316: Spencer Haws On How To Automate Internal Link Building For SEO

Podcast: Download (Duration: 44:46 — 51.5MB)

An article with zero internal links pointing to it is one Google has no reason to consider important, which is why adding just one or two often produces disproportionate ranking gains. Spencer Haws added internal links to roughly 30 articles on one site and watched over 75% of them improve, with several moving from page three or four into the top three results.

In this episode I spoke with Spencer Haws of Niche Pursuits, who created Long Tail Pro, built and sold an Amazon FBA business, and now runs Link Whisper, an internal linking plugin.

Below is everything: how to prioritize which pages get links, why internal anchor text follows different rules than external, the hub and spoke structure, and why he sold most of his businesses to focus.

Key takeaways

• Zero internal links is the biggest opportunity. Adding one or two can move a page substantially.
• Over 75% of articles improved in Spencer’s test, some from page three or four to the top three.
• Internal anchor text rules are looser than external, so keyword-rich anchors are safe.
• Cross-reference with analytics to find pages already ranking on page two with no links.
• Hub and spoke signals topical authority, which is what makes the pillar article rank.
• Skip pages you are not trying to rank, and link outward from them instead.
• The 100-site strategy is dead. One site with hundreds of articles beats a hundred thin sites.
• Ask developers what problems they see with your idea, which is how you find the good ones.

Why internal links move rankings

Google’s entire search engine is built on links as votes, and links from your own site count.

The familiar version is external links. A link from an authoritative site helps the page it points to rank.

Your own site also carries authority, which means a link from one of your pages to another functions as a vote in the same way.

Google has stated this and it is well established in the SEO community, and it remains the easiest ranking factor to control because you own both ends.

What happened in Spencer’s internal linking test

He built internal links to roughly 30 articles on a public case study site and changed nothing else.

Over the following months, more than 75% of those articles improved their rankings.

Some moved dramatically. Articles sitting on the third or fourth page of Google reached the top three results from three or four internal links.

Many were longer-tail keywords that were relatively easy to rank, and they were not ranking at all on their own.

Spencer’s framing of why this works is about signal. An article with no internal links gives Google nothing indicating it matters, and even two links change that.

How to prioritize which pages get internal links

Start with articles that have zero internal links pointing to them.

His plugin sorts articles by internal link count, and any site’s analytics can produce the same list with more manual work.

The refinement is cross-referencing. Pull up Google Analytics or Search Console and find which of those zero-link articles are already getting organic traffic or sitting at the bottom of page one or top of page two.

Those are the quick wins. Already ranking, already getting traffic, and receiving no internal support at all.

The pages to skip are ones with five or six internal links already that still rank on page ten. Those have a different problem.

Which pages do not need internal links

Not everything on your site is meant to rank.

Spencer’s caveat covers news posts, quick updates, and anything published for a viral or timely reason rather than for search.

My own version is podcast episode pages, which generally do not rank and are not intended to.

The useful move for those pages is the reverse direction. Link outward from them to the articles you actually want to rank, rather than trying to push links into them.

How internal anchor text rules differ from external

You can use keyword-rich anchor text internally in ways that would get you penalized externally.

The mechanism is the same in both cases. Anchor text tells Google which keyword the destination page is about, so “best survival knife” carries more weight than “click here.”

The penalty risk with external links comes from unnaturalness. If every external link to a page uses identical keyword anchor text, that pattern looks manufactured.

Google’s guidelines treat internal links differently, because it is your own site and the most relevant anchor text genuinely may repeat.

The practical consequence is that you can be more deliberate internally, and using a keyword research tool to pick anchor text is worth doing.

Rewriting a sentence to accommodate good anchor text is worthwhile when it fits naturally, and a link with imperfect anchor text still beats no link.

How many internal links to build

There is no strict maximum on links pointing to a page.

The guidance that exists came from Google representatives years ago suggesting more than 100 links coming from any single article may be too many.

Links coming to an article have no equivalent ceiling.

The one place to exercise restraint is on pages you are actively trying to rank. Three or four outbound internal links rather than ten or twenty keeps the link equity more concentrated.

Spencer still links outward from those pages when it genuinely serves the reader.

How the hub and spoke structure works

Topical clusters tell Google you are an authority on a subject rather than someone with one article about it.

The structure has a pillar article covering the core topic, with spoke articles covering specific subtopics.

For an ecommerce example, the pillar is how to start an ecommerce business, and the spokes are how to sell on Amazon, how to use Shopify, and individual platform reviews.

The linking runs both directions. The pillar links out to each spoke, and each spoke links back to the pillar.

Spencer’s explanation of why this works is about evidence. One article on a topic does not establish expertise, and ten tightly related interlinked articles using the associated vocabulary does.

That is what justifies ranking the pillar article higher.

How to make internal linking a habit

Build it into your publishing process rather than treating it as a separate project.

Spencer’s recommendation is a standing instruction, whether for yourself or for writers and editors. Every new article builds three or four internal links.

The reason it needs to be systematic is that good intentions decay. Nearly everyone starts with the intention and forgets after publication.

Why the hundred-site strategy stopped working

Spencer built over 100 niche sites around 2011 and the approach no longer works.

The original model was simple. Get an exact match domain, publish five to ten articles targeting specific long-tail phrases, and build a link or two.

At roughly $100 to $200 a month per site, a hundred sites cleared $10,000 monthly.

Google changed two things. Exact match domains no longer confer a meaningful boost, and content depth requirements rose substantially.

Site-level authority became the dominant factor, which means articles rank better across the board on an authoritative domain.

The consequence is that one site with hundreds of articles now beats a hundred thin sites, which is why Spencer sold most of his portfolio down to two or three.

Why he sold his Amazon business and most of his sites

He was giving thirty minutes each to too many things.

At the time he had seven or eight niche sites, an Amazon FBA business selling pillows, and software development starting up again.

The selection criteria for what to keep were current performance and future potential, weighted toward potential.

Amazon was his largest revenue generator when he sold it. His reasoning was competitive rather than financial.

New competitors were entering constantly, he had no patent or genuine defensibility, and maintaining existing revenue was going to be a grind before any growth.

He also simply had more interest in software, which is where his largest exit had come from.

WordPress plugin versus standalone software

The choice depends on how much the product needs to interact with the site itself.

Standalone software works anywhere regardless of platform, whether as a downloadable product or a hosted service.

A plugin only runs on WordPress and can interact far more directly with the site.

Link Whisper needed that interaction. Building links inside a WordPress site, generating reports, and modifying content all become substantially easier with code installed on the site.

Long Tail Pro made sense as standalone software for the opposite reason. Keyword research spans multiple sites, so tying it to any one installation would have limited it.

Listing on the official WordPress repository is free, and the review process verifies the code is safe and functional. Spencer’s took roughly two to three weeks.

How to find a good developer

Ask candidates what problems they see with your idea.

Spencer posts the job with a full description of the software concept, then asks what problems they foresee or how they would improve it.

He is not looking for feature suggestions. He is looking for whether they identified something he had not.

The filter works because most responses say they see no problems. The one or two who point out that it will not work on certain sites, or will break for a specific technical reason, are the candidates worth interviewing.

He is not concerned about idea theft from posting the concept publicly, and it has never been an issue.

His sources are freelance job platforms and referrals from people he knows.

What actually matters in SEO now

The core concepts have not changed, and the standard for each has risen.

Quality content is the foundation, paired with targeting a specific phrase. A well-written article nobody is searching for cannot receive traffic.

Keyword placement still matters. The title tag most of all, the phrase used a few times through the article, and increasingly the related vocabulary that demonstrates depth.

External links remain the backbone of Google, with natural links outperforming built ones and spammy links progressively devalued through successive updates.

Internal links are the piece you fully control, which is why they are the most reliable lever available.

On the disavow tool, Spencer used it once on a site receiving heavy automated spam links, without knowing whether it helped. He had received no penalty and acted preemptively.

Frequently asked questions

Do internal links actually improve rankings?

Yes. Spencer Haws added internal links to roughly 30 articles with no other changes and over 75% improved, with several moving from the third or fourth page of Google into the top three results.

Which pages should get internal links first?

Articles with zero internal links pointing to them, prioritized by whether they already get organic traffic or sit at the bottom of page one. Those are already close and receiving no internal support.

Can you use the same anchor text repeatedly for internal links?

Yes. Google’s guidelines treat internal links differently from external ones, because it is your own site and repetition of the genuinely relevant phrase is expected rather than suspicious.

How many internal links should point to a page?

There is no meaningful maximum for incoming links. The guidance that exists concerns outgoing links, where more than 100 from a single article may be excessive.

Should every page on your site get internal links?

Only pages you are trying to rank. News posts, quick updates, and podcast episodes often are not search targets, and the better move is linking outward from them to pages that are.

What is the hub and spoke model?

A pillar article on a core topic surrounded by spoke articles on subtopics, all linking to each other. It signals topical authority, since ten interlinked related articles establish expertise that one article cannot.

Does building many small niche sites still work?

No. Exact match domains no longer confer a meaningful boost and content depth requirements rose. One site with hundreds of articles now beats a hundred thin sites, which is why Spencer sold most of his portfolio.

How do you find a good software developer?

Describe your idea and ask what problems they foresee. Most say they see none, and the one or two who identify something you missed are the candidates worth interviewing.