The Two-Variable Scoring Model That Tells You Exactly Where to Spend Your Content Budget in 2026

Sixty-eight percent of sites with strong E-E-A-T signals gained rankings during Google's March 2026 core update, while 41% of AI-only content sites lost organic traffic. That single stat should reshape how every two-person B2B content team spends their next dollar.

But most teams aren't reshaping anything. They're still measuring traffic and publishing volume. Only 19% of teams using AI for content track AI-specific KPIs, which means 81% of programs are flying without instruments in airspace that changed overnight. We've watched this gap widen for months, and the economics are now clear enough to model precisely.

This post builds a scoring framework you can apply this week. Two variables, one budget reallocation, and a hard threshold below which new article spend is a waste of money.

What Actually Changed in March 2026

The March 2026 core update wasn't a tweak. AI Overviews coverage jumped from under 20% to 26-31% of queries during the March 27 through April 8 rollout. For B2B tech queries specifically, AI Overviews now appear above organic results for roughly 82% of searches.

That matters because AI Overviews have cut organic CTR by 61% on affected queries. A position-one ranking on a query with an AI Overview is worth roughly 40% of what it was worth eighteen months ago. And yet, getting cited inside that AI Overview now earns 35% more clicks than holding a traditional ranking alone.

So we have two populations of content now. Cited content, which benefits from AI's prominence. And uncited content, which gets squeezed from both sides: reduced organic clicks and zero overflow from AI summaries.

The gap between these two populations is not gradual. It's a cliff.

The Two-Variable Scoring Model

We've seen plenty of content scoring frameworks that try to rate everything on a 1-to-10 scale across twelve dimensions. They're useless. Nobody on a two-person team has time for that, and the precision is false anyway.

What we've found works is plotting content on exactly two axes.

Axis 1: E-E-A-T Signal Strength

This isn't subjective. You can audit it in under five minutes per article. Score each piece on four binary-ish factors, then sum them.

Experience evidence (0-3 points): Does the piece contain original data, first-person case studies, or screenshots of real results? A 2023 Semrush study rehashed without commentary scores 0. An article with your team's own A/B test data scores 3. Pages with evidence of real experience are 3.2x more likely to be cited by AI systems than purely theoretical content.

Author expertise (0-3 points): Is there a named author with verifiable credentials in the topic? A byline with a LinkedIn profile showing relevant work history scores 2. Add a point if that author has published on the topic across multiple platforms. Google's systems cross-reference this now, and 76.4% of ChatGPT citations come from content updated in the last 30 days, which suggests active expertise matters more than historical credentials.

Domain authority signals (0-2 points): Does the publishing domain already rank in the top 20 for related queries? 97% of sources cited in Google AI Overviews come from top-20 organic results. If your domain isn't there yet for the target topic cluster, you score 0 here. No shame in it; just math.

Trustworthiness markers (0-2 points): Third-party reviews, brand search volume, proper sourcing within the content, HTTPS (obviously), and structured data. Sites that performed well in the March update had stronger brand clicks and positive third-party credibility signals.

Maximum score: 10. We'll use this as Axis 1.

Axis 2: AI Citation Potential

This axis captures how likely AI systems are to pick up and reference your content. It's driven by structural and competitive factors.

Current organic position for target query (0-3 points): Already in the top 10? That's 3. Top 20? Score 2. Top 50? Score 1. Not ranking at all? Zero. 68% of AI Overview citations come from pages already ranking in the top 10, so this is the heaviest factor.

Content freshness (0-2 points): Updated within 30 days gets 2. Within 90 days gets 1. Older than that gets 0. The 76.4% ChatGPT citation stat for content updated in the last 30 days makes this non-negotiable.

Structural citability (0-3 points): Does the content contain clear, quotable definitions, numbered frameworks, or data tables that AI systems can extract cleanly? We've seen content with well-structured data tables get cited at 2x the rate of equivalent content in prose-only format. If your article buries its key insight in paragraph seven with no heading, AI systems skip it.

Topic citation concentration (0-2 points): The top 10 domains capture 46% of all ChatGPT citations on a given topic. If your domain is already one of those top 10, score 2. Top 30 (which captures 67%)? Score 1. Outside the top 30? Zero.

Maximum score: 10.

Plotting the Quadrants

With both scores in hand, every article in your portfolio falls into one of four zones.

High E-E-A-T / High Citation Potential (scores 7+ on both axes): These are your workhorses. Invest in keeping them fresh, adding new data, and expanding depth. They're already earning AI citations or are positioned to start. Budget priority: maximum.

High E-E-A-T / Low Citation Potential (E-E-A-T 7+, Citation under 5): Good content that isn't structurally visible to AI. The fix is usually formatting, freshness, or targeted link building to push organic rankings into the top 20. These are your best ROI opportunities because the hard part (building genuine authority) is already done.

Low E-E-A-T / High Citation Potential (E-E-A-T under 5, Citation 7+): This is rare, but it happens with older content on high-authority domains. The content ranks but lacks depth. AI systems might cite it today, but they'll drop it the moment a competitor publishes something better. Refresh aggressively or risk losing the position.

Low E-E-A-T / Low Citation Potential (both under 5): Stop spending money here. Genuinely. Every dollar spent producing new content in this quadrant is money you could redirect to quadrants one or two.

The Authority Threshold: Where Spend Becomes Indefensible

Here's the math that matters for a two-person team with a $3,000/month content budget.

If your domain's E-E-A-T score on a target topic cluster averages below 4 out of 10, and your citation potential score averages below 4, any new article you publish on that cluster has less than a 3% chance of being cited by AI systems. We're extrapolating from the concentration data: if 97% of AI Overview citations come from top-20 organic results, and your domain isn't in the top 50 for the topic, the probability of AI citation approaches zero.

An article that costs $200 to produce (writer time, editing, graphics) and has a 3% chance of earning AI citations has an expected value of about $6 in citation-driven traffic over six months. We're basing that on average B2B click values of $3-8 and the modest traffic volumes AI citations deliver to mid-authority domains.

That's indefensible. $200 in, $6 out.

The threshold we've identified: if your combined score across both axes is below 8 for a given topic cluster, do not publish new content in that cluster. Instead, redirect that budget to a cluster where your combined score is 12 or above.

The $3,000/Month Reallocation

For a two-person team, here's where the budget should go. This isn't theoretical. We've seen this allocation work.

$1,800/month on authority-building in one or two core topic clusters. That means original research (even small surveys of 50 customers count), bylined content from credentialed team members, guest contributions on higher-authority sites that link back, and active participation in industry conversations that generate brand signals. The Wellows study found pages with strong E-E-A-T signals are 2.3 times more likely to be cited in AI Overviews. This $1,800 is what builds that signal.

$900/month on refreshing and deepening existing content that scores above 8 on the combined axes. Update data, add new sections, improve structural citability. Freshness is the cheapest lever you have, and the citation data backs it.

$300/month for experiments in adjacent queries. Only if your core clusters are already above the authority threshold. If they're not, redirect this $300 to the first bucket. Experiments are a luxury until your foundation scores are solid.

Notice what's missing: no budget for new articles on topics where you don't already have authority signals. That's the hardest part for content teams to accept. It feels like giving up. It's not. It's concentrating force.

Why Volume-First Programs Are Bleeding Out

Content marketing teams that report pipeline contribution as a primary KPI outperform teams reporting traffic or engagement metrics by roughly 2.7x on year-two program growth. And yet the instinct is still to publish more. More keywords covered. More pages indexed. More chances to rank.

That instinct is wrong in 2026.

AI citation authority compounds. Each strong piece in a topic cluster reinforces the authority signals of every other piece in that cluster. Publishing ten thin articles on ten different topics gives you ten pieces with sub-threshold authority. Publishing three deep pieces on one topic, backed by original data and credentialed bylines, can push all three above the citation threshold within 60-90 days.

We are not saying volume doesn't matter ever. High-authority domains with strong E-E-A-T signals across broad topic areas can still benefit from volume plays. But those domains already have combined scores above 12 on most clusters. If you're a 30-person company with a DR of 45, you're not in that position, and pretending otherwise is expensive.

Tracking What Matters

The 19% stat is damning. If you're not tracking AI-specific KPIs, you're optimizing for a game that ended six months ago.

The minimum viable measurement stack for a two-person team in 2026 includes four metrics, and none of them are page views.

AI citation rate: What percentage of your published content appears in AI Overviews or is cited by ChatGPT, Perplexity, or Claude? Only 12% of surveyed B2B content teams track this. Tools like Semrush and Ahrefs now surface some of this data.

Content velocity per cluster: Not total output. Output per topic cluster. If you're spreading thin, this number will be high in aggregate but low per cluster, which is worse.

Cost per cited asset: Total content spend divided by the number of pieces that earned at least one AI citation in the measurement period. This is your real efficiency number.

Authority score trajectory per cluster: Month-over-month change in your combined two-axis score for each target cluster. If it's flat or declining, you're spending money without building anything.

The Uncomfortable Truth About 2026 Content Economics

Some topic clusters are simply not winnable for small teams. The concentration of AI citations among top domains means that entering a topic dominated by HubSpot, Gartner, or McKinsey with a $3,000/month budget is a losing bet. The math doesn't work.

That's genuinely messy, and we don't have a clean answer for it. The best advice we can give is this: find the clusters where the top-10 citation holders include at least three domains with authority profiles similar to yours. Those are your winnable battles. Everything else is aspiration, not strategy.

The two-variable scoring model isn't perfect. It compresses a lot of nuance into two numbers. But it forces a conversation that most content teams aren't having: "Should we publish this at all?" is a more valuable question in 2026 than "How do we rank this higher?"

Ask it before you spend another dollar.