isn't an afterthought bolted on via an extension. It's part of the browser. No extension overhead, no add-on to install, no configuration to fiddle with. It just works.
And the third-party numbers are striking too. Chrome made 935 third-party requests; nav0 made 612. That's 323 fewer requests to domains other than the site you're actually visiting. Every one of those requests is a potential tracking vector — a cookie being read, a fingerprint being built, a behavioral profile being updated.
Resource Types
| Type | Chrome Requests | Chrome Bytes | Nav0 Requests | Nav0 Bytes |
|---|---|---|---|---|
| Script | 415 | 15.66 MB | 331 | 12.26 MB |
| Image | 303 | 12.85 MB | 255 | 9.59 MB |
| Media | 2 | 10.80 MB | 6 | 10.83 MB |
| Document | 16 | 1,011.89 KB | 16 | 994.27 KB |
| Font | 24 | 1,004.28 KB | 23 | 975.89 KB |
| XHR | 157 | 756.93 KB | 47 | 147.08 KB |
| Stylesheet | 88 | 658.07 KB | 80 | 642.69 KB |
| Fetch | 87 | 303.09 KB | 34 | 126.30 KB |
| Ping | 21 | 8.19 KB | 9 | 3.74 KB |
The standout here is XHR. Chrome made 157 XHR requests totaling 756.93 KB. Nav0 made 47, totaling 147.08 KB. That's a 70% reduction in background data calls. Most of those XHR requests in Chrome are analytics, tracking beacons, and data synchronization calls that have nothing to do with rendering the page you're looking at.
Script and image data show meaningful reductions too — 3.4 MB less JavaScript and 3.3 MB fewer image bytes. A significant chunk of that is third-party ad and analytics scripts plus their associated tracking pixels.
Documents, fonts, and stylesheets — the things that actually make web pages work — were nearly identical between both browsers. Nav0 isn't breaking websites. It's cutting the fat.
The Idle Test
After loading all 15 pages, we let both browsers sit idle for 20 seconds with every tab still open. No scrolling, no clicking, no interaction. Just sitting there.
Chrome: 29 requests, 37.55 KB received, 11.14 KB sent.
Nav0: 0 requests, 0 bytes received, 0 bytes sent.
Zero. Nav0 produced absolutely no network traffic while idle.
Chrome, on the other hand, was busy. Here's a sample of what it was doing in the background:
- Pinging
chartbeat.netanalytics trackers - Making XHR calls to CNN's media streaming endpoints
- Fetching data from
ad-delivery.netandad.doubleclick.net - Sending data to
adsafeprotected.com - Calling
optimizely.comexperimentation endpoints - Hitting
scorecardresearch.comaudience measurement - Polling Twitch's GraphQL API
- Requesting ad resources from
fwmrm.net
This is with no user interaction. You're not browsing. You're not scrolling. You walked away from your computer, and Chrome is still sending and receiving data on your behalf — to ad networks, analytics companies, and data brokers.
This isn't a privacy policy debate. This is measurable network activity. 29 requests in 20 seconds, while you're doing nothing, to domains whose entire business model is tracking you.
Nav0 does nothing. Because there's nothing to do. The page is loaded. You're not interacting. So the browser sits quietly and waits.
On Honesty
We said we'd be honest, so let's talk about where Chrome won.
Chrome used less data on 4 of the 15 pages: Hacker News (+0.3%), Craigslist (+66.7%), GitHub Explore (+30.8%), and YouTube (+3.5%).
The Craigslist result is the most notable outlier. Nav0 transferred 310 KB more than Chrome on craigslist.org. We investigated this and believe it's related to Craigslist's redirect and bot-detection behavior — different browser signatures can trigger different response paths, and nav0's Electron-based user agent may have caused additional redirects or a heavier page variant. This is worth investigating further and we plan to.
The GitHub Explore difference (+30.8%) likely reflects similar server-side content negotiation or caching behavior differences between the two browsers.
On Hacker News and YouTube, the differences were negligible — 0.3% and 3.5% respectively.
The important thing: on the pages where Chrome "won," the margins were small in absolute terms (except Craigslist). On the pages where nav0 won, the savings were measured in megabytes.
What About First-Party Data?
You might have noticed that nav0 actually transferred more first-party data than Chrome: 4.38 MB vs 3.87 MB. That's a 13% increase.
We think this is a feature, not a bug. Nav0 blocks third-party tracker scripts that Chrome loads. Some of those scripts dynamically modify what first-party content gets loaded — for example, a tracker script might lazy-load certain images or defer content loading. When those scripts are blocked, the browser may load the full first-party content directly instead.
In other words: nav0 loads more of the actual content you came for and less of the surveillance infrastructure around it.
Testing Methodology
Hardware and Software
- Machine: MacBook Pro with Apple M1 chip, 8 CPU cores, 16 GB RAM
- OS: macOS (Darwin arm64)
- Chrome: Launched at
/Applications/Google Chrome.appwith default settings - Nav0: Launched at
/Applications/nav0-browser.appwith default settings
How We Measured
Both browsers were automated via the Chrome DevTools Protocol (CDP). We used CDP's Network.requestWillBeSent and Network.loadingFinished events to capture every network request, including:
- URL, domain, and resource type
- Bytes sent and received (wire-level
encodedDataLength, not decoded size) - Whether the request was first-party or third-party
- Whether the request was to a known tracker domain
- Whether the request was served from cache
- Whether the request was blocked
For each page, we navigated to the URL, waited for the network to settle, then collected all request data. After all 15 pages were loaded, we monitored idle traffic for 20 seconds.
The Test URLs
We used the same 15 real-world websites from our performance benchmark, spanning a range from minimal text sites to heavy media platforms:
- news.ycombinator.com
- text.npr.org
- en.wikipedia.org
- craigslist.org
- lite.cnn.com
- developer.mozilla.org
- docs.github.com
- stackoverflow.com
- npmjs.com
- github.com/explore
- reddit.com/r/programming
- youtube.com
- bbc.com/news
- edition.cnn.com
- twitch.tv/directory
What We Didn't Do
- Multiple runs. This was a single test run. Network responses vary between requests — CDN routing, A/B tests, and ad auction results all introduce variability. The trends are directionally clear, but individual page numbers should be taken as representative, not exact.
- Extension testing. Chrome was tested without extensions. A user running uBlock Origin would see numbers closer to nav0's. But that's the point — nav0's protections are built in. You shouldn't need to install and maintain a third-party extension to stop your browser from hemorrhaging data.
- Mobile testing. This was a desktop test. Data consumption on mobile, where bandwidth is metered and more expensive, is an even more important benchmark. That's coming.
- VPN or proxy analysis. We measured browser-level network activity. We didn't attempt to capture OS-level network traffic that might bypass the browser's network stack.
Why This Matters
Data consumption isn't just an abstract privacy concern. It has real, tangible consequences.
Bandwidth costs money. If you're on a metered connection — mobile hotspot, satellite internet, travel roaming — every megabyte counts. Chrome using 17.7% more data means 17.7% more of your data plan consumed by content you didn't ask for.
Data means battery drain. Every network request wakes the radio, processes a response, and consumes power. 340 fewer requests per session means measurably longer battery life on laptops and mobile devices.
Background traffic means background surveillance. Those 29 idle requests Chrome made in 20 seconds weren't loading content for you. They were reporting on you — your behavior, your viewing patterns, your engagement metrics — to companies whose business model is selling that information.
Request volume means attack surface. Every third-party request is a trust decision. You trust the site you're visiting, but do you trust the 935 third-party domains it connects to? Each one is a potential vector for malvertising, cryptojacking, or drive-by downloads. Fewer requests means fewer opportunities for things to go wrong.
The Bottom Line
Chrome transferred 44.02 MB across 15 websites. Nav0 transferred 36.22 MB. That's 7.8 MB of data that served no purpose other than tracking, profiling, and monetizing your browsing.
Chrome made 1,168 requests. Nav0 made 828. Those 340 extra requests weren't loading the news, rendering a video, or fetching a stylesheet. They were pinging ad servers, firing analytics beacons, and reporting your behavior to data brokers.
Chrome generated 29 background requests in 20 seconds of idle time. Nav0 generated zero.
These aren't opinions. They're measurements. The test script is open source in our repository — run it yourself.
Your browser should load the page you asked for and then stop. Nav0 does that. Chrome doesn't.
Download nav0 — free, open source, and light on your data.
nav0 is a minimal, open-source browser built on one principle: your browser should work for you, not for an ad company. Zero telemetry. Zero bloat. Zero BS. Learn more.