Meta tags and structured data, without a developer

Most of what a small business needs from technical SEO is a handful of tags in the head of a page and a block of structured data telling a search engine what the business is. Neither requires a developer. Both are routinely wrong.

Title and description lengths

Roughly 60 characters for a title, 155 for a description. Treat those as a working range, not a rule — search engines truncate on rendered pixel width, so a title in capitals runs out of room well before the same character count in lower case.

The description does not affect ranking directly. It affects whether anyone clicks, which affects everything downstream. Write it as the sentence that makes someone choose your result over the one above it.

What JSON-LD is actually for

Structured data tells a search engine what a page is about in a form it does not have to infer: this is a business, with this name, this phone number, this address, these opening hours. Organization, LocalBusiness, Restaurant and ProfessionalService cover most small businesses.

There is one hard rule. Structured data must match what a visitor can read on the page. Claiming a review score, a price or an FAQ answer that is not visible is what search engines classify as spam, and the penalty is not worth the rich result.

If the JSON-LD and the page disagree, the JSON-LD is the lie — and it is the one with your name on it.

Half an address is worse than no address

A country on its own is not an address. A PostalAddress node with a country and nothing else tells a search engine you have structured data and no information, which is worse than being quiet. A generator worth using refuses to emit it until there is at least a street, city, region or postcode.

The four small checks

• UTM parameters — without them, newsletter traffic and paid traffic look identical in analytics and you cannot tell which worked.
• Slugs — titles with accents, ampersands or non-Latin characters break in ways you only notice after the link is shared.
• robots.txt — one wrong Disallow line takes a section of the site out of search, silently.
• Sitemaps — a sitemap promising URLs that no longer exist wastes the crawl and undermines the rest of it.

Redirects that fail silently

Two redirect mistakes never announce themselves: a rule that redirects a path to itself, which loops, and duplicate sources, where the second rule never runs. Both are trivial to generate by accident and painful to diagnose in production, so the generator refuses to emit either — for Apache, nginx and Netlify alike.

Nothing here fetches your site

These tools generate from what you type. No URL is fetched, no URL is logged. If you want a crawl of your live site, that is a different kind of tool and it should tell you plainly that it is one.