AEO has gone mainstream. Maturity has not.

Answer engine optimisation has moved from emerging idea to mainstream marketing activity with remarkable speed. The stronger finding, however, is not simply that adoption has grown. It is that adoption has run far ahead of maturity.

HubSpot says 58% of marketers report that their businesses are already optimising content for answer engines. Minuttia and Growth Memo found that 82% of the SEO, content and growth professionals they surveyed had implemented AEO to some degree. Semrush found that 85% of marketers, business owners and SEO professionals said AI had changed their search strategy.[1] [2] [3]

Those figures make AEO sound almost universal.

Look further into the same evidence and a second story appears. Only 22% of Semrush respondents said SEO and AI search were fully integrated across strategy, execution and reporting. Just 9% said they could measure all the metrics that matter. Minuttia found that 9.2% had a dedicated AEO specialist or team. When Webflow assessed 2,000 live US company websites rather than asking marketers what they were doing, the median company's overall AEO maturity was 2 out of 5.[2] [3] [4]

The studies are not contradictory. They are observing different levels of the same diffusion process.

<blockquote>AEO is no longer struggling for adoption. Organisations are struggling to turn adoption into a repeatable capability.</blockquote>

<h2>The best working estimate is approximately 60%</h2>

There is no representative census of AEO adoption. The available studies use different populations, questions and thresholds.

HubSpot's 2026 State of AEO report provides the broadest direct adoption anchor. Based on insights from more than 4,000 global marketers, it reports that 58% say their businesses are optimising content for answer engines.[1]

Minuttia's 82% figure comes from a different population. Its 599 respondents were SEO, content, growth and related professionals, with particularly strong representation from technology companies, agencies and consultants. These are people who would reasonably be expected to adopt AEO earlier than the general marketing population.[2]

Its headline also combines very different levels of activity. A quarter of respondents were running experiments or pilots. Another 21.5% were actively implementing a strategy. For 26.9%, AI search was already a core part of SEO. Only 7.7% had a dedicated AEO programme or team.[2]

Semrush's 85% asks a broader question again. It measures whether AI changed search strategy, not whether an organisation has built an intentional AEO programme. Only 32% described the change as significant, while 53% reported some adjustments.[3]

The most responsible interpretation is therefore not to average 58%, 82% and 85%. They do not share a denominator or an adoption threshold.

AEO Updates' working estimate is that <strong>approximately 55–65% of marketing organisations are undertaking some intentional AEO activity</strong>. If one central planning figure is needed, <strong>60%</strong> is defensible.

That range is anchored primarily to HubSpot's 58% broad-marketer finding. Minuttia and Semrush act as contextual upper signals. It is a judgement band, not a statistical confidence interval or formal meta-analysis.

<h2>Five studies are measuring five different things</h2>

<div class="overflow-x-auto"><table><thead><tr><th>Evidence set</th><th>What it measured</th><th>Finding used here</th><th>What it cannot establish</th></tr></thead><tbody><tr><td>HubSpot</td><td>Self-reported activity among 4,000-plus global marketers</td><td>58% say their businesses optimise content for answer engines</td><td>Depth, consistency or operational maturity</td></tr><tr><td>Minuttia and Growth Memo</td><td>Self-reported stages among 599 SEO, content and growth professionals</td><td>82% report some implementation, from pilots to dedicated programmes</td><td>General-market prevalence</td></tr><tr><td>Semrush</td><td>Strategy, workflow, ownership and measurement among 481 retained respondents</td><td>85% report strategy change; 22% report full integration</td><td>Objective maturity or causation</td></tr><tr><td>Webflow</td><td>Proprietary assessment of 2,000 live US company websites and their visibility across four answer engines</td><td>Median overall maturity of 2/5</td><td>Self-reported organisational adoption or a universal maturity standard</td></tr><tr><td>McKinsey</td><td>September 2025 survey of approximately 30 Fortune 500 consumer-brand CMOs</td><td>16% systematically tracked AI-search performance</td><td>Broader market prevalence or a directly comparable 2026 adoption rate</td></tr></tbody></table></div>

The table explains why the headline figures can coexist. One source measures stated activity. Another measures strategy change. Another scores observable website and answer-engine conditions. A small earlier survey measures systematic tracking inside a specific group of large consumer brands.

They should not be pooled. They can still support a common directional conclusion.

<h2>“Doing AEO” covers an enormous maturity range</h2>

A marketer who rewrites ten articles to answer questions more directly can legitimately say the organisation is doing AEO.

So can a company with a dedicated owner, repeatable prompt research, multi-engine monitoring, evidence mapping, coordinated SEO and PR, structured product facts, and a measurement system connected to pipeline.

Those organisations are not at the same stage.

Treating adoption as a binary measure hides the difference between experimentation and institutional capability. A maturity model is more useful, provided its estimates are labelled honestly.

<div class="overflow-x-auto"><table><thead><tr><th>Maturity stage</th><th>AEO Updates working estimate</th><th>Primary evidence anchor</th><th>Evidence status</th></tr></thead><tbody><tr><td>Aware or responding to AI search</td><td>~85%</td><td>Semrush strategy-change question</td><td>Observed in one vendor survey</td></tr><tr><td>Doing some intentional AEO</td><td>~60%</td><td>HubSpot 58% plus cross-study context</td><td>Editorial synthesis</td></tr><tr><td>Active implementation</td><td>~40%</td><td>Minuttia stage distribution and Semrush workflow evidence</td><td>Editorial synthesis</td></tr><tr><td>Systematic measurement</td><td>~30–35%</td><td>Semrush tracking and consistency findings</td><td>Editorial synthesis</td></tr><tr><td>Integrated operating model</td><td>~20–25%</td><td>Semrush 22% full-integration finding</td><td>Anchored interpretation</td></tr><tr><td>Dedicated AEO capability</td><td>~10%</td><td>Minuttia 7.7% and 9.2% findings</td><td>Editorial synthesis</td></tr><tr><td>Mature cross-functional AEO</td><td>Under 10%</td><td>No direct prevalence study</td><td>Working hypothesis</td></tr></tbody></table></div>

These stages are not measured by one instrument. They are not necessarily nested, and organisations may not move through them in a fixed order. Uncertainty increases toward the bottom of the model.

The shape of the curve is more defensible than any individual percentage. Awareness and initial activity are widespread. Integration, measurement and cross-functional execution are much less common.

<h2>The first wave of AEO is mostly content</h2>

Semrush provides one of the clearest pictures of what early adoption looks like. Among the tactics respondents used to improve AI visibility, 54% were creating structured, well-organised content, 43% were improving product and service pages, and 37% were publishing more authoritative content.[3]

This is a logical entry point. SEO and content teams already have publishing systems, search expertise and the authority to change owned pages. Existing material can be made clearer. Answers can become more explicit. Product facts can be organised. Comparisons and question-led formats can be added.

Ownership patterns reflect the same path. Minuttia found that 52.8% of organisations placed AEO responsibility with the existing SEO team, while 9.2% reported a dedicated specialist or team.[2]

The first large wave of AEO is therefore not an entirely new discipline. It is largely existing search and content organisations adapting to a new discovery environment.

That is a rational beginning. It is not the complete operating model.

<h2>Measurement is spreading, but the methods remain immature</h2>

Once organisations realise that AI systems are describing and recommending their brands, they naturally ask how they are showing up.

Semrush found that 40% of respondents still relied on manual ChatGPT checks as their main tracking method. Thirty-eight per cent used a traditional SEO platform, 36% used a specialised AEO or GEO tool, and 13% had no consistent approach.[3]

These categories may overlap. Tool use is not the same as systematic measurement. A company can own an AEO platform and still lack a stable prompt set, disclosure of its <a href="/articles/behind-aeo-dashboard-ai-search-measurement">measurement frame and calculation</a>, or connection to business outcomes.

Only 9% of Semrush respondents said they could measure all the metrics that matter. Nearly half identified AI's impact on pipeline or revenue as a major challenge.[3]

McKinsey offers a useful historical reference. In September 2025, approximately 16% of a small survey of roughly 30 Fortune 500 consumer-brand CMOs said their brands systematically tracked AI-search performance.[5]

That figure should not be treated as a general 2025 baseline or compared directly with the 2026 tool-use percentages. It does, however, show that systematic AI-search measurement was still unusual in that specific large-brand cohort.

The category is moving quickly. Measurement adoption is spreading. Measurement discipline is still catching up.

<h2>Live websites provide a useful reality check</h2>

Self-reported surveys reveal what marketers believe they are doing. Webflow examined observable conditions instead.

Its July 2026 study assessed 2,000 live US company websites across 123 scoring gates and four broad dimensions: content, technical readiness, authority and measurement. It also queried ChatGPT, Gemini, Claude and Perplexity using buyer-style questions.[4]

The median company scored 2 out of 5 overall. Content and technical maturity each scored 1/5, authority 2/5 and measurement 3/5. The median company appeared in 16% of responses and was cited with a link in 6%.[4]

Webflow's scale is proprietary. It is not a universal maturity standard, and its cross-sectional results do not prove that improving a score causes better commercial outcomes. The sample is also restricted to US companies and four answer engines.

Even with those limits, the findings are a valuable counterweight to self-report. Many organisations have started doing AEO. Far fewer have built the full capability.

The basics remain a large part of the gap. Webflow reported broken internal links on 62% of sites, missing SEO metadata on 60%, and less than 10% of content updated during the previous six months at 54% of companies.[4]

Sophisticated conversations about vector retrieval, knowledge graphs and citation engineering can obscure a simpler reality. Many organisations still need clearer information, sound technical foundations, current content and credible evidence.

<h2>Adoption is strongest where AEO looks most like SEO</h2>

The execution pattern matters as much as the adoption rate.

Semrush found comparatively low reported use of tactics that sit farther from traditional owned-content workflows: 11% were building external mentions through communities and social platforms, 9% were managing online reviews, and 8% were using digital PR.[3]

Those figures do not prove that third-party evidence is always more important than first-party content. They indicate that owned-content activity is diffusing faster than broader evidence-building work.

That gap matters because AI answers can draw on product documentation, publishers, experts, reviews, communities, video and other sources. The relevant mix changes by engine, category and question. AEO Updates' <a href="/articles/ai-citation-benchmarks-evidence-functions">cross-study citation analysis</a> shows why there is no universal first-party versus third-party ratio.

A mature programme begins with the question a customer is asking, identifies the evidence needed to answer it credibly, and then determines which source is suitable to carry that evidence.

<h2>The work left to do is organisational</h2>

The next stage of AEO is unlikely to be won by producing more content alone.

It requires several capabilities that currently sit across different functions. Brand defines what the company should stand for. Customer insights identifies the decision criteria that matter. Product marketing substantiates claims. Content and SEO make evidence accessible. PR and communications help create independent validation. Community and customer teams influence lived-experience evidence. Technical teams support access and interpretation. Analytics connects visibility to outcomes.

No one function controls all of those inputs.

That does not mean AEO should have no owner. The recent AEO Updates analysis <a href="/articles/who-owns-aeo-marketing-operating-model">Who owns AEO? Three surveys give three different answers</a> proposes a more workable model: broad inputs, narrow accountability and distributed execution.

One leader should own the final outcome, the diagnostic cadence and the prioritisation process. Contributing functions should remain accountable for the evidence they control.

This is how AEO becomes an operating capability rather than a collection of tactics.

<h2>Integration is associated with results, but causation is not established</h2>

Semrush reports a striking association. Among respondents who said they were seeing more traffic or leads connected to AI platforms, integrated execution was more common than it was among teams not reporting those results.[3]

That does not prove integration caused the improvement. Teams seeing results may invest more, have stronger brands, work in more AI-visible categories or define success differently. Reverse causality is also possible.

The finding is still useful as a hypothesis. Organisations should test whether clearer ownership, common workflows, shared evidence priorities and better measurement improve decision speed and outcomes.

They should not cite a vendor survey as proof that adopting one operating model will produce a predetermined lift.

<h2>The market is moving from experimentation to institutionalisation</h2>

Minuttia found that approximately 40% of respondents had not allocated a dedicated AEO budget. It also found that 50.2% planned to moderately or significantly increase AI-search investment during the following 12 months.[2]

That second finding should not be rewritten as 50.2% planning to establish a new dedicated budget. It measures intended investment growth, not the creation of a separate budget line.

The direction is nevertheless clear. Organisations are not simply noticing AI search. Many are allocating more time, tools and money to it.

The first phase of AEO was awareness. The second was experimentation. The current phase is institutionalisation: turning experiments into workflows, budgets, roles, measurement systems and cross-functional decisions.

The transition will not occur uniformly. Some companies will remain at the content-optimisation stage. Others will build dedicated teams before they have reliable outcome measurement. A few may develop strong cross-functional execution without using the AEO label at all.

The maturity model is therefore best used as a diagnostic, not a league table.

<h2>How practitioners should use the maturity model</h2>

The first step is to identify the organisation's current constraint rather than aiming immediately for the final stage.

If activity is still experimental, the priority may be defining a stable set of customer questions and a repeatable measurement frame. If owned content is strong but the brand remains absent, the constraint may be authority, independent evidence or retrieval eligibility. If visibility is improving without business impact, the problem may be the connection to consideration, traffic, pipeline or revenue.

The organisation should then choose one intervention that can plausibly affect that constraint, measure the result and preserve a record of the decision.

The <a href="/state-of-aeo-2026">State of AEO 2026</a> report's Measurement Declaration offers a practical disclosure structure. The <a href="/manus-storage/aeo-governance-worksheet_89ad3e3e.pdf">AEO governance worksheet</a> can be used to assign ownership, diagnose the bottleneck and establish a 30/60/90-day operating cadence.

The objective is not to score five out of five on someone else's proprietary model. It is to create a disciplined loop that improves the evidence available to AI systems and connects those changes to decisions that matter.

<h2>AEO Updates Takeaway</h2>

AEO has crossed an important threshold. The accumulated evidence supports the conclusion that it has moved from niche experimentation into mainstream marketing activity.

AEO Updates' best current estimate is that approximately 60% of marketing organisations are doing something intentional about AI search. Roughly 40% may be moving beyond isolated experiments. Around 20–25% appear operationally integrated. Dedicated capability sits near 10%, and mature cross-functional AEO is probably still below that level.

The percentages below the adoption estimate are working interpretations, not measured universal prevalence. They will change as better studies appear and as organisations define the discipline more consistently.

The durable conclusion is the gap between adoption and maturity.

The next competitive advantage will not come from being able to say that a company is “doing AEO”. It will come from knowing what the programme is trying to change, which evidence supports that change, who owns the outcome, and whether the work affects how the brand is understood, considered and selected.

<h3>Sources and methodology note</h3>

This article synthesises five studies that use different populations and measurement frames. Their percentages are not pooled. HubSpot, Minuttia, Semrush and Webflow sell marketing, SEO, website or AEO-related products and have commercial interests in category adoption. McKinsey's 16% figure comes from a small, specific CMO sample. The 55–65% adoption range and most maturity-stage percentages are AEO Updates editorial estimates whose uncertainty increases toward the lower stages.

<h3>References</h3>

1. HubSpot, <a href="https://offers.hubspot.com/state-of-aeo">The State of AEO in 2026</a>.

2. Minuttia and Growth Memo, <a href="https://minuttia.com/state-of-aeo/">State of AEO Report 2026</a>, published 19 June 2026.

3. Semrush, <a href="https://www.semrush.com/blog/the-operational-gap-ai-seo-study/">Only 22% of marketers have fully integrated AI search and SEO. They're pulling ahead</a>, published 3 June 2026.

4. Webflow, <a href="https://webflow.com/blog/aeo-maturity-index">The AI discovery gap: We analysed 2,000 websites, and almost nobody is ready for answer engines</a>, updated 16 July 2026.

5. McKinsey & Company, <a href="https://www.mckinsey.com/capabilities/growth-marketing-and-sales/our-insights/new-front-door-to-the-internet-winning-in-the-age-of-ai-search">New front door to the internet: Winning in the age of AI search</a>, published 16 October 2025.