Type a web address, press enter, and a name like example.com has to become a numeric address before any page can load. That translation is DNS, the internet’s directory, and it runs quietly before every connection you make. Most of the time it finishes in a few thousandths of a second and you never think about it. When it goes wrong, though, sites feel slow or refuse to load for no obvious reason, even on a fast line. Following the lookup step by step shows where the time goes and why swapping one setting sometimes helps.
What a name points to
People remember words; the network routes numbers. A domain name is a label sitting on top of one or more IP addresses, the actual identifiers that packets travel toward. When you ask for a site, what you really need is its current address, and DNS is the lookup that provides it. A single name can map to several addresses to balance load, and to both older IPv4 numbers and newer IPv6 ones.
Names carry more than web addresses too. The same system holds records that direct email, verify ownership and point subdomains at different servers. For everyday browsing, though, the record that matters is the one linking the name you typed to the address your browser should open a connection to.
The chain that answers a lookup
When the answer is not already known, resolving a fresh name is a relay through several specialised servers, each responsible for one slice of the job. Your device asks a recursive resolver, usually run by your provider or a public service, and that resolver does the walking on your behalf.
- Your device asks its configured recursive resolver for the address of the name.
- If the resolver has no cached answer, it asks a root server, which points it toward the servers for the top-level domain, such as .com.
- The top-level domain server replies with the authoritative name server for that specific domain.
- The authoritative server returns the real IP address for the name.
- The resolver hands that address back to your device and stores it for next time.
Only the first fresh lookup runs the whole relay. It sounds long, yet it usually completes in tens of milliseconds, because the servers are fast and most of the chain is skipped on repeat visits.
Caching at every layer
Repeating that full journey for every request would be wasteful, so answers are stored at several points and reused until they expire. Your browser keeps a short-lived cache, the operating system keeps its own, and the recursive resolver holds a large shared one for all its users. Each cached record carries a time to live, a countdown chosen by whoever runs the domain, fixing how long an answer stays valid before it must be looked up again.
The effect is that the great majority of lookups never leave your resolver, and many never leave your own device. Caching is the main reason a well-run resolver feels instant, and how much a change to it really helps day to day is the question the guide on DNS and browsing speed examines.
Why a poor resolver feels slow
Every new domain your browser meets needs an address before it can fetch anything, and a modern page pulls images, fonts, scripts and trackers from many domains at once. Each unfamiliar one triggers a lookup. If your resolver is far away, overloaded, or often missing its cache, that small delay is added to the start of nearly everything, and it stacks up across all those separate requests.
The delay shows up as a pause before a site begins to appear, after which it may load quickly. That pattern, hesitation and then speed, is a classic sign of a sluggish resolver rather than a slow connection. It is also separate from the speed of your wireless link itself, the throughput the overview of what the wifi standards mean covers.
When lookups fail
Sometimes the problem is not slowness but a lookup that returns nothing or the wrong answer. If a resolver cannot reach the authoritative server, or holds a stale record after a site has moved, that one name fails while every site with a working cached answer loads normally. This is why some pages break while others on the same connection are fine; the fault sits in a single lookup, not the line.
Stale caching cuts both ways. After a website changes servers, a cached record with a long time to live can send you to the old address until it expires, which is why a page can work on one device and fail on another for a while. Clearing the local cache or waiting out the countdown usually resolves it.
Changing your resolver
You are not tied to the resolver your provider assigns. Pointing your device or router at a different recursive service can cut lookup times if the new one is closer or better run, and many offer extras such as category filtering or encrypted queries that hide your lookups from the network. The change affects only how names are resolved, not how much data your line can carry.
Setting the resolver on each device changes only that device; setting it once in the router hands the same choice to everything on the network, which is part of what the overview of the home router’s settings covers. After switching, load a few new sites and notice whether the initial pause shrinks, since that first delay is what a better resolver improves.
What to actually expect
A faster resolver sharpens the moment between clicking a link and the page starting to build, especially across many unfamiliar domains. It does nothing for the size of a download or the ceiling of your connection, so a speed test will read the same before and after. The two are different measurements, and mistaking one for the other leads to disappointment, the gap the guide on what real wifi speed means explains.
For most people the honest expectation is modest but real: a tidier resolver trims a little delay from the start of browsing and can add filtering or privacy along the way. Understanding the lookup, the chain, the caches and the point where it stalls turns a vague complaint that the internet feels slow into a specific thing you can test and, often, fix.
Frequently asked questions
What is DNS?
DNS, the domain name system, is the directory that turns readable names such as example.com into the numeric IP addresses computers use to reach each other. Every time you open a site, send an email or use an app, a DNS lookup quietly finds the right address first. It is distributed across many servers worldwide and leans heavily on caching, so most lookups finish in a few thousandths of a second without you noticing.
Does changing DNS speed up browsing?
It can help, but only in one specific way. A closer or better-run resolver returns addresses faster, which shortens the pause before a new site starts to load, most noticeably across pages that pull from many domains. It does not raise your download speed or bandwidth, so speed tests look unchanged. The benefit is largest if your current resolver is slow or unreliable, and small if it is already quick.
Why do some sites fail to load but not others?
Usually because a single DNS lookup is failing while the rest work. If your resolver cannot reach one site’s authoritative server, or holds an outdated record after that site moved, that name fails while every site with a valid cached answer loads normally. The connection itself is fine. Clearing your device’s DNS cache, or trying a different resolver, often fixes the one stubborn site without touching the others.
