How to Plan Internal Links Before You Publish

A link map is a small table, not a diagram. It holds one row per planned link. Four columns are enough:

Keep the row count low. If a page needs more than five planned 1 rows for outgoing links, the page is probably doing two jobs and should be split.

Two habits make the map useful rather than decorative. First, write the anchor idea at the same time as the target, because Google asks for anchor text that is descriptive and tells people and search engines something about the linked page per the link best practices page. Second, store the map as plain text (Markdown or CSV) in the same repository or folder as the draft. A map in a chat thread disappears.

Here is a minimal map file you can copy:

| page | target | anchor idea | priority |
| --- | --- | --- | --- |
| /guides/invoice-reminders | /pricing | see pricing for volume tiers | 1 |
| /guides/invoice-reminders | /guides/payment-terms | how payment terms affect reminders | 1 |
| /guides/invoice-reminders | /blog/collections-playbook | the collections playbook | 2 |

At publish time, the map doubles as a checklist. Every priority 1 row either exists on the page or has a written reason why it was dropped.

A budget converts a vague preference ("do not over-link") into a number you can check. A workable starting point for body content:

• Out: a maximum of five contextual links in the body, of which at most one points to a commercial page such as pricing or a product page.
• In: at least one link from an existing page at publish, rising to about three links in from related pages within a reasonable window.
• Structural links (menu, footer, breadcrumbs) are outside the budget; they carry navigation rather than editorial intent as Yoast's internal linking guide sets out.

These numbers are conventions, not rules from Google. Their value is that they force a decision. When a sixth candidate link appears, the writer must either drop it or demote an existing row to priority 2.

The one-link-per-commercial-page rule matters more than the total. Pages that link to pricing from every paragraph dilute the signal and annoy readers. If a guide genuinely needs three references to pricing, link one and mention the other two in text.

Budgeting also protects the reader. Google's people-first guidance asks whether a reader leaves a page feeling they learned enough to achieve their goal per the helpful content documentation. A page that sends the reader away every other sentence rarely passes that test.

The hardest part of planning links is not the mechanics but the editorial call. Use this three-way test on each candidate:

1. Link when the reader would plausibly stop reading this page to get the detail on another page, and the link target answers that need completely.
2. Mention without a link when the other page is background context the reader does not need mid-sentence, or when the target page does not yet exist.
3. Skip when the link would only exist to pass signals, not to help a reader.

A worked example makes the arithmetic concrete. Suppose you publish 20 pages in a quarter and plan five outbound body links per page. That is 100 link rows to review, which sounds heavy until you split it: roughly half are repeats pointing at a small set of core pages (home, pricing, one or two pillar guides). In practice you are deciding links to about 10 core targets, not 100 unique decisions.

The same arithmetic exposes hubs. If one page is the target of 40 of those 100 rows while the other pages get one or two each, you have built a hub by accident. Spread the priority 1 rows so that no single page receives more than a modest share of inbound body links; that keeps a page from soaking up all the internal authority you are handing out as Backlinko's internal linking guide notes about link equity.

One more decision rule on anchors: vary them. Repeated identical anchor text across many pages looks mechanical. Google's guidance asks for descriptive anchor text, and Semrush's internal links guide describes how anchor text and surrounding context combine to signal what the target page covers in Semrush's internal link guide. Descriptive and varied beats keyword-exact and repeated.

Run four checks on the map, in this order, every time a page ships.

Orphans. Filter the map by target and confirm the new page appears as a target at least once. If it does not, add one priority 1 inbound row from the most closely related existing page before you publish. Orphan pages are the most common and most avoidable internal linking failure, and Semrush's guide lists them among the problems a deliberate structure prevents per Semrush's internal links guide.

Over-linked hubs. Count rows per source page. Any page above the budget gets a trim. Then count rows per target page. A target receiving an outsized share of the site's body links is a candidate for demotion: move some of those links to a more specific page.

Depth. Somewhere in the site, each page should be reachable in a few clicks from the homepage. Moz describes the pyramid model, where important, revenue-driving pages sit at low crawl depth in its internal links best practices. If a new page can only be reached through three hops of blog archives, plan a link from a higher-level page instead.

Reachability. Confirm each planned link renders as a real <a href> in the final HTML. Links attached to script handlers or built as non-anchor elements may not be extracted by Google's crawlers per the link best practices page. This check takes a minute in view-source and catches framework-specific breakage before it ships.

If you need a repeatable program rather than a one-off read, Ahrefs frames technical work as ensuring pages are crawlable, indexable, and understandable before anything else in its technical SEO guide. Link planning is the editorial half of that same foundation.

A link plan decays in predictable ways: pages get deleted, URLs change, and priority 1 rows silently disappear. Three maintenance habits limit the damage.

• Audit monthly, on a fixed date. Check for rows whose target now returns an error, and for new pages that never gained an inbound row. A broken internal link sends readers to a 404 and wastes the signal the link carried as Moz notes.
• Review hubs quarterly. Look at any page that has become the target of a large share of body links and ask whether it still deserves them.
• Re-plan during migrations. When URLs change, the map is the checklist of links to update. Teams that skip this step ship redirects for external links but forget internal ones.

A useful sanity check for a growing library: sample 10 recent pages and count their inbound body links. If more than two have none, the plan has stopped being followed, not just been forgotten. Fix the process (a publish checklist that blocks release without an inbound row) rather than patching pages one at a time.

The underlying logic ties back to how Google judges content. The helpful content guidance evaluates whether content is people-first and whether the site has a clear purpose and focus per Google's guidance on creating helpful content. A link plan is a focus tool: it forces you to name the few pages that matter and point to them consistently.

If the map work is the bottleneck rather than the writing, Quanta Intelligence can take it off your hands: describe the site, app, document or plan you need and a specialist picks it up and builds it with you, including a link map that survives as the site grows.