Five situations, five workflows

The pipeline is the same everywhere. What changes is the volume, the quality of the titles, and how much review the result deserves.

A content blog

The situation. 200+ articles. Focus keywords exist on some posts and not others. Internal linking is manual and inconsistent; most posts are only reachable from category archives.

Step 1, fill the gaps. Run Auto Focus Keyword on posts and pages. It identifies the 60 to 70 % of articles with no focus keyword and populates them from titles. Then review the twenty most-visited posts and adjust by hand where the title is a poor target.

Step 2, build the network. Run Automatic Internal Links. Articles that share keyword themes become connected through contextual anchors in the body text, not just through navigation.

What to expect. Orphan pages drop. Average depth decreases. Related content becomes discoverable inline. Older evergreen articles start receiving links again, from both readers and crawlers.

A WooCommerce catalogue

The situation. 500+ products. No focus keywords assigned. Internal linking between products and editorial content is minimal or absent.

Step 1, bulk assignment. Product support requires Pro on both plugins. Select Posts, Pages and Products, then FETCH and SYNC. Every product now carries its product name as focus keyword.

This is where automatic assignment is at its strongest: a product name is specific by nature, and it is what customers actually search.

Step 2, link creation. Configure Automatic Internal Links for all post types and run SYNC. Blog posts that mention product names link to the product pages, and related products link to each other.

What to expect. Product pages receive links from editorial content, which is usually where their authority was missing. Articles gain navigation paths towards products. The catalogue becomes connected through keyword-based anchors rather than category pages alone.

An agency with ten client sites

The situation. Different content volumes, different SEO plugins, inconsistent keyword coverage. Manual internal linking at that scale is not a plan.

The workflow, per site:

1. Audit. Install Auto Focus Keyword and run FETCH without SYNC. The count of items with no focus keyword is an instant coverage metric, and it costs nothing to produce.
2. Populate. Run SYNC to fill the gaps, then review the client's top pages.
3. Link. Install Automatic Internal Links and run SYNC to build the network.
4. Maintain. Enable AUTO LINKS in Pro for clients who keep publishing.

Why it scales. Both plugins are install-and-run, with no per-page configuration in the bulk workflow. What would be hours of manual work per site becomes a short, repeatable pass.

The caveat that matters for agencies. Repeatable does not mean unsupervised. Step 1 writes a target onto every page, and a bad target propagates into every anchor built from it. Budget the review, especially on the pages the client cares about.

A multilingual site

The situation. One WordPress install serving two or more languages through WPML or Polylang. Each language has its own pages, and the keyword fields are uneven across them.

What to establish first. From version 2.0.8, the plugin filters out links that would cross a language boundary, when it can determine the language of both pages. It does so through the filter WPML provides. Confirm on your own stack whether that filter is available, because the rest of the workflow depends on the answer. Polylang has not been tested.

The workflow when the filter is active. Run the pipeline normally. Spot-check a word your languages spell identically to confirm no link crosses. Remember the check is fail-open: it allows a link whenever either side's language cannot be resolved, so a review pass is still worth its time.

The workflow when it is not. Run one language at a time. After each SYNC, read the activity log and delete the rules that cross a boundary. Add the words your languages share to the keyword exclusion list, and use per-page exclusion on the pages where a wrong-language link would be most costly.

What to expect. Within a language, the result is the same as on any other site. Across languages, expect to spend review time proportional to how much vocabulary your languages have in common. Two distant languages need almost none; French and English need a real pass.

Full details in multilingual sites.

A site with weak SEO hygiene

The situation. Keyword fields filled inconsistently, internal links added by whoever happened to think of it, no convention anywhere.

This is the case where the two plugins are worth more together than separately, because the problem is not a missing feature but a missing baseline. Step 1 establishes that every page has a declared target; step 2 establishes that every page participates in the link graph. Neither is clever on its own. The value is that they are applied uniformly, which is exactly what a site in this state has never had.

The caveat. A uniform baseline built from titles is better than nothing and worse than editorial judgement. Treat the first run as a floor to improve on, not a finished state.

Read next

• The three steps — what each step produces
• The pipeline — what changes, and why it matters
• Best practices — the hybrid automatic and manual strategy
• Requirements and compatibility — what has to be true before any of this