How an SSL/TLS certificate works
A certificate is a small signed file issued to a specific hostname. It contains a public key, the names it is valid for, the dates it is valid between, and a signature from a certificate authority that browsers already trust. When a visitor arrives, the browser checks three things: that the name on the certificate matches the address being requested, that the signature traces back through any intermediate certificates to a root the browser trusts, and that today falls inside the valid dates. If all three hold, the two sides agree on keys and everything after that is encrypted.
The naming is confusing and worth clearing up. SSL was the original protocol and has been retired for years; what actually runs today is TLS. The industry kept saying “SSL certificate” out of habit, so the two names mean the same thing in ordinary conversation.
Certificates differ mainly in what they cover and how much the authority checked before issuing. A domain validated certificate confirms only that the applicant controls the domain, which is what most sites use. Organisation and extended validation certificates involve checks on the business itself, though browsers no longer display that distinction in a way visitors would notice. A wildcard covers a level of subdomains beneath a name, and a multi-domain certificate lists several hostnames at once.
Why the certificate matters
Without one, a site cannot serve HTTPS at all, so everything downstream depends on it: the padlock, the absence of a browser warning on your enquiry form, the newer and faster protocol versions, referrer data surviving in analytics.
It is equally important to understand what it does not prove. A certificate says the connection to that hostname is private and unaltered. It says nothing about whether the business behind the site is honest — a fraudulent site can hold a perfectly valid certificate, and many do. Do not present the padlock to a client as evidence of trustworthiness; it is evidence of a secure pipe, not a good company.
Where certificates go wrong
Expiry causes more outages than any other certificate fault, and it is entirely avoidable. Automated renewal quietly stops working after a hosting change, a control panel migration or a DNS move, and nobody notices until visitors meet a full-page warning that most will not click through.
Name mismatch is the next most common. A certificate issued for the www form while visitors arrive at the bare domain produces a warning on an address that works perfectly well otherwise. Then there are incomplete chains: the server sends its own certificate but not the intermediate one, so it looks fine in a modern desktop browser while failing on older phones, on some crawlers and in command-line tools. Finally, a self-signed certificate meant for a staging environment sometimes follows the site into production.
What to do about it
Cover every hostname visitors might use, including both the www and bare forms, and automate renewal. Monitor expiry with something independent of the host, so a broken renewal is caught by you rather than by a customer. Test with an external checker rather than your own browser, because your browser may have cached an intermediate certificate that other visitors will not have.
Once the certificate is solid, work through the rest in order: redirect every plain address as part of moving from HTTP to HTTPS, clear any mixed content, and only then consider HSTS. Doing them out of order is how a site locks visitors out of an address it can no longer serve.
