Blog / How the internet works /

Traceroute: seeing the path your data takes

A website loads slowly for one customer and perfectly for you. Is it the server, their Wi-Fi, their internet provider, or something in between? “Have you tried turning it off and on?” won’t tell you.

Traceroute might. It’s a small tool already on your computer that shows the route your data takes to reach a destination, one stop at a time.

What traceroute does

When you load a site, your data doesn’t jump straight to the server. It passes through a chain of routers: your home router, your provider’s equipment, bigger networks, and finally the network the server sits in. Each stop is called a hop.

Traceroute lists those hops and shows how long each took to reply. Think of a parcel tracking page that shows every sorting depot, with the time at each one.

How it works, briefly

Every data packet has a “time to live” (TTL), a number that drops by one each time a router passes it on. When the number hits zero, the router throws the packet away and sends back a short “time exceeded” message.

Traceroute uses that. It sends a packet with TTL 1, which dies at the first router, which replies. Then TTL 2, which dies at the second router, and so on. Each reply reveals who that router is and how long the round trip took. Lined up, the replies draw the path.

How to run it

Windows: open Command Prompt and type:

tracert example.com

Mac or Linux: open Terminal and type:

traceroute example.com

Replace example.com with the site you’re checking. It takes anywhere from a few seconds to a minute.

If you can, run it on both sides: from your own computer to the server, and (where you can) from another network to your own connection. A single trace is one view of one route.

How to read the output

Each line is a hop. You’ll usually see:

  • A number, the hop count.
  • Up to three times in milliseconds, one for each probe.
  • A name or IP address for the router.

A made-up example, trimmed:

1  192.168.1.1     2 ms   2 ms   3 ms
2  10.20.0.1      18 ms  21 ms  19 ms
3  * * *
4  core-router.isp.example   45 ms  44 ms  46 ms
5  203.0.113.9    120 ms 118 ms 121 ms
6  server.example.com  122 ms 120 ms 123 ms

Hop 1 is almost always your own router. The next hops are usually your provider. The last line is the destination.

What to look for

A big jump that stays

If latency jumps from 20 ms to 120 ms at hop 5 and stays high for every hop after it, that link is slow. That might be a long physical distance (a hop crossing an ocean adds delay naturally) or a congested connection.

A slow hop that recovers

If one hop shows 300 ms but the next hop is back to 40 ms, ignore it. Routers treat replying to traceroute as a low-priority job. A slow reply from one router doesn’t mean your traffic was slow.

Stars (* * *)

Stars mean no reply came back. Often that’s harmless: many routers are set to ignore these probes for security. If the trace carries on after the stars, nothing’s wrong. If it ends in stars and never reaches the destination, the problem might be a firewall or a real break, and you need more evidence.

The trace stops at your provider

If it never leaves your provider’s network, you have something to show their support. Which brings me to the useful part.

Using it with support teams

Good support teams love evidence. When you contact your internet provider or hosting company, include:

  • The command you ran and the full output, pasted as text.
  • The time you ran it.
  • What you saw in the browser (slow, timing out, certain pages only).

It saves the first three messages of “please send more details”. I wrote more about asking for help well in hosting support: how to ask so you get help fast.

Why this matters if you run a site

If you serve readers far from your server, the physical path matters. More hops and more distance mean more delay, which is one reason to host near your audience or use a CDN. I covered the reasoning in choosing a hosting location for your readers and why the internet is slow sometimes: latency, bandwidth and routing.

Common myths

  • “Any slow hop is the culprit.” Not unless the slowness continues to the end.
  • “Stars mean it’s broken.” Often they just mean a router stays quiet.
  • “The path is always the same.” Routes change from minute to minute and differ between directions, so the way back can look different from the way there.

Better tools when you need them

Plain traceroute is a snapshot. Tools like mtr (Mac and Linux) or WinMTR (Windows) run the trace repeatedly and show packet loss per hop over time, which is far better for intermittent problems. Start with traceroute, and reach for these if the issue comes and goes.

Key takeaways

  • Traceroute lists each router between you and a server, with timings.
  • Look for latency that jumps and stays high, not isolated slow hops.
  • Stars are often harmless.
  • Paste the full output when you contact support.

FAQ

Is traceroute safe?

Yes. It sends a few tiny probe packets. Don’t run it against sites you don’t own repeatedly, though, as heavy probing can look rude.

Why does traceroute differ from ping?

Ping measures the round trip to one destination. Traceroute shows the steps along the way.

Try it now on a site you visit every day and count the hops. How many did you expect?