In this week’s episode of the Niche Pursuits podcast, Brian Winum and I discuss advanced authority-building for Google, AI visibility, affiliate sites, and content systems that can scale without relying on large manual teams. Brian is a repeat guest, and his latest strategy builds on the authority stacking approach he shared two years ago.
This is a dense, high-level conversation for site owners, SEOs, and affiliate marketers who like technical systems. We cover WP multi-site networks, subdomain authority, LLMS.txt files, tiered link building, entity silos, EEAT, cryptographic trust layers, and the AI workflows Brian now uses to get more done with fewer people.
Watch the Full Episode
Brian’s Background and Current Focus
Brian is an agency owner based in Long Island, New York. He is a partner in MAXBURST Web Design and MAXPlaces Marketing, two businesses that have been around since the mid-2000s.
That gives him roughly 20 years of experience across web design, e-commerce, nationwide SEO campaigns, affiliate sites, and owned web properties. He also teaches courses and uses AI heavily in his current work.
His work today spans several areas:
- Agency client campaigns
- Affiliate site networks
- AI-assisted development
- Authority-based SEO systems
- Content and link automation
Brian’s current strategy isn’t a clean break from what he shared on his first podcast appearance. It’s an expanded version of the same authority-first system, with new layers added for AI platforms and modern indexing challenges.
His Authority Stack That Still Works
Two years ago, Brian talked through a system built around EEAT, authority stacking, and trust signals. At the time, his affiliate-style sites had survived the Helpful Content Update and several Google core updates that damaged many similar sites.
That older system included detailed author profiles, social signals, Google News approval, Reddit polls, expert input, and content that could attract engagement early. Brian says he still uses nearly every micro-strategy from that period. Several pieces remain part of his process:
- Deep author or brand profiles
- Social proof from outside platforms
- Original data from polls
- Expert quotes and commentary
- Authority signals across the web
- Content designed to attract early engagement
The biggest change isn’t the foundation. It’s the way Brian now packages those signals for Google, AI systems, and users who are closer to making a purchase decision.
The Monetization Shift Away From Display Ads
Brian’s revenue mix has changed since his last visit. Display ad income used to be a larger driver across many content sites, including some of his own properties.
Now, display ads sit lower in the mix. He’s still earning from content, though the emphasis has moved toward more focused monetization. His current revenue stack leans toward:
- Guest post income
- Amazon-style affiliate revenue
- Product-focused affiliate offers
- Service-based affiliate offers
- Commercial pages with a tighter intent
This shift also changed his content strategy. Instead of publishing large volumes of broad informational posts, Brian is creating deeper assets aimed at middle-of-funnel and bottom-of-funnel traffic. That means fewer thin question-and-answer posts. It also means more content built around product research, comparisons, buyer needs, and authority signals that support trust.
WP Multi-Site Network Strategy
The biggest change in Brian’s system is his use of WP multi-site networks. He called this one of the main drivers of success in his current approach.
Instead of building a single website on a single domain for every project, Brian builds subdomain-based WordPress multisite networks. A single main domain can support hundreds of tightly focused subdomains. For example, a broad home services domain could include subdomains for:
- Kitchen services
- Backyard services
- Roofing services
- Patio furniture
- Outdoor grills
- Cordless power tools
Brian said he can run 700-800 subdomains on a single domain. This gives him a way to test many niches without buying a large number of separate domains. Each subdomain can focus on a narrow topic. At the same time, the whole network can benefit as individual subdomains gain traction.
The Indexing Advantage of Subdomains
Indexing has become a major problem for many site owners. Brian says his subdomain-based WP multisite setup has helped him overcome that issue. He shared an example from PressClone, one of his press-release-style networks. According to Brian, the network had around 100 unique press releases, while Google had indexed roughly 5,000 to 7,000 pages from it.
That means syndicated or duplicate content across subdomains was still getting indexed at scale. Brian’s read on this is that Google appears to treat the subdomains with enough separation to give them individual value.
The workflow also saves time. Brian said he can clone a site to a new subdomain in less than 30 seconds, then make keyword swaps and have a new property running within minutes. That speed matters because it allows him to test markets quickly. If one topic performs better than another, he can lean into it without rebuilding the whole system.
What Is the Role of LLMS.txt Files
A major part of the interview focused on LLMS.txt files. Brian knows many people in SEO are skeptical of them since they aren’t a formally accepted standard like robots.txt. His point is that a basic LLMS.txt file is unlikely to do much. He compares it to a generic plugin schema, where the default output is rarely sufficient to produce a measurable result.
Brian’s files go much deeper. He uses custom header and footer content to explain the brand, services, audience, authority signals, and external evidence behind the site. His LLMS.txt setup can include:
- Brand summaries
- Product and service details
- Ideal customer profiles
- Citation sources
- Awards and mentions
- Google review summaries
- Important backlinks
- Supplemental AI files
He also builds subdirectory-level files for sections such as services or products. That gives each major part of the site its own focused data layer.
The Crawl Path System Behind the Files
Brian doesn’t simply publish LLMS.txt files and hope AI bots find them. He creates crawl paths that direct bots to the files from several locations across the site.
He mentioned robots.txt, header meta tag injection, XML sitemaps, internal file links, and supplemental markdown or JSON files. The goal is to make the AI-facing authority stack easy to find. The supplemental file stack may include:
- Review profile files
- Brand snapshot files
- Citation files
- Backlink files
- Service files
- Product files
Brian said he has seen traffic hitting these files in server logs. He also reported increases in AI referrals and mentions after adopting this setup. The larger idea is simple, even though the execution is complex. Brian wants AI systems to receive the same brand and authority signals he has already built for search engines.
Tiered Link and Entity Silo Model
Brian’s link building has also become more structured. He now talks about building entity silos through tiered links.
The money site contains the main content, schema, LLMS.txt setup, and core authority signals. Then tier-one links support the site through interviews, listicles, guest posts, press mentions, and other third-party articles.
After that, tier-two links support the tier-one links. One example he gave was publishing a press release about a third-party interview, then syndicating it across hundreds of subdomains. The structure might look like this:
- Money page with deep topical content
- Tier-one interview or feature article
- Tier-two press release about that feature
- Schema added to supporting pages
- Entity references are repeated across the chain
Brian’s goal is to remove ambiguity. The content, schema, LLMS.txt files, and links all reinforce the same person, brand, product, or service entity.
Updated EEAT Focus
EEAT has been central to Brian’s approach for years. Brian’s early experience with Google News approval gave him a close look at the types of signals Google wanted from trusted publishers.
He still thinks trust and authority signals matter. The difference is that he’s now placing more emphasis on genuine people, original content, and visible proof.
Brian said he doesn’t rely as much on fake author personas as he did in the past. They can still work in some cases, though he sees greater value in using people with a public history and third-party validation. His current EEAT stack includes:
- Original reporting
- Expert commentary
- Real people behind content
- Third-party mentions
- Awards and citations
- Clear authorship
- Brand consistency
He’s also going deeper at the page level. Instead of broad surface-level content, he wants each important page to show experience, expertise, and a clear reason to trust the source.
The Cryptographic Trust Layer
One of the most technical parts of the conversation was Brian’s discussion of cryptographic trust layers. This is his way of claiming authorship and creating timestamped proof around published content.
Brian wants a clear record of who published a piece of content and when it appeared online. That matters more as AI-generated content becomes more common across the web. His process includes several signals:
- DMCA badge timestamping
- Internet Archive submission
- Wayback Machine archive links
- Archive schema markup
- Blockchain-based certificates
Brian mentioned ScoreDetect as one of the tools he uses for blockchain certificates. He said the service is around $10 per month for 100 certificates. The purpose is both protection and trust. By timestamping content and tying it back to an author or brand, Brian is creating another layer of proof around originality.
His AI Workflow Behind the Operation
Brian used to need more manual help to run these systems. Today, AI and automation have replaced much of the work once handled by VAs or employees.
He shared one striking example of the scale these systems once required. One of his contacts previously ran a network of about 14,000 websites and later exited for $29 million.
Additionally, Brian shared custom skills, cron jobs, overnight workflows, and AI tools for researching, creating, publishing, and reporting. This allows him to manage larger systems with fewer people involved. Some of the tasks now handled through AI include:
- Guest post inquiry replies
- Price list delivery
- Form collection
- Payment flow handling
- Content creation
- Link delivery
- Site research
- Morning status reports
One example was his guest post sales process. An inquiry comes in, AI sends the pricing, the buyer fills out a form, payment is made, content is created, and the approval link is sent without Brian manually managing the exchange.
The AI-Generated Consensus Strategy
Brian also shared an unusual AI use case around brand discussion. He has been building Reddit-style sites that create threaded discussions around a product, service, or brand.
These aren’t Reddit. They are his own discussion platforms, filled with AI-generated topics and replies. The discussions can cover:
- Product pros
- Product cons
- Brand comparisons
- Service questions
- Interview references
- Use case debates
The aim is to create more discussion around an entity. Brian wants those discussions to add context and support the wider authority system he is building across content, links, files, and schema.
Final Thoughts
This episode with Brian Winum is one of the more advanced SEO conversations we have had on the podcast. It moves quickly, and several sections could easily become full episodes on their own.
The central takeaway is that Brian isn’t relying on one tactic. His system combines authority stacking, subdomain networks, AI-facing files, tiered links, entity signals, authorship proof, and automation. The full system includes:
- WP multi-site subdomain networks
- Deeper commercial content
- LLMS.txt files and supplemental AI files
- Tiered link building
- Entity silos
- EEAT signals
- Timestamped authorship
- AI-powered operations
Some listeners may be skeptical of parts of the system, and Brian says he expects that. His view is that the results are coming from the full stack, not from any single file, link, or plugin.
For site owners, affiliate marketers, and SEOs, this conversation is worth revisiting more than once. The ideas around subdomains, LLMS.txt files, entity silos, and AI-assisted operations offer a detailed look at where advanced authority building may be heading.
Links & Resources
