The Authority Pie: What Actually Deserves Your Marketing Investment?

Someone told me last month that social media is now 40% of GEO. I asked forty percent of what. The conversation moved on.

The dishonest part of that number isn't the figure itself. It's the implication underneath it: that digital authority has a single formula, that somebody has worked it out, and that your job is to fund it evenly across every channel on the list. Website, blog, Instagram, LinkedIn, TikTok, YouTube, reviews, press, Google Business Profile, email, product feeds, podcasts, influencers, SEO, GEO. You could spend a full year's budget just trying to be present everywhere. Plenty of companies do exactly that.

The better question is which parts of that list actually carry the business, and which parts only become useful once the pieces that hold the weight are already standing.

I keep coming back to a house. Some things are foundation. Some are framing. Some connect the structure to the systems around it. And some things are furniture. Furniture matters, and it changes how the whole place feels, but buying a better sofa is a strange response to a cracked foundation. Marketing gets this wrong constantly. A company will have thin service pages, no independent reputation, and business information that contradicts itself across three platforms, then spend heavily on short form video because somebody decided it needs more reach.

Reach to what?

---

There is no universal authority pie

A high value professional, a residential contractor, an outdoor ecommerce brand, a consulting firm and a software company should not divide their effort the same way, because the customer is evaluating something different in each case. One buyer needs trust and judgment. Another needs proof that somebody can physically finish a project on a real house. Another needs to know whether a product survives a season outdoors. Another wants documentation. Another cares mostly about what customers and creators are showing online.

And there is no public formula showing that Google assigns articles one percentage, Reddit another and YouTube a third. Google's own guidance on AI features, updated in May 2026, is fairly blunt that established SEO fundamentals still apply to AI Overviews and AI Mode. Those experiences are built on the same ranking and quality systems as the rest of Search, not on a published weighting by content channel.

So the percentages below are something else entirely. They are an OBVS planning model: a starting position for where I would concentrate authority building effort for a given type of business. Not ranking factors. Not leaked weights. Just a way of forcing a decision about what gets funded first.

Highlighted figures mark where the weight sits in each row. These are planning allocations, not ranking factors.

I wouldn't argue about whether one box should read 20% instead of 25%. The useful part is how far the mix travels from one row to the next. And these are hypotheses, not permanent settings. Customer behavior, conversion economics, search visibility, reputation gaps and whatever is currently strangling growth should all move the numbers over time.

---

Start with the parts you actually own

For almost every company, the website and its core commercial information are still the foundation: service pages, product pages, categories, documentation, company information, professional biographies, case studies, portfolio, conversion paths. The exact pieces vary by business. What they have in common is that the company controls them outright.

Sitting on top of that is authority content: articles, original analysis, technical explanations, research, comparisons, testing, the material that shows the company genuinely understands its subject. On paper those two categories look similar. They do different jobs. A product page tells me what the product is; a serious durability test tells me whether to believe it. A consulting service page explains what the firm does; a well written case study shows me how the people inside it think.

Then there are the systems that live outside the website: business profiles, product feeds, structured product and business data, accurate contact information, inventory, merchant data. For the right company these are not minor technical housekeeping. Google points ecommerce and local businesses toward accurate Merchant Center and Business Profile data specifically because product and local information can surface in both conventional results and AI generated ones.

And then there is the layer nobody controls: reviews, legitimate press, independent product reviews, industry references, associations, links, customer discussion. Corroboration. It's last on the list of things you can influence and near the top of the list of things that decide whether anyone believes you.

Social, video and creator distribution come last in the sequence, which is not the same as least. Sometimes that's the furniture. Sometimes it's very nearly structural.

---

A professional and a contractor need different houses

Take an independent professional advising clients through expensive decisions. Part of what the client buys is the service. A larger part is judgment: experience, knowledge, communication, and confidence that this person knows what happens when things get complicated in month four.

I'd put more than half the authority effort into the owned site plus real demonstrated expertise. A polished social presence helps, but I'd rather have ten useful explanations of difficult client situations than two hundred inspirational posts about excellence. The website should tell me what this person knows. Case studies help. Real analysis helps. Detailed answers to the awkward questions clients are slightly embarrassed to ask help most of all.

Outside validation matters here too, because professional authority can't be built entirely out of what somebody says about themselves. Reviews, legitimate media contributions, professional recognition, independent references, visible evidence of work. Social can then carry that expertise outward, and I would use it. I just wouldn't mistake posting frequency for authority.

A residential contractor is a different and usually simpler house. Business information and reputation move much closer to the foundation. A homeowner wants to know whether the company does this kind of work, whether it has handled similar projects, whether previous customers are happy, and whether anyone answers the phone. A beautifully written 2,000 word article does very little when the review profile is thin and there's no visible proof of finished jobs.

For that business I'd push far more effort into local search infrastructure, reviews and project documentation, then publish where the contractor actually has something to say. Three exterior materials we won't use again on walls that face west is worth reading, because it comes from having been wrong once. Five reasons home maintenance is important exists only because the SEO package included four blogs a month.

---

For product brands, social moves into the framing

Now consider a premium outdoor brand selling physical goods. The catalog itself is a major marketing asset: titles, materials, dimensions, variants, images, availability, reviews, returns, shipping, feeds. That's foundation, and it's the part most often treated as data entry and handed to whoever is available.

But visual content deserves a much larger share here, because the customer wants to see the thing. How does it fit? How large is it next to a person? What does it look like after a season? How does it pack down? Does it shift around when you actually move in it? Ninety seconds of video answers questions a thousand words of polished copy can't. Creator and customer content produces independent proof on top of that.

For this company, social and video aren't decoration applied once the SEO work is finished. They carry buying information. I'd still want the owned core solid first. If every video is beautiful but the product pages are thin, the specifications are inconsistent and the inventory data is wrong, the structure under the brand is weak. Once those basics are right, I'd spend aggressively on showing the product in the real world.

---

Consulting is mostly intellectual inventory

A B2B consulting firm nearly inverts the product brand model. There's often no physical thing to demonstrate. What's being sold is thinking, so I'd put roughly a third of the effort into substantive authority content, not "leadership insights" but actual arguments. Why these projects fail. Where companies misdiagnose the problem. What a particular operating model gets wrong. Frameworks, research, case studies, and useful disagreement with conventional wisdom in the places where the firm has evidence to back it.

I want to be able to read the work and understand how these people see the problem before I ever get on a call with them.

LinkedIn, podcasts and video can be strong distribution here, especially when individual partners carry the brand. But the strongest ideas should still live somewhere the firm owns. Otherwise the intellectual property is being built inside somebody else's feed.

Hospitality and wellness lean further toward reputation, visual proof and guest experience. Software pulls hard in the opposite direction. For SaaS, documentation is often the best marketing the company owns: integration pages, migration guides, security information, technical comparisons and implementation docs answer far more valuable questions than another top of funnel post. That's why the SaaS row puts 60% into owned core and authority content combined. It isn't glamorous. People trying to implement software rarely want glamour.

---

Activity is not authority

The most common failure is confusing the two. A company can publish every day for a year and accumulate almost none. Twenty posts calling the founder an expert don't produce the evidence that one piece of serious original research does. Fifty generic articles don't create subject depth. A paid press release isn't independent editorial coverage. Ten thousand directory listings aren't credibility.

Authority accumulates when the pieces reinforce each other. Say the outdoor company runs a six month durability test. It becomes a detailed article with original data and photographs. The test becomes a long form video, and sections of that video become clips. The product page cites the findings. Customers get an email about it. An independent reviewer notices and discusses the product. Another publication references the data. Somebody asks an AI assistant a question about durability in that category, and part of the article is what gets retrieved.

One piece of real knowledge, several different kinds of evidence. That's what compounding looks like. It isn't about being everywhere. It's about having something worth carrying from one place to the next.

---

The pie should move

None of these allocations should stay fixed. If a business has a genuinely bad website, the first quarter might be 70% owned core. If the site is excellent but there's no independent reputation anywhere, stop publishing for a minute and go work on corroboration. If a product brand has good search visibility and weak visual demonstration, video deserves the money. If a consultant has a decade of excellent research nobody has read, the problem is distribution, not production.

The allocation's real job is to expose the bottleneck. That's a better prompt than "what should we post this month?"

Ask instead:

• Does the web clearly understand what this company does?
• Are the pages closest to revenue actually good?
• Does the business have anything original to say?
• Are its important claims supported anywhere other than its own website?
• Do customers describe the business the way the business describes itself?
• Does the content format match how people actually evaluate this kind of purchase?
• And are we funding the weak part of the structure, or feeding a channel because it's already on the calendar?

The house analogy isn't perfect. I keep using it because it forces the order of operations. Build what carries the weight. Then spend money making it louder, prettier and easier to find.

---

Google's guidance on AI features in Search is published at developers.google.com/search/docs/appearance/ai-features, last updated May 2026.