How the trailing slash works
A trailing slash is the mark at the end of a URL path: the difference between an address ending in the word contact and one ending in contact followed by a slash. To a server and to a search engine those are two different addresses. Whether they both return a page, and whether they return the same page, depends entirely on how the site is configured.
The convention came from the days when a path with a slash meant a directory and a path without one meant a file. Modern platforms have made it a style choice instead. WordPress adds the slash when the permalink structure ends in one; several static hosts and JavaScript frameworks default to leaving it off. Neither is better for rankings. The only thing that matters is that a site picks one and never contradicts itself.
One address always keeps its slash: the home page. A bare domain resolves to a single slash whatever style the rest of the site uses, which is why home page URLs look inconsistent with the pattern below them.
Why the trailing slash matters
If both forms return a page with a success status, the site has two copies of everything. Links and shares land on whichever version somebody happened to copy, so the signals split. Analytics reports the same page twice, once in each form, and every report that ranks pages by traffic understates both.
There is a subtler effect worth knowing. Relative links are resolved against the current address, and the slash changes what the browser treats as the current folder. A relative link written on a page without a slash can resolve to a different destination than the identical link on the same page with one, which is how sites end up with images that load in one form of a URL and break in the other.
Where the trailing slash goes wrong
The common failure is a redesign that quietly flips the default. Every existing address changes form, nothing redirects, and the site starts again with new URLs while the old ones return errors. The second is applying the rule too broadly — a redirect that adds a slash to everything will happily do it to files with extensions, breaking a sitemap or a feed.
The third is stacking rules until they chain. A request that is redirected from plain to secure, then from one hostname to another, then to add a slash, has made several hops before anything is served. Each hop costs time and slightly weakens the link.
Getting it right
Choose the form the site already uses and the one most of your existing links point at, then enforce it once, at the server, in a single hop. Make everything else agree: canonical tags, the XML sitemap, internal links in templates and content, hreflang if you have it, and the final URLs in your ad accounts.
Verify with a crawl rather than by eye. Request a handful of pages in the wrong form and confirm each one redirects straight to the right form with no intermediate stop. Then check your analytics for pages appearing in both shapes, which is the quickest sign that something is still generating the wrong style. It is a small piece of technical SEO that sits next to canonical tags and avoiding a redirect chain.