Site settings

Settings is split in two. Settings itself is account-wide: Domains holds everything about one site and Integrations holds Search Console. A site's settings sit behind Domains, in five tabs: General, Hreflang, Canonical, Export and Danger Zone. Each of the first three has its own Save button.

The summary row on the account page counts Active sites against your plan's limit, and API keys, which always equals the number of sites because every site gets a key when it is created.

General

The domain name is read-only. Default language is used as the fallback language for breadcrumb structured data when a page's own language is not known. Auto breadcrumbs, on unless you turn it off, adds a BreadcrumbList to every page's structured data from the breadcrumbs the crawl found.

The API key field shows only the first characters. Regenerate shows the new key once, in full, and it disappears when you save or leave the page; the key is otherwise shown on the Integrate page, where the plugin guide needs it. Regenerating stops every installation for the site until each one has the new key, and there is no grace period, so update the installations right after, not before you have the new key.

Hreflang

Take over hreflang decides whether we serve anything at all. Off, the site's own alternate tags stay as they are; we still verify them and report what we find. On, we serve our own verified set instead of the site's tags, and only for pages where the pairing has been proven.

Strategy says how the site's languages are laid out. Subdirectory, /sv/ and /en/, is the supported layout: each locale row names a path prefix, and the crawl's own URLs are paired by that prefix.

The Locales list needs exactly one row marked as the default and at most one marked x-default. A locale code is sv or sv-SE; the same code twice, or a row without a prefix, blocks Save with a message under the row. The tab opens with a Swedish and an English row filled in as an example; nothing is saved until you press Save.

Pairing never invents a URL. A page whose translation has a different slug does not get an alternate pointing at a guessed address; it gets none. Only pages the crawl actually found are paired, and a pair needs two languages.

Canonical

The Canonical tab records the site's canonical policy: which query parameters to strip, utm_*, gclid and fbclid by default, which to keep, whether to lowercase parameters, what to do with trailing slashes, and how paginated pages should point. The Path defaults table can record the same per path prefix.

The canonical that is served on a page is the one from the category's template on the Content page, or the page's own if the template leaves it empty.

Export and import

Export downloads every generated SEO field for every page of the site, as CSV or JSON: title, description, H1, keywords, canonical, robots, the social fields, H2s and image alts, with the category and the source each value came from. The CSV opens correctly in Excel.

Import takes the same file back, or any CSV or JSON with a url column. Existing pages are updated and new ones added; nothing is deleted. In a CSV, a blank cell means keep the current value. A large file is written in batches, and if the import is interrupted, running it again is safe: rows are updated, not duplicated.

Imported values become the pages' own values, which the category template then leaves alone. Templates and publishing explains how to hand a field back to the template.

Deleting a site

Danger Zone removes the site from your account: its categories, templates, variables, rules, structured-data selections and published content. The crawl data for the hostname is kept. A mapped Search Console property stays in your account, unmapped.

Adding a site

Enter the hostname; https://, www. and anything after the first slash are stripped for you. A hostname we have crawled before comes with data at once. Adding a new hostname starts sitemap discovery, and if the sitemap holds 5,000 URLs or fewer, a first crawl follows on its own. A larger site waits to be mapped; Crawling and categories explains why.

You can add sites up to your plan's limit.

Search Console

Connect Google account sends you to Google to grant read access to Search Console; more than one account can be connected. On return, the account's properties are listed already. Map one property to one site: the mapping is one to one, so a site already mapped to another property has to be unmapped there first. A domain property covers every subdomain; a URL-prefix property covers exactly one prefix, so if the numbers look low, check that the mapped property covers the URLs you expect.

Data fills in over the next day rather than at once, starting with 90 days of history, and it always ends three days ago because Google's own numbers lag that far. Sync properties is only needed after adding a property in Search Console; it is disabled while a connection needs re-authentication. Disconnect keeps the mappings and the history, so reconnecting the same account later picks up where it left off.

If the connection comes back with an error about a missing scope, the "See Search Console data" permission was unticked on Google's consent screen; connect again and leave it on.