Most advertising groups I discuss to are doing genuinely good search engine optimisation, and but once they open ChatGPT or Perplexity and kind within the prompts their consumers are literally utilizing, their model is nowhere to be discovered. That is the precise downside the FSA Framework was constructed to resolve.
For the final decade, standard knowledge has been, “Do good search engine optimisation, and the remainder takes care of itself.” That assumption was secure, and lots of manufacturers benefited from a well-executed search engine optimisation technique (good day, income!). But it surely doesn’t work anymore.
The mismatch isn’t as a result of search engine optimisation is damaged. search engine optimisation is doing precisely what it was designed to do. The issue is that search engines like google prioritize rating one of the best useful resource, and reply engines prioritize offering one of the best reply..
These are two very completely different machines, they usually reward two very various things.
Desk of Contents
What’s the FSA Framework?
The FSA Framework stands for Freshness, Construction, and Authority — the three indicators that answer engines actually evaluate when deciding which sources to cite inside a generated answer. It’s the diagnostic lens I use to figure out why a brand is or isn’t showing up in ChatGPT, Perplexity, Gemini, and Google’s AI Overviews, and what to fix first when they’re not.
Each pillar does a different job:
- Freshness determines whether your content gets reconsidered when new prompts come in.
- Structure determines whether a model can actually lift a clean answer out of your content.
- Authority determines whether the model comes back to your brand the next time a related prompt shows up.
Miss one, and the others can’t fully compensate. When all three are working together, your content stops being a candidate and starts being the obvious choice inside an AI-generated answer.
Where the FSA Framework Came From
In 2025, I started using my own website as a testing ground for answer engine optimization. I had a hunch about AEO, and no one was running the experiments I wanted to read. So, I ran them myself across ChatGPT, Perplexity, Gemini, and Google’s AI Overviews, tracking what surfaced for each prompt and — more importantly — what didn’t.
In one experiment, I up to date a single web page utilizing the ideas I’d been growing, and tracked AI Share of Voice throughout your complete window. The web page was on a subject the place Search Engine Journal — a legacy writer with the sort of area authority most entrepreneurs would kill for — had been the dominant cited supply for months.
Inside 96 hours, AI Share of Voice for Cassie Clark Advertising on that matter moved from round 27% to 72.7%. Search Engine Journal dropped to 0% visibility in the identical window. There have been no new backlinks and no promotional push. I simply had a better-structured, more energizing, extra extractable model of the identical thought.
Underneath conventional search engine optimisation logic, this could not have been attainable. A solo strategist’s website shouldn’t displace a legacy writer in 4 days. That doesn’t occur — particularly that rapidly — in conventional rankings.
However underneath AEO logic, it made excellent sense. The legacy web page had stopped being maintained, and its construction was constructed for crawlers, not for extraction.
Once I went again by means of each take a look at I’d run that 12 months, I seen engines had been usually skipping high-authority domains. As an alternative, they cited content material that was not too long ago up to date, cleanly structured, referenced persistently throughout a number of sources, and simple to raise into a solution.
Freshness, construction, authority. The identical three indicators, each time, throughout each mannequin.
Why We Want a New Framework within the First Place
Conventional search engine optimisation was constructed round a easy premise: A person varieties a question, the search engine identifies probably the most related pages, and people pages compete for place on a outcomes web page. Pages are the vacation spot, and the entire job of search engine optimisation is getting your vacation spot greater up the checklist than the following particular person’s.
That mannequin assumed two issues that reply engines not assume:
- The person needs an inventory of choices.
- The person will consider these choices themselves.
AI fashions don’t work that means. They retrieve info from a number of sources, synthesize it, and hand the person a single, assured reply. The person will get a abstract, not an inventory. And inside that abstract, sources are talked about, not as a reward for rating nicely however as proof that the reply will be trusted.
So the query the engine is asking has modified utterly. It’s not “which web page ought to we present?” It’s “which sources assist us clarify this clearly and precisely?”
That feels like a small distinction once you learn it on a web page, however in follow, it adjustments all the pieces about what your content material has to do so as to be helpful to the system. Your content material is not a vacation spot, however an enter.
And, when you internalize that shift, the FSA Framework stops feeling like a brand new set of techniques. It turns into the one logical response to how reply engines truly work.
Featured Useful resource: How AEO is altering the search panorama.
The FSA Framework Breakdown
Freshness
In AEO, freshness is a weight — one which influences how confidently a mannequin reuses your content material, how usually it will get reconsidered when new prompts are available, and whether or not it stays eligible to seem in assembled solutions in any respect. Stale content material will get dropped from the candidate pool fully.
The way in which I give it some thought is that this: Freshness is recency, relevance, and reinforcement.
- Recency is the time-based piece. When was this final touched?
- Relevance is contextual. Does this nonetheless match how the subject is definitely mentioned immediately with the language persons are truly utilizing?
- Reinforcement is behavioral. Has this supply continued to indicate up, get cited, and maintain its place over time?
All three feed the identical sign, and a web page can fail on any considered one of them and lose floor.
What Freshness Actually Means
Reply engines don’t want a “final up to date” badge to find out whether or not content material is present. As an alternative, they discover when the language doesn’t match how a subject is being mentioned now, once you reference a instrument that doesn’t exist anymore, or when the encircling matter house has developed previous what your web page is describing.
In fast-moving verticals — SaaS, AI, fintech — content material has roughly a 90-day shelf life earlier than it begins shedding relevance indicators. For extra evergreen subjects, you will have nearer to 6 months. After that, you threat falling out of the reply pool fully.
The sensible takeaway is straightforward:
- Don’t simply replace the date.
- Add a present instance.
- Pull in a current stat.
- Reference one thing that’s truly modified within the house.
The amount of updates issues means lower than their consistency and their substance. One actual replace each quarter beats 5 beauty adjustments a month.
Freshness will get your content material reconsidered, however getting reconsidered isn’t sufficient by itself. The mannequin nonetheless has to have the ability to use what it finds.
Construction
Construction for AI is completely different from construction for crawlers, and the 2 don’t all the time align.
AI fashions don’t learn your web page the way in which people do. They parse it and scan for clear hierarchies, self-contained explanations, and clearly labeled sections they’ll raise into a solution while not having the remainder of the web page to make sense.
Content material that performs nicely in AI solutions shares numerous the identical structural traits:
- Clear H2s and H3s.
- Brief paragraphs that resolve one thought at a time.
- Specific definitions close to the highest of a bit, earlier than the reason unfolds.
- Labeled steps.
- FAQ sections.
- Callouts.
In case your greatest thought is buried three paragraphs into a bit that requires the earlier part to observe, the mannequin goes to skip it. Not as a result of it’s a foul thought, however as a result of it may possibly’t be extracted cleanly.
Why Structuring for Reply Engines is Completely different From Conventional search engine optimisation
In case your content material forces the mannequin to do interpretive work, the mannequin will discover one thing structured in a means that’s simpler to interrupt aside.
The error I see most frequently is groups optimizing construction for crawlers — meta tags, clear header hierarchy, inner hyperlinks — and assuming that’s the identical job. It’s not. Crawler construction focuses on navigability, whereas AI construction prioritizes extractability.
The correct query to ask of any web page is: Can ChatGPT raise a clear, correct reply out of this while not having the remainder of the web page?
If the reply is not any, you will have a construction downside, regardless of how nicely your headings are nested.
Authority
In search engine optimisation, authority meant area authority. It took years to construct and was virtually not possible to displace as soon as a model had it. Whole company enterprise fashions had been constructed round hyperlink acquisition.
In AEO, authority is now entity authority. The query isn’t “how robust is that this area?” It’s “is that this model the one which persistently explains this particular matter, throughout each channel I can discover them on?”
Entity authority will get constructed one point out at a time, in a means that has virtually nothing to do with backlinks. Each time your model seems someplace a mannequin can study from — a podcast, a Reddit thread, a visitor put up, a quote in a third-party article, a LinkedIn put up, your individual web site — it provides to what the mannequin is aware of about you.
One point out is a knowledge level. However repeated mentions in comparable contexts throughout a number of channels assist construct a sample and create mannequin confidence. Confidence is what will get you cited.
Why Smaller Manufacturers Have Robust Entity Authority
Inside AI solutions, smaller manufacturers are all of the sudden successful fights they don’t have any enterprise successful on paper. Digging deeper, the rationale why is apparent.
Smaller manufacturers usually create content material just for their core viewers and depend on social media or influencer advertising to construct model authority throughout surfaces, not simply their very own web site. When a mannequin encounters these manufacturers repeatedly, it features confidence in reusing the reason.
The huge writer, against this, has 100 contributors writing about all the pieces. None of them is constructing a recognizable entity round a particular, user-focused matter. Distribution is commonly nonexistent as a result of conventional search engine optimisation knowledge says that area authority is sufficient. When this occurs, the mannequin has nothing to anchor to.
Authority work is now nearer to popularity administration throughout channels than hyperlink constructing. None of this appears like an search engine optimisation marketing campaign, but it surely’s precisely the way you turn out to be the model the mannequin acknowledges.
The way to Apply the FSA Framework
So if that is how reply engines truly work underneath the hood, the following query is: What ought to groups be doing in a different way to place the FSA Framework to work?
Right here’s the way in which I body it for purchasers. search engine optimisation will get you into the room. AEO will get you chosen when you’re there. Right here’s learn how to apply the FSA framework in follow.
1. Begin with an audit — and discover your cash prompts
Earlier than you contact a single web page, you must audit your visibility to know where you actually stand inside AI answers. That means running real prompts in ChatGPT, Perplexity, and Gemini for the topics tied to your pipeline — not the topics tied to your keyword list.
These are your money prompts. Think about the questions your buyers are actually typing when they’re evaluating a solution, comparing options, or trying to figure out if you’re the right fit. They usually sound like:
- “Best AEO tool for [specific use case]”
- “[Your brand] vs. [competitor] for [buyer context]”
- “How do I [solve the problem your product solves] as a [your ICP]”
- “What should I look for in a AEO tool if [specific constraint]”
Run your money prompts across multiple engines and pay close attention to whether your brand shows up at all, who’s showing up instead, and what the AI-generated answer actually says about your space. That single exercise will tell you more about your real AI visibility than any keyword report.
Pro Tip: You can measure mentions with HubSpot AEO — monitor prompts throughout ChatGPT, Perplexity, and Gemini, and see precisely the place your model stands.
When you’ve achieved the preliminary scan, audit your prime 5 pages by means of the FSA lens with an sincere eye towards the place every pillar is or isn’t holding up:
- Is the content material present and reflecting how the subject is being mentioned immediately, or is it quietly growing old out of relevance?
- Is it structured in a means {that a} language mannequin might raise a clear reply out of the primary few hundred phrases?
- Is your model persistently represented throughout the channels the place consumers in your house are literally paying consideration? Or are you primarily invisible in all places besides your individual area?
Prognosis earlier than techniques, each single time.
2. Exchange quantity targets with refresh targets
Sustaining and updating current content material on a constant cadence does extra for AI visibility than publishing net-new content material each week. In case your editorial calendar is constructed round what number of posts you ship, rebuild it round what number of of your top-performing pages get meaningfully refreshed every month.
3. Construction for extraction, not simply indexing
Audit your prime pages with one query in thoughts: Can a mannequin raise a clear, full reply out of the primary few hundred phrases?
If not, restructure with:
- Definitions up prime.
- Labeled sections.
- FAQ blocks.
- Comparability language for prompts the place consumers consider you in opposition to alternate options.
4. Construct entity authority throughout channels
Your web site alone isn’t doing all of the work anymore. Reply engines study from content material diversification, that means:
- Podcast appearances.
- LinkedIn firm and worker content material.
- Reddit feedback and threads.
- Visitor articles.
- Professional quotes.
- Group participation.
The manufacturers that construct a constant presence throughout a number of surfaces are those fashions begin to belief.
5. Measure AI Share of Voice, not simply rankings
AI Share of Voice tracks how usually your model seems inside AI-generated solutions in comparison with competing sources. It’s a zero-sum metric — when one model features share, one other model loses it.
HubSpot’s AEO options now allow you to see how your model is exhibiting up throughout reply engines and the place rivals are being cited as a substitute — which is genuinely helpful as a place to begin, since most groups don’t know the place their gaps are till they’ll see the information.
6. Choose one pillar to repair first
As soon as you understand the place you stand, decide one pillar to repair first reasonably than attempting to handle all three directly:
- In case your content material is stale, begin with freshness. That’s the quickest sign to maneuver.
- In case your content material is complete however dense, restructure for extractability.
- In case your model is invisible regardless of having genuinely good content material, the issue is sort of actually entity authority, and the repair lives outdoors your web site.
Most AI visibility issues fall cleanly into a type of three buckets. Loads of what appears like a visibility downside is definitely an authority downside in disguise.
Professional tip: Pair the FSA Framework with these AEO greatest practices for a extra complete method.
What This Means for Your Content material Technique
The FSA Framework is a diagnostic lens for determining why visibility is or isn’t taking place in your model inside AI solutions. You’ll be able to cease guessing and begin engaged on the precise factor in the precise order.
The particular indicators reply engines weigh will change because the fashions evolve. The techniques constructed on prime of the framework will have to be adjusted because the surfaces shift. However the underlying logic — favor freshness, reward readability, belief consistency — has held regular throughout each mannequin I’ve examined, and I count on it to proceed to carry because the engines evolve.
The manufacturers that win inside AI solutions over the following few years aren’t going to be those chasing each new tactic. They’re going to be those who perceive how AEO truly works, diagnose their visibility gaps actually, and repair the precise pillar first.
Construct on these ideas, and the FSA Framework adapts because the floor adjustments.
