AI made content a commodity, but you can harness expertise to stand out

Generative AI was supposed to free content marketing teams to do their best work. It promised more capacity and less busywork, which is a beautiful idea in theory. Instead, teams often end up using that new cranial real estate to publish more of the same, just faster.

The result is a web flooded with AI-generated output that doesn’t rank, has no meaningful impact, and is largely indistinguishable from the countless other pieces of content that cover the same ground.

The brands that successfully incorporate AI into content workflows recognize how to use AI to handle commodity content while humans provide the expertise.

In this article, we’ll walk through how to use AI to build a winning content workflow — so you can maximize the value of human expertise.

What is commodity content? 

Commodity content is any piece of content that’s interchangeable across brands or publishers. It includes the same points, structure, subheadings, and ideas as other published pieces. There’s no nuance, new perspective, or tangible long-term value.

Think about it like pre-shredded mozzarella cheese. It’s widely available, made by a bunch of different manufacturers, and essentially all the same at a certain price point. Nobody’s driving across town for a specific brand because every grocery store carries some acceptable version of it that gets the job done without leaving a lasting impression 

But say you want the really good stuff. You’re probably going to that well-loved local spot that makes fresh mozzarella by hand every day. They offer something unique that stands apart from the generic, mass-produced product. Once you’ve tried it, you crave the quality and will go out of your way to get it again. 

The same concept applies to content. Think of AI-generated content like pre-shredded mozzarella. As a commodity, it’s widely available and indistinguishable from site to site. Its handmade equivalent is expertise-driven content. This content stands out thanks to the original data and first-hand experience it contains. It’s memorable in the way you want your brand to be.

The current state of AI in content creation

At first, AI tools helped teams publish content faster. Now, some important shifts are changing how both publishers and audiences use AI.

AI usage in content creation

According to the HubSpot 2026 State of Marketing report, the vast majority of marketers now use AI in some part of their workflow.

Four in five marketers use AI for content creation, indicating it’s now a standard part of the workflow. Any head start it gave to early movers has since evaporated. Almost everyone else has caught up. 

Pressure to produce more

Of the marketers surveyed, 83% say they’re expected to produce more than before, showing that pressure from leadership keeps climbing. Nearly three-quarters credit AI with helping them produce significantly more content. But that same group reports their content is struggling to perform. 

More is not always better

Over 50% of marketers say the ease of AI production has made content weaker across the board. A similar percentage (56%) say they can’t get their content to stand out in a crowded market. Saturation used to be a problem for specific niches or keyword sets, but marketers now describe it as the default state for any category.

The study highlights a potential fix. Over 60% of surveyed marketers say they want more original, people-driven content to cut through the noise — the kind of work AI can’t generate on its own.

Building for volume ignores quality concerns

But identifying the fix and executing on it are different challenges. Restructuring a team built for volume into one that produces original work is difficult.

So for now, the pattern holds across the board. Teams produce more commodity content than they used to, under more pressure than ever, without achieving the desired result.

Zero-click stats

As content volume is increasing, opportunities for search traffic are shrinking, because zero-click searches are on the rise. In the first four months of 2026, only about 32% of Google searches led to a click. This is down from roughly 40% in 2024, according to data reported by Search Engine Land.

This decline has had the greatest impact on informational content, since that’s the type of content AI Overviews summarize best. In January 2025, 91.3% of searches that resulted in an AI Overview were informational queries, according to Semrush.

But AI Overviews haven’t wiped out clicks across the board. When Semrush tracked the same keywords before and after an AI Overview appeared, the zero-click rate barely moved, slipping from 33.75% to 31.53%. Many of those search engine results pages (SERPs) weren’t producing clicks, with or without AI Overviews. People were already getting what they needed from the SERP and moving on.

The payoff for ranking for those terms is eroding. But that doesn’t mean you should stop making informational content. Instead, reorient around the current state of zero-click search.

A citation in an AI Overview or an AI engine answer puts your brand in front of someone at the moment they’re looking, even with no click. It won’t drive traffic the way it once did. So you measure the outcome by visibility, not clicks.

AI and LLM conversion rates

It’s logical to expect traffic from AI engines like ChatGPT to convert at a higher rate than organic search. By the time someone clicks through from an AI answer, the model has already walked them through their options in a conversational back-and-forth.

This means they typically land on your site further along in their customer journey, which means they’re more qualified. In fact, the average AI search visitor is 4.4 times as valuable as one from organic search, according to Semrush.

But that edge doesn’t always apply. An Amsive study found that LLM traffic converted at 4.87%, barely ahead of organic search at 4.6%. Higher conversion rates seem to apply to buyers doing middle-of-funnel research as they weigh their options, not across the board. 

No matter which stage of the customer journey your website attracts, volume is an issue. LLM referrals are still a tiny fraction of total traffic compared to organic search. AI traffic is a real channel worth tracking, but it isn’t a high-volume traffic source on its own yet. 

By making your content visible across AI engines, you can capture the upside. And by continuing to invest in organic search, you can maximize clicks where the search volume is. You need both.

The trouble with commodity content

Commodity content has value. It provides thorough topical coverage for humans and search engines alike. For example, explainer content and how-to articles are key parts of many content libraries. They help search engines understand your topical authority and give readers the foundational answers they need.

The trouble starts when brands overindex on commodity content. Build your whole strategy around it, and weaknesses like these start to show:

• Zero information gain: Recycled facts add little value when search engines and LLMs can synthesize this information on demand. Content that brings nothing new to the topic quickly becomes redundant. 
• Interchangeability: If your article sounds exactly like multiple others on the same topic, you’ve produced a commodity. When the piece includes nothing that’s distinctive to your brand, it gives users no reason to click on your content over the next result. 
• Zero-click obsolescence: AI engines surface basic answers instantly via definitions, listicles, or how-to summaries. Content that restates what’s already accessible gives users no incentive to click deeper. 
• Hallucination risk: Generative AI tools write with total confidence, whether or not what they’re saying is correct. High-volume commodity content is vulnerable to invented stats, inaccurate sourcing, and factual errors. If you skip human verification, you risk those mistakes going live and causing reputational damage. 
• Algorithm vulnerability: Commodity content has little to fall back on when Google updates its algorithm. If the content never built much authority or earned backlinks, not much keeps its rankings from slipping.

None of this makes commodity content a waste of effort. You have to create some of it, and AI can help scale production. Treating it as the centerpiece of your strategy is a mistake, though. The majority of your team’s effort should go toward non-commodity content: expertise-driven work that existing sources can’t synthesize.

How do you make content stand out when everyone is using AI?

Don’t omit AI from your content workflow. Instead, use it for what it’s good for, like creating outlines, writing meta descriptions, or even generating first drafts. But remember that expertise is what makes your content stand out.

When all content becomes a commodity, expertise wins

Placing expertise at the center of your content strategy has always been the play. It creates more compelling content. An article written by someone who’s done the work has a specificity and conviction that an AI-generated summary can’t fake.

Readers can feel the difference, and so can the search engines and AI models that decide what to surface. 

Two-thirds (65%) of B2B marketers consider the relevance and quality of their content key to improving outcomes, according to the Content Marketing Institute. They rank these qualities well ahead of budget, tools, or market conditions. 

Here’s how to make expertise the center of your content strategy:

• Assign content to a named expert with real credentials in the subject — or have one substantively review and edit. This kind of byline signals that someone with authority is behind the work. This is exactly the kind of credibility readers and AI engines look for in a source.
• Build in first-person accounts that reflect real experience in your field. LLMs can’t fabricate lived details.
• Gather quotes from practitioners that add depth the piece couldn’t achieve without them. This gives readers and AI systems a reason to treat your content as new information rather than a summary of existing content.

By focusing on expertise, you elevate your most important work above commodity content. 

Incorporate original research, data, and statistics 

There’s a higher premium than ever right now on original data. Search engines and LLMs have already indexed most of what’s been published across the web. That means repurposing the same facts from a long-published study won’t be as citable as something new. 

Marketers are moving in the same direction. Many are shifting budgets away from generic production and toward proprietary research. In fact, 86% of B2B SaaS marketers plan to put more resources behind research in the next year, according to the Datalily 2026 State of Data-Driven Content Marketing report. Teams that lean on proprietary research are more likely to see leads convert and organic traffic climb, with 64% reporting stronger conversion rates and 61% reporting a lift in organic traffic.

For buyers, original statistics and concrete figures offer proof and credibility. Data can increase trust in your brand or move buyers closer to a decision.

Compelling stats can also lead to citations from other writers, links from journalists, and references from thought leaders. Every one of those instances creates a new path back to your brand.

Original research can often stretch further than a single content asset. One research initiative can anchor a full report. From there, you can distribute it across blog posts, social media posts, and webinars.

In many cases, original research is easier to collect than teams might assume. You don’t need a formal study or a big budget, just a question worth answering. Simpler data collection methods include:

• A survey of your customer base on a trending topic
• A proprietary analysis of your own product or usage data
• A fresh analysis of public data that surfaces a finding no one’s reported
• An aggregation of firsthand observations from your team or network

Whatever method you choose, point the research at questions your audience is already asking. Data that answers real questions your audience has often gets pulled into AI Overviews and AI answers, with your brand cited as the source.

How to create compelling content: A complete workflow

Creating content that stands out in this market will look a little different depending on your industry, brand, and resources. But the basic framework (audit, assign, rebuild) holds up regardless. 

Phase 1: Plan and audit

Auditing your content once won’t solve the production problem in perpetuity. Without changes to how you brief, source, and review, there’s nothing stopping you from drifting back to a library full of commodity content. These are the basic steps you can take to ensure that doesn’t happen. 

1.1: Restructure your briefs by track

Create different briefs for each content track:

• Commodity briefs: Optimize for structure. Include the target keyword, search intent, preferred format, and brand guidelines. Make it possible for AI to handle most of the drafting. 
• Hybrid briefs: Optimize for judgment. Clarify the audience, evaluation criteria, examples, and brand perspective. Guide the decisions AI can’t make on its own.
• Expertise briefs: Optimize for original knowledge. Identify the subject matter expert (SME), document relevant firsthand experience or proprietary data, define the unique angle, and specify the original contribution the piece needs to make.

Commodity and hybrid briefs are relatively straightforward to create. But expertise briefs will require the biggest overhaul. They’re responsible for capturing information that doesn’t exist in AI training data. 

Building that kind of brief usually requires a conversation with an SME.

1.2: Build your SME pipeline

Consistently sourcing SMEs can be challenging. Use this workflow to make the process simpler.

• Identify and map your experts. Review the topics you want to own, and then find people who can credibly speak on them. For example, you might recruit product managers, support leads, salespeople, engineers, or executives with a point of view. Tag them by topic so you know who to ask for what.
• Set a cadence that aligns with your content calendar. For experts who anchor your highest-priority topics, consider setting up a standing monthly or quarterly block. Doing this removes the negotiation every time you need something. For other SMEs, schedule a quick conversation before drafting a relevant piece. 
• Treat SME interactions like interviews, not writing assignments. Invite SMEs to share their perspective and knowledge, while you transcribe the conversations. Bank the material, as one in-depth interview may contain enough material for several pieces. Capture it once, and mine it over and over again.

The goal is to make contributing insights frictionless for the expert and useful for the writer.

1.3: Audit your content library

First, clarify what you’re publishing and how much of it needs human input.

Start by inventorying your content library. Pull together every piece of content across every format, including blog posts, guides, landing pages, videos, thought leadership, case studies, and newsletters.

Sort each piece into one of three tracks:

• Commodity: Content with primarily established, widely available information. These are foundational pieces you don’t want to skip but they probably won’t reinvent the wheel. Think glossary entries, FAQ pages, and basic how-tos. 
• Expertise: Content where the value comes from something an AI model can’t recreate, such as first-person experience, proprietary data, or a unique point of view. Think original research, first-person case studies, and nuanced opinion pieces. 
• Hybrid: Content that needs human judgement but not lived experience. Think buying guides, comparison pages, and category explainers.

Then, pressure-test your sorting process.

If you’d be comfortable publishing a piece without substantial human edits, it’s commodity content. If it falls apart without proprietary research, a first-person point of view, or deep industry knowledge, it’s expertise content. And if a human has to edit it substantially but doesn’t need to have lived it, it’s hybrid content. 

As you sort, note roughly how much human time each piece takes to produce. Then, tally the effort by track. By the end, you’ll see exactly how your team’s time maps against these three tracks.

Phase 2: Assign and create

Now it’s time to route each piece to the appropriate production workflow and allocate resources accordingly. 

2.1: Commodity track

On the commodity track, AI generates the content and a human verifies before publication.

This shifts the human role from creator to editor — confirming the facts are right, the structure holds, and the piece aligns with your brand standards. The content requires a real review, but the total human time is much less than what other tracks need.

Take glossary entries, for example. An AI model can produce definitional content in seconds. This is one area where you claw back time and redirect it toward work that moves the needle. 

2.2: Hybrid track

With hybrid content, AI can produce structure and a working draft. But this content requires substantial editing. A human needs to shape it using judgment, relevant examples, brand perspective, and voice.

Here are criteria that signals content fits cleanly into the hybrid track:

• The underlying information is publicly available
• AI can gather and organize it competently
• A human must decide what’s relevant, how to frame it, or what recommendations make sense for a certain audience
• The value doesn’t hinge on personal experience or developing proprietary data

Take an article on the best CRM software for small law firms for example. Generative AI can gather features, pricing, reviews, and integrations. It can also create comparison tables and summarize strengths and weaknesses based on publicly available reviews. 

But you need human judgment to decide which criteria matter most to law firms. A human should also recommend which tools belong at the top and why. Their editorial decisions are a key factor in the content quality.

2.3: Expertise track

Put any reclaimed time toward the expertise track, which requires first-person experience, original research, and practitioner insights. AI can support the research and even help organize the structure. But the soul of the piece has to come from an expert. 

Here are criteria that signal content belongs in the expertise track:

• The most valuable information comes from direct experience, proprietary data, or original observations
• AI can’t independently generate the core insights because they haven’t been published yet
• The value comes from revealing a new insight rather than reorganizing information that’s already widely available

For example, take a breakdown of how a law firm increased qualified consultations by 45% over six months. Generative AI can assist in organizing the case study structure and summarizing interview notes and performance data. It can also draft sections based on original data and suggest visualizations or headlines. 

But you need human expertise to share mistakes, false starts, and lessons learned through the process. An expert should also interpret results and provide context from their involvement in the work. 

Phase 3: Review and publish

The work isn’t done when a draft is. Phase three is where you decide what’s good enough to publish and how hard to push it once it is. Both decisions scale with the track: The more human effort a piece required going in, the more scrutiny and promotion it earns coming out.

3.1: Establish editorial checkpoints

Every track should pass through human review before publishing.

• Commodity: Confirm accuracy with an AI quality assurance workflow. Check that the facts are right, the structure holds up, the links work, and the content aligns with your brand.
• Hybrid: Confirm the judgment is sound. Verify that the recommendations, examples, and framing make sense for the audience.
• Expertise: Confirm the human contribution provides added value. Check that the firsthand experience, original data, or unique perspective makes the piece worth reading.

Setting this up takes some upfront work, but it’s mostly a one-time cost. Once you’ve built the tracks, briefs, and checkpoints, the day-to-day gets simpler. And more of your team’s hours can go toward impactful work.

3.2: Publish to the appropriate channels 

After a piece passes editorial checkpoints, give it as much promotion as the track calls for. Some pieces need a real push, while others just need to go live.

• Commodity: Publish and move on. These pages exist to cover a topic and answer a basic query, so they don’t need a formal launch. 
• Hybrid: Give these a modest push. A buying guide or comparison page can pull real traffic, so it’s often worth a spot in your newsletter or a social post. 
• Expertise: Spend most of your promotion budget here. Original research and thought leadership can earn links, citations, shares, and mentions — but only if people see the content. Pitch it, build a landing page around it, and repurpose it into social and email initiatives. 

This logic mirrors the rest of the workflow. The less effort a piece needs from a human during production, the less it needs after the content. When the content involves more expertise, it’s worth amplifying. 

Reclaim the hours, then spend them well

AI can free your team members to do their best work, but you have to set up the production workflow first. It starts with seeing your content clearly — separating commodity from expertise and identifying where your best people are spending their hours. 

That’s where the right tooling helps. Semrush’s AI Visibility Toolkit can show you which of your pages get cited in AI Overviews and LLM, and which ones are invisible. Start there, and then redirect your efforts toward the expertise content that earns mentions.