How HTTP Works - Technical Support

Module numbers

Where this page says "module 05" or "module 09", it means that module of the Rank Math Crash Course, where each setting is made.

1The trip a page takes

The short version

HTTP is the set of rules a browser and a web server use to talk. The browser sends a request for an address. The server sends back a response: a status code, some headers, and the page. Google's crawler does exactly the same thing a browser does, so everything Google knows about a page came through one of these responses. The SEO plugin's job is to decide what goes in that response.

The six steps. The whole trip usually takes less than a second, and each step has a place where an SEO decision lands.

2Step 1. The address

An address, properly a URL, has parts, and the SEO courses treat each part differently.

The four parts of an address. The host and the path are the two the courses spend time on.

Scheme. https means the connection is encrypted. Google expects it, browsers warn without it, and the launch checklist confirms every page has it.
Host. Ours is www.hermeticnetworks.com. The version without www is a different host as far as a browser is concerned, so one of them has to redirect to the other. Today the site's canonical tag says www is the real one. Module 05 and the launch checklist keep it that way.
Path. In WordPress this is the slug. It is what the page map in module 02 decides, what the redirect map in module 09 changes, and what the "keyword in URL" test in module 07 reads.
Query. Anything after a question mark. Tracking codes from an email or an ad live here. Google treats the page as the same page with or without them, which is what the canonical tag is for.

3Step 2. DNS, and step 3, HTTPS

DNS (the domain name system) is the lookup that turns a name into the numeric address of a server. The browser asks, a DNS server answers, and the browser now knows where to send the request. Two SEO tasks touch DNS: verifying the domain with Google Search Console (module 06 recommends the DNS record method, which proves we own the whole domain, www and not), and any change of host, which is a DNS change first and a redirect second.

HTTPS is HTTP over an encrypted connection. Before the first request, the browser and the server agree on a key, and the server proves who it is with a certificate. The padlock in the address bar is that certificate. For SEO the rule is simple: every page on https, every http address redirecting to its https twin, and a certificate that has not expired. The launch checklist in module 09 has all three.

Cloudflare sits in front

Our site is behind Cloudflare. That means the DNS answer points at Cloudflare, Cloudflare receives every request first, and it decides whether to answer from its own copy, pass the request on to our WordPress server, or challenge the visitor. It is why the crawl for these courses could read the homepage and the sitemaps but was blocked from the other pages. It matters in three places: Cloudflare can serve a cached copy of a page after you changed it, Cloudflare can hold redirect rules of its own, and Cloudflare's bot settings decide whether Googlebot gets through. When a change does not show up, purge the Cloudflare cache before you doubt the plugin.

4Step 4. The request

A request is a few lines of text. The first line says what the browser wants; the rest, called headers, say who is asking and what they can accept.

A request for one of our pages, simplified
Line	What it means	Why it matters here
`GET /managed-services/ HTTP/1.1`	Give me this path. GET means read; POST means send data, as a form does.	Googlebot only ever GETs. A page that needs a form submission to appear is invisible to it.
`Host: www.hermeticnetworks.com`	The host part of the address, sent separately because one server can hold many sites.	This is how the server knows to redirect the non-www host to www.
`User-Agent: Mozilla/5.0 ... Chrome/128`	What kind of browser is asking. Googlebot announces itself as Googlebot.	Cloudflare and some plugins treat known crawlers differently. robots.txt rules are written per user agent, which is why ours lists Googlebot, Bingbot, GPTBot and the rest by name.
`Accept-Language: en-US`	Preferred language.	No effect on our single-language site.

Two special requests happen before the first page is crawled. Googlebot asks for /robots.txt to learn which paths it may fetch, and it asks for /sitemap_index.xml to learn which pages we want it to fetch. Both are plain files served by the same rules as any page, and both are configured in the plugin (modules 04 and 06).

5Step 5. The server builds the page

Our server does not have a file called managed-services. WordPress builds the page on request: it looks up the post, runs the theme (Beaver Builder), and asks every active plugin whether it wants to add anything. This is the moment the SEO plugin works. It writes the lines into the page's <head> that Google reads before the content, and it can change the response itself.

What the plugin puts in the head

<title>: the title link Google starts from
<meta name="description">: the snippet Google starts from
<link rel="canonical">: which address is the real one for this page
<meta name="robots">: index or noindex, follow or nofollow
Open Graph and X tags: how the page looks when shared
JSON-LD: the structured data (Organization, WebPage, BreadcrumbList, FAQ, LocalBusiness)

What the plugin can change about the response

The status code: a redirect answers 301 instead of building a page at all
The Location header: where the redirect goes
Which addresses get a page at all: attachment pages, author archives and date archives can be turned off so they return a redirect or a 404 instead
The contents of /robots.txt and /sitemap_index.xml

Redirects can live at three layers, and the courses mention all three. Cloudflare can redirect before the request reaches us. The web server can redirect from its own rules, the .htaccess file on Apache. And WordPress can redirect from PHP, which is what Rank Math's Redirections and Yoast Premium's default method do. The plugin layer is the easiest to manage and the slowest by a few milliseconds; Yoast's "Web server" method and Rank Math's .htaccess export exist to move the rules down a layer once they are settled. Whichever layer answers, the browser sees the same 301.

6Step 6. The response

A 200 response for a service page. Google reads the status line, then the headers, then the head of the HTML, then the content.

Three things in the response are worth knowing by name, because Search Console reports on them and the courses set them.

The status line carries the code. The status codes section below is entirely about it.
The canonical tag says "of all the addresses that show this page, this is the one to index". It is how the query string, the non-www host and any accidental duplicate all resolve to one address. Both plugins write it automatically and let you override it per page.
The robots meta tag says index or noindex and follow or nofollow. It is the per-page switch behind module 02's noindex decisions. The same instruction can also travel as a header called X-Robots-Tag, which is how it is done for files that have no HTML head, such as PDFs.

Cached responses

Cloudflare, the host and the browser can all keep a copy of a response and serve it again without asking WordPress. That is good for speed and bad for the ten minutes after you change a title. If Search Console's URL Inspection shows the old title, fetch the live page in Inspection, purge the Cloudflare cache, and look again before changing anything in the plugin.

7Seeing it yourself

Open www.hermeticnetworks.com in Chrome or Edge. Press F12, open the Network tab, and reload. The first row is the HTML document. Click it.
Under Headers, read the Request URL, the Status Code, and the Response Headers. Find server: cloudflare. That is step 3's aside made visible.
Type hermeticnetworks.com without the www and reload. Watch the first row come back as a 301 with a Location header, followed by a 200 for the www address. That is the canonical host redirect.
Right-click the page and choose View page source. Find <title>, <meta name="description", rel="canonical" and the application/ld+json block. Everything you find between <head> and </head> that mentions Yoast is what module 05 will move into whichever plugin we keep.
Open /robots.txt and /sitemap_index.xml the same way. Both are ordinary responses with a 200 and a body.

8Status codes: what the first line says

The short version

A status code is the server's one-line answer to "can I have this page?" 200 means here it is. 301 means it moved, permanently, go here instead. 404 means there is nothing at that address. 410 means there was something here and it is gone for good. The plugin settings called "redirects" are you choosing which of those answers the server gives for an old address, and the right choice is what keeps the Google ranking a page earned when its address changes.

The three you set

301 for almost everything: merged duplicates, renamed slugs, the non-www to www hop, http to https.
410 for a page you are deleting with no replacement. A 404 does the same job more slowly.
302 only for something genuinely temporary. If you never use it, nothing is lost.

The three you only read

200 is what every page you want found should return. You check it; you never set it.
404 is the default when nothing is there. You watch for it in Search Console and the 404 log, then decide whether a 301 or a 410 is the fix.
307 and 451 are in the dropdowns because the plugins list them. No use at Hermetic.

Delete the old page

A 301 is a rule about the address, so once the rule exists the old page has no job. Merge any words worth keeping into the survivor, delete or trash the old page, and add the redirect. If the old page stays published, both addresses return 200 and Google sees two pages again. Keep the redirect rule itself indefinitely; old links and bookmarks can send traffic for years, and the rule costs nothing.

When you type an address, your browser sends a request to our web server. The server sends back a response, and the first thing in that response is a status code: three digits that say what kind of answer this is. Then, if there is a page, the page follows. You never see the code in normal browsing because the browser acts on it for you. Google's crawler sees every one of them and decides what to do with the address based on the code alone.

One request, one response. The status code is the first line of the answer, and for a redirect the new address is the second.

Codes are grouped by their first digit. 2xx means success. 3xx means "go somewhere else". 4xx means "your request cannot be answered", usually because nothing is there. 5xx means the server itself failed. Only seven individual codes matter for these courses.

9The seven codes that come up

What each code says, and what Google does with it
Code	Name	What the server is saying	What Google does
200	OK	Here is the page.	Crawls it and can index it. The normal state of every page you want found.
301	Moved Permanently	This address is retired. The page lives at the new address for good.	Follows to the new address, indexes that one, and over time transfers the old address's ranking signals to it. The code for almost every redirect in the page map.
302	Found (temporary)	The page is somewhere else for now, but this address will come back.	Follows to the new address but keeps the old one indexed, because you said it is coming back. Use only when that is true.
307	Temporary Redirect	The modern version of 302, with a stricter rule about resending the request exactly.	Treated like a 302. Rank Math and Yoast both offer it; you will rarely need it over 302.
404	Not Found	There is nothing at this address.	Drops the address from the index after repeated visits. The default answer for any address that never existed or was deleted without a redirect.
410	Gone	There was a page here. It was removed on purpose and is not coming back.	Drops the address from the index, and can do so a little faster than a 404 because you have said it was deliberate. Use for pages you are deleting with no replacement.
451	Unavailable For Legal Reasons	The page exists but cannot be shown because of a legal demand.	Treats it as unavailable. Listed in both plugins; Hermetic has no use for it today.

Two more you will meet outside the plugin. 500 means the server crashed while building the page, and 503 means the site is deliberately down for maintenance. Both are your host's business rather than an SEO setting, but if Search Console reports them, the pages are not being read.

Index and noindex are separate

"Noindex" is a line inside a page that says "you may read this, but do not put it in search results". The page still returns 200. It is how the thank-you and appointment pages are handled. A redirect or a 410 is for an address you are removing; noindex is for a page that stays but should not be found. The page map in module 02 uses both, for different pages.

10Choosing the code

The decision is one question: after the change, does the searcher have somewhere better to go? If yes, redirect them there with a 301. If no, tell Google the page is gone with a 410. Everything else is a special case.

Three questions get you to the right code. The left branch is the redirect map from module 02; the right branch is the tourism pages decision in module 10.

Worked on our own pages

Real addresses from the crawl of www.hermeticnetworks.com on 2026-09-29
Address today	What happens to it	Code	Why
`/backup-disaster-recovery-old/`	Merged into the surviving backup page	301	A searcher who lands on the old address wants the backup page. Send them there and give the survivor the old page's standing.
`/managed-it-support-richmond-va/`	Merged into the one managed IT page	301	Same search, one survivor. Every duplicate in the page map takes this code.
`/phising-protection-service/`	Renamed to fix the spelling	301	The page stays; only the address changes. Both plugins can create this one automatically when you change the slug.
`/metro-richmond-zoo/`	Deleted, if that is the module 10 decision	410	Nobody searching for the zoo wants an IT company, so there is no better page to send them to. Tell Google it is gone.
`/thank-you/`	Stays, hidden from search	200 + noindex	The page has a job after a form is sent. It is a noindex decision, and no code changes.
A page taken down for a week while it is rewritten	Points at the closest service page meanwhile	302	It is coming back at the same address, so you do not want Google to move its standing elsewhere.

Redirect chains

A chain is A redirecting to B, which redirects to C. Each hop is another request, and Google follows only a limited number before giving up. When you retire a page that was already the target of an older redirect, point the old redirect straight at the new destination. Module 09 has the rule and the diagram.

11Seeing the code a page returns

In a browser. Open the page, press F12 for the developer tools, open the Network tab, and reload. The first row is the page itself; its Status column shows the code. A redirect shows as a 301 or 302 row followed by a 200 row for the destination.
In Google Search Console. URL Inspection on any address reports how Google last fetched it and what code it got. The Pages report lists every address that returned a 404, a redirect or a server error, grouped by reason. Module 06 connects it; module 09 reads it.
In the plugin. Rank Math's 404 Monitor logs every address that returned a 404, and its Redirections screen lists every redirect and its code. Yoast Premium's Redirects screen does the same for redirects; Yoast has no 404 log.

12Glossary

HTTP: The rules a browser and a web server use to ask for and deliver pages. HTTPS is the same over an encrypted connection.
URL: A web address: scheme, host, path and query.
Host: The server name part of the address. www.hermeticnetworks.com and hermeticnetworks.com are two hosts; one redirects to the other.
Path, slug: The part after the host that names the page. In WordPress, the slug.
Query string: Everything after a question mark. Usually tracking. The canonical tag keeps it from creating duplicates.
DNS: The lookup that turns a host name into a server's numeric address. Also where domain ownership is proved to Google.
Certificate: The file that proves a server is who it says it is and makes HTTPS possible. The padlock.
Request: What the browser sends: a method (GET), a path, and headers.
Response: What the server sends back: a status line, headers, and usually a body.
Header: A named fact about a request or response, such as Location or Content-Type.
Head (of the HTML): The part of the page before the visible content, where title, description, canonical, robots and structured data live. The SEO plugin writes it.
Canonical: The tag naming the one true address for a page.
Robots meta tag: The per-page index/noindex and follow/nofollow instruction. Also possible as the X-Robots-Tag header.
Cache: A saved copy of a response served instead of asking the server again. Cloudflare, the host and the browser each keep one.
Cloudflare: The service in front of our server that handles DNS, caching, bot checks and its own redirect rules.
Googlebot: Google's crawler. Makes the same requests a browser does and reads the same responses.
Status code: The three-digit number a web server sends first in every response. Says what kind of answer this is.
200 OK: Here is the page. The normal state of every page you want found.
301 Moved Permanently: The address has moved for good. Google follows and transfers the old address's standing. The code for the redirect map.
302 Found, 307 Temporary Redirect: The address is elsewhere for now and will come back. Google follows but keeps the old address.
404 Not Found: Nothing at this address. Google drops it over time.
410 Gone: There was a page and it was removed on purpose. Google drops it, and can do so faster than a 404.
451 Unavailable For Legal Reasons: Blocked by a legal demand. Offered by both plugins; no use at Hermetic.
Redirect: A response that sends the visitor to a different address. 301, 302 and 307 are all redirects.
Redirect chain: A redirect whose destination is itself a redirect. Point old redirects straight at the final page.
Noindex: A line inside a page telling search engines not to list it. The page still returns 200. A different tool from a redirect.
Soft 404: A page that returns 200 but says "not found" in its content. Google flags these; return a real 404 or 410 instead.

13Sources

MDN Web Docs, An overview of HTTP and HTTP messages: requests, responses, headers.
Google Search Central, How Google Search works and Robots meta tag and X-Robots-Tag.
Google Search Central, Consolidate duplicate URLs with canonicals.
Cloudflare, How Cloudflare works: the proxy in front of the origin server.
Google Search Central, How HTTP status codes, and network and DNS errors affect Google Search: what Google does with each code.
Google Search Central, Redirects and Google Search: permanent and temporary redirects, redirect chains.
MDN Web Docs, HTTP response status codes: the codes and their names.
Rank Math, Setting up redirections, and Yoast, The redirect manager in Yoast SEO Premium: the codes each plugin offers.
Site facts from a crawl of www.hermeticnetworks.com on 2026-09-29: the www canonical, the Cloudflare response, robots.txt, the sitemaps and the addresses in the worked examples.

Documentation read 2026-09-30.