Stop ranking, start getting cited 🤌

First things first: This newsletter has a new name!

Welcome to Context Window, where I'll be sharing everything I'm learning about AI search. (Your girl is over here cooking on a website rebrand, and this is part of it.)

Today, we're going to talk about the fact that your content might rank on page one and still be invisible.

As you know, Google AI Overviews now answer queries directly at the top of the results page, pulling from multiple sources to construct answers, often without anyone having to click through to your site.

In fact, a Pew Research study tracking 68,000 real search queries found that users clicked on results just 8% of the time when AI summaries appeared, compared to 15% without them (a 47% relative reduction year-over-year).

For SaaS companies, this is a very big deal.

The good news: getting cited isn't (entirely) a black box.

Much of the work comes down to structure, originality, and what I call "source signals" that you can build into your content starting right now.

Here's what I've been telling my SaaS clients.

Shifting from keywords to citations

Stopping at traditional SEO tactics (keyword density, backlink volume) isn't enough anymore.

The goal isn't to "rank" for a keyword; it's to be cited as a trusted source inside an AI-generated answer.

Data backs this up. Seer Interactive analyzed 3,119 informational queries across 42 organizations and found that brands cited in AI Overviews earn 35% more organic clicks and 91% more paid clicks than uncited brands on the same queries.

That's the clearest argument for why getting cited (not just ranking) is the actual goal now.

Most SaaS content struggles here for two reasons:

1. It's too product-focused with marketing language instead of direct answers.

2. It's too generic and is just rehashing information found on 50 other sites.

Your content needs to be "answer-ready" for an AI model to cite it; not just optimized for a search crawler. It needs to provide value that can't easily be replicated by AI itself.

The irony here is that if AI could have written it, AI probably wouldn't cite it.

I know that sounds a little bleak, but it's also the clearest filter I've found for evaluating whether a piece of content is actually doing its job.

"Most SaaS companies are overthinking AI overviews and underinvesting in authority," says Daniel Horowitz, Enterprise SEO at Salesforce.

"The biggest shift I've seen is that AI systems reward clear, consistent signals across your entire ecosystem instead of isolated content. In my work, pages that perform well in AI overviews are supported and sit inside a broader system that reinforces what the company does and why it’s credible. If you’re trying to get cited, the focus should be less about using short-term tactics and more about making it easy for AI systems to understand, connect, and trust your content at scale."

5 tactics that help boost AI citation rates for SaaS companies

Getting cited in an AI Overview isn't random.

After working on this with SaaS clients across the last year, I've noticed the same patterns showing up again and again in content that gets pulled. Here are the five structural moves I’ve consistently seen make a difference.

1. Lead with the answer.

Format like a journalist and put the clearest, most concise answer in the first paragraph, directly under the heading. If your conclusion is buried after a long intro, you're making it harder for AI to find and cite you.

2. Use descriptive headings, not clever ones.

AI systems rely on heading tags to understand content hierarchy. "The Secret Sauce" tells an AI nothing. "How to Structure Content for AI Visibility" tells it everything.

I know clever headings are fun to write, but save them for the newsletter subject line. On the page, be boring and specific.

3. Answer the follow-up questions, too.

AI Overviews favor comprehensive content that addresses the initial query and the likely next questions. These are called “fan out queries.” This is a great use case for AirOps. I use it to build workflows that automatically surface related queries and subtopics around a target subject, so I can map the full question landscape before I start outlining.

Instead of manually combing through related searches, you can systematize that research step and make sure you're covering the angles that actually matter, not just the ones that are easy to find.

4. Add original data and/or expert perspectives.

This is the one most teams skip because it takes the most work. Include proprietary survey data, anonymized client examples, and quotes from SME interviews. Even a single original statistic can differentiate your content from dozens of generic competitors. This is the moat!

5. Refresh your high-value content regularly.

Freshness is a key relevance signal: Data shows 65% of AI overviews target content published in the past year. That "definitive guide" from 2023 you worked so hard on is being actively passed over in favor of newer pieces, even though the older content is technically higher quality.

Build a regular audit cadence into your editorial operations. I do this for clients at least quarterly.

The technical table stakes

Before any of the above matters, the foundational stuff needs to be right.

Crawlability: Check your robots.txt for accidental blocks and verify indexing in Search Console. Anything sitting behind a login wall or hard paywall won't get used. (You'd be surprised how often this is the culprit.)

Structured data: Schema markup helps Google understand the context of your content. It has to match what's visible on the page. Mismatches erode trust, and AI systems are increasingly good at catching them.

Page experience: Core Web Vitals matter. A slow, clunky, or ad-heavy page is less likely to be surfaced as a trusted source. This one often gets deprioritized because it lives on the engineering side, but it's worth having the conversation.

Visuals with alt text: AI Overviews are becoming multimodal. Relevant images with descriptive alt text give you an additional citation vector (low-effort and underutilized).

Building the authority that gets your brand named

On-page optimization is half the battle. “Source signals” are what determine whether your brand name actually appears in the AI answer or whether your information is used anonymously.

Three things I’ve seen *actually* move the needle here:

Topical depth. A single blog post isn't enough. You need a hub of interconnected articles (see also: topical ownership) covering a subject from multiple angles. This signals deep expertise to AI systems.
Third-party mentions. AI systems cross-reference sources, so being cited in industry publications, expert roundups, or news coverage significantly increases the likelihood that Google will name your brand in an answer rather than just use your information. If your digital PR strategy has been on the back burner, this is a good reason to dust it off.
E-E-A-T signals. Author bios with real credentials, cited sources, and first-hand case studies are how LLMs decide to trust you. Generic bylines and content that could've been written by anyone aren't going to cut it.

What this looks like in practice

I love this advice from Leigh McKenzie, Head of SEO & AI Search at Semrush:

“Google your core topics and actually look at the SERP. What's in the AI Overview? What sources is it pulling from? Brand sites, niche publishers, Reddit, YouTube? Then build a content strategy that goes deep on those topics and shows up across those channels. The companies winning in AI Overviews aren't doing one clever thing. They're impossible to ignore because they're everywhere that matters.”

The fix isn't to just write more blog posts.

It's to build the kind of deep, differentiated coverage that signals genuine expertise: a point-of-view-driven comparison page, original survey data from your customers, case studies with real numbers, participation in off-site forums, and a connected cluster of content that covers the topic from every meaningful angle.

Six months of building that, and you're the source AI keeps returning to. Why? Because you've done the work to deserve it.

A note on AI-assisted content

AI-assisted content can absolutely appear in AI Overviews, but only if it clears a quality threshold. The key is to use AI as a starting point, not a final product. Verify all claims, edit for voice consistency, and add original value on top of the draft (e.g., personal experience, expert insights, proprietary examples).

Seer Interactive research notes that they can't definitively prove that citations cause higher CTRs, by the way. It's equally possible that brands with stronger authority are simply more likely to be cited. What they can say with confidence is that queries where you're cited consistently outperform those where you're not, regardless of the causal direction.

Purely AI-generated content is unlikely to perform well here because it lacks the originality and E-E-A-T signals these systems are built to look for.

Think long-term about how you’ll tackle surfacing in AI overviews

Optimizing for AI Overviews is an ongoing process, not a one-time project.

I'll be honest: this stuff is still evolving, and anyone who tells you they have it completely figured out is selling something.

What I do know is that the teams sitting on the sidelines waiting for more certainty are already behind. Pick one section of your content, apply the tactics above, and see what moves.

That's how you learn what works...and right now, that's the whole game.

'Til next time (see you on LinkedIn),