The Pillar Cluster Model Is Dead. Here’s What Replaced It.

I need to tell you something that might annoy you: if you’ve spent the last five years building out pillar pages and topic clusters the way every SEO blog told you to, there’s a decent chance you’ve been wasting your time.

Not entirely, mind you. The foundational idea, that comprehensive topical coverage matters, was never wrong. But the specific implementation model we’ve all been following? The hub-and-spoke structure where everything links back to a central pillar page? The carefully constructed silos mixing commercial intent with informational content like some kind of SEO smoothie nobody asked for? That’s becoming increasingly problematic.

I know this because I’ve spent the past eighteen months systematically dismantling and rebuilding content structures for clients who couldn’t understand why their “perfectly optimised” pillar cluster models weren’t performing. And the pattern that emerged was pretty clear: the traditional model doesn’t align with how search engines (or users, frankly) actually behave anymore.

This isn’t going to be a short article. Sorry about that. I’m going to walk you through why the pillar cluster model worked when it did, why it’s failing now, what intent separation actually means in practice, and why internal linking has become both more important and more complicated than anyone wants to admit. I’ll also acknowledge where I’m still figuring this out myself, because honestly, we’re all making educated guesses about some of this, and anyone who tells you otherwise is selling a course.

Let’s start with why we’re even having this conversation.

Why the Pillar Cluster Model Made Sense (Past Tense)

The pillar cluster model emerged around 2016–2017 as a response to how Google was evolving. The logic was sound:

• Google was getting better at understanding topics, not just keywords, so organising content by theme made sense
• Comprehensive coverage signalled expertise, having multiple related pieces demonstrated authority
• Internal linking passed authority, connecting everything to a central hub theoretically concentrated ranking power
• It solved organisation problems, gave content teams a framework for planning and structure (which, let’s be honest, most of them desperately needed)

And for a while, it worked. I built dozens of these structures between 2017 and 2022. Pillar pages ranked, clusters supported them, clients were happy, case studies were written. The model became orthodoxy because it delivered results.

But here’s the thing about SEO orthodoxy: it’s usually about three years behind the actual algorithm. By the time something becomes a “best practice” on every blog, it’s often already on its way out. We just haven’t noticed yet because we’re too busy writing the blog posts.

What We Misunderstood About Why It Worked

Looking back (and I’ve had plenty of time to look back whilst fixing broken pillar structures at 3 AM, fuelled by increasingly questionable coffee), I think we misattributed causation. The pillar cluster model didn’t work because of the specific hub-and-spoke structure. It worked because:

• It forced people to create comprehensive content when most sites had gaps the size of the Grand Canyon
• It encouraged topical depth when shallow content was still the norm
• It provided internal linking when many sites had essentially none (genuinely, some of these sites were navigational wastelands)
• It came with content planning when most people were throwing darts at a keyword spreadsheet and hoping for the best

The structure itself was fine, but it wasn’t the magic ingredient. The comprehensiveness was. And now that everyone and their dog is building pillar clusters, that comparative advantage has evaporated. You’re not demonstrating topical depth anymore, you’re just doing what everyone else is doing, which is roughly the opposite of a competitive advantage.

But there’s a bigger problem we didn’t see coming.

The Mixed-Intent Death Spiral

Here’s where traditional pillar cluster implementations fall apart, and it’s not pretty: they mix intents indiscriminately.

Your typical pillar structure looks something like this:

Pillar Page: "Complete Guide to Email Marketing"
├── "What is Email Marketing?" (informational)
├── "Email Marketing Benefits" (informational)
├── "How to Build an Email List" (informational)
├── "Best Email Marketing Software" (commercial)
├── "Email Marketing Services" (commercial)
└── "Hire Our Email Marketing Agency" (transactional)

All neatly connected, all linking back to the pillar, all sitting in the same silo. Looks great in a content strategy spreadsheet. Looks fantastic in a pitch deck. Performs terribly in reality.

Why This Structure Is Broken

I’ve tested this across enough sites now to be confident about this (I say “confident” because “certain” would be a lie, certainty in SEO is a red flag): search engines treat different intents as fundamentally different queries, even within the same topic. And they don’t want you mixing them in the same structure.

When someone searches “what is email marketing,” Google understands this is an educational query. They want clear, objective, informational content. When someone searches “email marketing services,” Google knows this is commercial intent. They want comparison pages, pricing information, vendor options.

These aren’t variations on the same query. They’re different query types that happen to share keywords. It’s like saying a cookbook and a restaurant menu are the same thing because they both mention food.

And here’s the killer: when you interlink them aggressively in a traditional pillar structure, you’re creating confusing signals. Your informational content is linking to commercial pages. Your commercial pages are linking back through informational content to reach the pillar. The whole structure becomes a muddle of mixed intent that Google increasingly struggles to categorise properly. It’s trying to figure out what your content is, and you’re essentially responding with “everything, all at once.”

I first noticed this on a financial services site in 2023. Beautiful pillar structure, comprehensive content, proper internal linking, textbook stuff. Rankings were mediocre and getting worse. We split the structure into separate intent-based silos, informational separate from commercial, and within two months, the informational content jumped an average of 12 positions. The commercial pages followed three months later.

Could have been coincidence. Could have been something else entirely. But I’ve replicated the pattern seven times since then across different industries, and the results are consistent enough that I’m reasonably confident about the principle. (I keep a spreadsheet. I’m that sort of person.)

The User Behaviour Problem

There’s also a user experience angle here that’s worth considering. When someone lands on “what is email marketing” and you’re aggressively internal linking to “hire our email marketing agency,” you’re not being helpful. You’re being that person at a party who steers every conversation back to their start-up.

Yes, some informational searchers will convert commercially. But forcing commercial links into informational content doesn’t increase conversion, it degrades the informational value and makes the whole page feel less trustworthy. (I learned this the hard way after a client insisted we link every informational article to their pricing page. Bounce rate went up 23%. Conversions went down. We removed the links. Metrics improved. I resisted the urge to send a “told you so” email. Professional growth.)

Search engines are increasingly good at detecting this kind of user dissatisfaction. Dwell time, pogo-sticking, bounce rate, these signals matter. And mixed-intent structures create poor user experiences more often than they create conversions.

What Intent-Based Silos Actually Mean

Right, so if traditional pillar clusters are out, what’s in? Intent-based silos with strategic, limited crossover.

The principle is straightforward (the implementation, predictably, is not): group your content by primary intent first, topic second. Keep your informational content largely separate from your commercial content, even when they’re on the same broad topic.

The Practical Structure

Instead of one pillar with mixed-intent clusters, you build separate intent hierarchies:

Informational Silo:
"Email Marketing Knowledge Hub"
├── What is Email Marketing?
├── How Email Marketing Works
├── Email Marketing Statistics
├── Email Marketing Best Practices
├── Email Deliverability Guide
└── Email Marketing Mistakes to Avoid

Commercial Silo:
"Email Marketing Solutions"
├── Email Marketing Software Comparison
├── Best Email Marketing Platforms 2025
├── Email Marketing Tools for Small Business
├── Email Marketing Pricing Guide
└── Email Marketing Service Reviews

These silos can acknowledge each other’s existence, they’re not pretending the other doesn’t exist, they’re just not aggressively holding hands. They’re separate content ecosystems serving separate intents.

Where (and When) They Should Connect

This doesn’t mean never linking between intents. It means being strategic about it. Think of it less like a motorway interchange and more like a country lane connecting two villages.

When it makes sense to link informational → commercial:

• At the natural end of the user journey (“Now that you understand email marketing, here’s how to choose the right platform”)
• When genuinely relevant to solving the user’s problem
• As a single, contextually appropriate link, not five scattered throughout like confetti

When it makes sense to link commercial → informational:

• To support commercial claims with educational content (“Learn more about deliverability”)
• To provide necessary context for decision-making
• As supplementary resources, not as the main internal linking focus

When it doesn’t make sense:

• Aggressive cross-linking just because keywords overlap
• Forcing commercial links into purely educational content
• Using informational content as a covert pipeline to pass authority to commercial pages (Google’s not daft)

The mindset shift is this: these are separate user journeys that occasionally intersect, not variations of the same journey.

Why Internal Linking Suddenly Got Complicated

Here’s where this all gets a bit tricky. If you can’t rely on traditional hub-and-spoke structures where everything connects back to a pillar, internal linking becomes both more important and more difficult to execute properly. Welcome to the fun part. (I’m using “fun” loosely.)

The Old Model Was Easy (Maybe Too Easy)

Traditional pillar clusters had simple internal linking rules:

1. Every cluster page links to the pillar
2. The pillar links to every cluster page
3. Cluster pages can link to each other if relevant
4. Everything flows through the hub

Easy to implement. Easy to audit. Easy to explain to clients. And increasingly ineffective.

The New Model Requires Actual Thinking

Intent-based silos don’t have the same tidy structure. Your internal linking needs to be contextual, relevant, and genuinely useful rather than following a formulaic pattern. I know, I know, we all loved the formula. Formulas are comforting. But comfort and effectiveness parted ways a while ago.

This means:

Hierarchical authority still matters. You still need some form of hierarchy within each intent silo. More general, authoritative pages at the top, more specific pages branching off. But the hierarchy is based on specificity and comprehensiveness within that intent, not just traffic potential.

Contextual relevance trumps keyword matching. Just because two pages share keywords doesn’t mean they should link to each other. The link needs to make sense in context and serve the user’s immediate needs. If you wouldn’t naturally recommend that page to someone reading this one, the link probably shouldn’t be there.

Depth of linking varies. Some topics warrant deep, multi-layered internal linking within the intent silo. Others work better with shallow, direct structures. There’s no universal pattern, which is maddening but also just how things are.

Anchor text becomes more important. When your structure isn’t formulaic, your anchor text needs to do more work. Generic “click here” or “learn more” anchors don’t cut it anymore. You need descriptive, contextual anchors that tell both users and search engines what they’re about to find.

What Good Internal Linking Looks Like Now

I’m going to show you an example from an insurance client I work with (details changed enough that they won’t ring me up annoyed). We completely restructured their content last year, and the internal linking strategy was crucial to making it work.

Informational Silo: Life Insurance Knowledge

The top-level page is “Understanding Life Insurance”, comprehensive, authoritative overview of the entire topic. This isn’t trying to rank for commercial terms. It’s pure education. It wants to be the page you’d send to a friend who asked “so what actually is life insurance?”

From there, we have second-tier pages on specific aspects:

• “Term Life Insurance Explained”
• “Whole Life Insurance Guide”
• “Life Insurance Underwriting Process”
• “Life Insurance Policy Types”

Each of these links back to the overview when providing context, but they primarily link laterally to related informational content:

From "Term Life Insurance Explained":
→ "How Life Insurance Premiums Are Calculated" (relevant to understanding costs)
→ "Life Insurance Medical Exam Requirements" (relevant to application process)
→ "Term Life vs Whole Life Insurance" (comparison page, still informational)

Notice what’s missing: aggressive links to “Get a Term Life Insurance Quote” or “Compare Life Insurance Providers.” Those exist on the site, but they’re in the commercial silo. They’re over there, doing their own thing.

Commercial Silo: Life Insurance Products

Separate structure, separate purpose:

"Life Insurance Quotes and Comparison"
├── "Best Term Life Insurance 2025"
├── "Affordable Life Insurance Providers"
├── "Life Insurance for Seniors"
└── "No-Exam Life Insurance Options"

The result? Informational content ranks higher for educational queries. Commercial content ranks higher for comparison and buying queries. Overall visibility increased. Client happy. Me slightly less sleep-deprived.

The Authority Flow Problem

Here’s a question I get constantly: “But doesn’t separating silos mean you’re not concentrating authority? Doesn’t the pillar model work because it consolidates ranking power?”

Short answer: that was always somewhat mythical.

Longer answer: yes, internal linking passes authority (PageRank, or whatever Google’s calling it this week, they rebrand it more often than a struggling start-up). But the assumption that funnelling everything through a hub page concentrates authority only works if search engines actually want to rank that hub page for all related queries. They increasingly don’t.

Search engines want to rank the most relevant page for a query, not the most internally linked page. If someone searches “what is term life insurance,” Google wants to rank your specific term life insurance explanation page, not your general “Complete Guide to Life Insurance” pillar page. All that concentrated authority is pointed at the wrong target.

What does help is building topical authority across an intent category. If all your informational life insurance content is comprehensive, well-structured, and properly interlinked within that informational silo, search engines recognise you as authoritative on life insurance information. That authority lifts the entire silo.

Similarly, if your commercial content is thorough and well-structured within its silo, you build commercial authority. These types of authority are related but distinct, and trying to combine them in a single structure dilutes both. It’s like trying to be both the teacher and the salesperson at the same time, you end up being a slightly untrustworthy version of both.

Rebuilding Existing Pillar Structures (Without Destroying Everything)

Right, so you’ve probably got a bunch of pillar cluster content that follows the traditional mixed-intent model. What do you do with it? (Please don’t say “burn it down.” That’s never the answer, however satisfying it might feel.)

Option 1: Gradual Intent Separation

This is what I usually recommend because blowing up your entire site structure in one go is terrifying and potentially disastrous. I speak from experience. Well, a colleague’s experience. Fine, it was me. Once.

Step 1: Audit your current structure. Map every page in your clusters by primary intent, Informational, Commercial/Comparison, or Transactional. Be honest about this. That page you’re calling “informational” but which is actually 60% sales pitch? It’s commercial. Label it accordingly.

Step 2: Identify natural groupings. Look at how your content clusters by intent rather than topic. You’ll often find that the intent groupings are cleaner than the topic groupings, which tells you something about how users actually think about your content.

Step 3: Adjust internal linking first. Before changing URLs or structure, adjust your internal linking to reflect intent separation. Strengthen links within intent groups. Weaken or remove aggressive cross-intent links. Keep strategic, contextual cross-intent links where they genuinely help users. Wait 2–3 months. Monitor ranking changes. Try not to refresh Search Console obsessively. (You will anyway. We all do.)

Step 4: Consider structural changes. If the linking changes produce positive results, consider restructuring. Create separate parent pages for each intent silo. Adjust URL structure if necessary.

Step 5: Don’t rush it. I’ve seen people panic-restructure their entire site based on one article (hopefully not this one), and it usually goes badly. This is a months-long process, not a weekend project.

Option 2: Keep It, But Fix the Linking

If your pillar structure is performing reasonably well, you might not need to blow it up. Sometimes the answer is simply establishing clear intent zones within the cluster and minimising the crossover between them. Think of it as putting up partition walls rather than demolishing the building.

Option 3: Start Fresh on New Topics

The safest approach: leave existing performing content alone, implement intent-based silos for new content initiatives. Compare performance over time. Let the data make the case for you.

What Usually Goes Wrong

I’ve implemented (or fixed) enough of these transitions to know the common failure modes. Consider this the “learn from my mistakes” section:

• Over-correction. People swing from aggressive mixed-intent linking to zero cross-intent linking, as if informational and commercial content are going through a divorce. Some strategic connection is fine. Encouraged, even.
• Impatience. Expecting results in three weeks. Give it at least 2–3 months. Google isn’t in a rush, and neither should you be. (Easy to say, hard to do when the client’s asking for updates weekly.)
• Forgetting about users. Getting so focused on search engine signals that you remove genuinely helpful links. If a user reading your informational piece would genuinely benefit from seeing your product page, keep that link. Intent separation is a guideline, not a religion.
• Inconsistent implementation. Applying intent separation to half your content and leaving the other half as a tangled mess. Google sees your whole site. It notices.

The Complications Nobody Talks About

Let me be honest about where this model gets messy. And it does get messy. If anyone presents this as a clean, simple framework, they’re either oversimplifying or they haven’t tried to implement it at scale.

Multi-Intent Queries Exist

Some queries genuinely have mixed intent. “Email marketing” could be informational or commercial. Google themselves sometimes can’t decide, which is why you see SERPs that mix both types of results like a confused DJ switching between genres.

My current approach: create separate pages for each dominant intent. Let search engines figure out which to rank. Does this always work? No. Do I have a better solution? Not yet. If you do, genuinely, email me.

Brand Queries Complicate Things

When someone searches “your brand + email marketing,” they might want either your informational content or your commercial offering. This is more of a business decision than an SEO one, and I’ve found it’s best handled on a case-by-case basis rather than with a universal rule.

Enterprise Sites with Massive Content Libraries

If you’ve got 10,000 pages across hundreds of topics, implementing intent-based silos is a multi-year project. I won’t pretend otherwise. Start with highest-value topics. Measure. Expand gradually. Try not to think about the total scope of work too often, because it will make you want to lie down.

The Measurement Problem

How do you measure success when you’re changing fundamental structure? Rankings shift during transitions. Traffic is unreliable. Your stakeholders are nervous. I usually look at: rankings for clear informational queries, rankings for commercial queries, overall topical visibility, and conversion rates on commercial content. But I’ll be honest, the first month or two after a structural change is mostly just managing anxiety (yours and the client’s).

What This Means for Content Planning Going Forward

Plan by Intent, Then by Topic

Rather than “Let’s create a pillar cluster about email marketing,” think:

“We need informational coverage of email marketing for people learning about it”
AND
“We need commercial coverage for people evaluating email marketing solutions”

These are separate briefs. Separate editorial calendars. Separate success metrics. Possibly separate writers, depending on your team. (Your best educational writer might not be your best comparison page writer, and that’s perfectly fine.)

Resource Allocation Changes

In an intent-separated model, you might create:

• One informational overview + 5–6 supporting informational pages
• One commercial comparison page + 4–5 commercial evaluation pages
• Strategic connection points between them (two or three thoughtful links, not twenty haphazard ones)

Keyword Research Gets More Nuanced

You need to bucket by both topic and intent, then plan separate content structures for each intent category. More work upfront, yes. But better results, and less time spent six months later wondering why your beautiful pillar page is stuck on page three.

The Honest Limitations of What I’m Suggesting

I wouldn’t trust anyone who presents an SEO framework without caveats, so here are mine:

• I can’t prove this is why rankings improved. Correlation, not proven causation. I’m confident, but “confident” and “certain” are different words for a reason.
• This might be temporary. Search engines evolve constantly. What works now might not work in two years. (Though I suspect intent separation is more fundamental than a passing trend.)
• It requires more strategic thinking than formulaic pillar clusters. That’s either a feature or a bug, depending on how you feel about spreadsheets.
• It’s harder to scale. Formulas scale beautifully. Judgment doesn’t.
• Search engines might not see it the way I do. I’m interpreting patterns, not reading source code.

Where to Go From Here

Look, I’m not suggesting you immediately blow up your carefully constructed pillar clusters. (If you do and it goes wrong, I absolutely do not want that email.)

What I am suggesting:

Test the thesis. Pick one topic where you have a traditional mixed-intent pillar structure. Separate the internal linking by intent. Wait three months. See what happens. If it works, expand. If it doesn’t, you’ve learned something valuable either way.

Be intentional about new content. Try the intent-separated approach from the start for new topics. Compare its performance against your existing pillar structures. Let the data argue with me on your behalf.

Think about user journeys. Make sure your internal linking actually helps users rather than following a formula. Ask yourself: “Would I genuinely recommend this link to someone reading this page?” If the answer is no, the link probably shouldn’t be there, regardless of what any structural model says.

Accept ambiguity. There’s no perfect answer here. There rarely is in SEO. Anyone offering you The Definitive Framework is either oversimplifying or hasn’t been doing this long enough to know better.

And maybe, just maybe, in three years someone will write an article about how intent-based silos are dead and here’s what replaced them. That’s how this industry works. We build models, they work until they don’t, we build better models. The trick is being honest about which phase we’re in.

For now, based on what I’m seeing across dozens of sites and multiple industries, intent separation with contextual internal linking is outperforming traditional pillar clusters. Your mileage may vary. Test it yourself. Let me know if I’m wrong, I’m genuinely interested, not just saying that.

Just… maybe keep a backup before you change everything, yeah?