How to Audit a Webflow B2B SaaS Site Against Google's May 15 AI Guide

On May 15, 2026, Google Search Central published its first official guide to optimizing for AI Overviews and AI Mode, with explicit guidance that AEO and GEO are still SEO and that tactics like llms.txt and content chunking are unnecessary. The guide gives Webflow Partners a Google-confirmed audit framework anchored on indexability, unique non-commodity content, query fan-out coverage, agent-friendly DOM, and Schema where it supports rich results. From Phoenix Studio in Bengaluru, I run this exact audit on my own pravinkumar.co practice site plus one anonymized B2B SaaS client site each week. In this piece I walk through the 10-step audit, the actual Webflow Designer panels I use, the CMS query I run to surface non-commodity content gaps, and what each step costs in time on a typical B2B SaaS marketing site.

Why audit a Webflow site against Google's May 15 guide?

Auditing a Webflow site against Google's May 15 guide is valuable because the guide is the first Google-published reference framework for AI search optimization. Earlier audits referenced third-party SEO frameworks that included some tactics Google has now explicitly called unnecessary. Running an audit against Google's own document produces a result you can defend to a B2B SaaS founder with the actual Google URL as the source.

For Phoenix Studio's retainer clients, the audit serves a second purpose: surfacing which existing content categories are likely to lose visibility to AI Overviews and AI Mode versus which are likely to gain. The piece on pre-I/O AEO prep covered the answer-block discipline that the audit operationalizes. The audit itself runs in roughly two hours per site, which is small relative to the durability of the resulting changes.

How do I check if my Webflow site is indexable?

To check if a Webflow site is indexable, open the site in the Webflow Designer, navigate to Project Settings, then SEO, and verify that the site is not set to disallow indexing. Then open robots.txt at yoursite.com/robots.txt and confirm it does not block important paths. Then submit the site to Google Search Console and confirm pages are appearing in the Coverage report as Indexed.

The Phoenix Studio audit step adds one Webflow-specific check that catches a common error: verify that the SEO meta tags on each page template do not include a noindex directive in the Custom Code section. Some clients accidentally add this during staging and forget to remove it before launch. The check takes 5 minutes per site and catches issues that would otherwise produce zero AI Overview citations and zero traditional Search rankings.

What counts as "non-commodity content" on a B2B SaaS site?

Non-commodity content on a B2B SaaS site is content that cannot be generated by a generic AI assistant from public information alone. Examples include original product documentation, named-customer case studies, internal-data analysis, regulatory-context content specific to the company's jurisdiction, and comparison content based on direct experience with named competitors. The opposite (commodity content) is generic advice that an AI can produce from training data alone.

The Phoenix Studio CMS query I run in the Webflow Designer is a filter on the Blog Posts collection that surfaces posts older than 12 months and shorter than 800 words. These posts are most likely to be commodity content that no longer earns citations. The query produces a list of 5 to 20 posts on a typical B2B SaaS site, which becomes the prioritization list for either deprecation or rewrite during the audit follow-up sprint.

How do I map content to query fan-out coverage?

To map content to query fan-out coverage, take a high-priority user query (for example, "best B2B SaaS analytics tools for fintech") and list the three to five related queries that AI Mode is likely to fan out into during retrieval. Then check whether the site has answer-block content for each fanned-out query. Sites that cover the full fan-out earn citation surface across multiple related Search queries; sites that cover only the exact query lose surface to competitors.

The Phoenix Studio audit pattern uses a simple spreadsheet with the primary query, the fan-out queries, and a column for each query indicating whether the site has dedicated answer-block content. Gaps in the spreadsheet become the prioritization list for content production. The piece on Reddit citations in AI Mode covered the parallel pattern of evaluating which third-party content already earns citation for fan-out queries the site does not yet cover.

Does my Webflow site need schema markup for AI Overviews?

No, a Webflow site does not need schema markup specifically for AI Overviews. Google's May 15 guide states that there is no special schema.org markup needed for AI features. Schema markup remains valuable because it powers rich results in traditional Search, which produces direct visibility benefits, but it is not a requirement for AI Overview or AI Mode citation eligibility.

For Phoenix Studio's audit pattern, the schema check still runs because the rich-results benefits are durable and worth maintaining. The check uses Google's Rich Results Test tool at search.google.com/test/rich-results on three to five representative templates (article, product, FAQ, breadcrumb, organization) per site. The check takes 15 minutes per site and surfaces missing or malformed schema that would block rich-result eligibility. The piece on well-known files tutorial covered the parallel server-side configuration audit.

How do I check Webflow's DOM for agent-friendliness?

To check a Webflow site's DOM for agent-friendliness, view the page source in the browser, look for clean semantic HTML structure, and verify that important content is in the static rendered DOM rather than injected by JavaScript. Webflow Designer-built sites generally produce clean semantic markup by default, but custom-code embeds and JavaScript-driven content patterns can create agent-readability issues.

The Phoenix Studio audit uses the browser DevTools Elements panel to spot-check three to five representative pages per site. The check focuses on whether headings (H1 through H4) form a logical outline, whether main content is inside a main element or has clear semantic landmarks, and whether interactive elements have proper accessible labels. The check takes 10 minutes per site and surfaces patterns that affect both AI agent readability and traditional accessibility.

What should I drop from my current SEO checklist?

Drop llms.txt deployment, content chunking specifically for AI, AI-specific schema markup beyond what rich results require, and the practice of rewriting SEO-performant content into AI-optimized variants. Google's May 15 guide explicitly states these tactics are not necessary for AI features. Keeping them in the SEO checklist wastes time and signals to clients that the practice has not read Google's actual current guidance.

The Phoenix Studio service-page copy update on May 18 removed two line items that previously referenced AEO-specific work. The line items are now bundled inside the standard Webflow SEO scope, which matches Google's framing and simplifies the founder conversation about pricing. The change is small and easy to implement, and it pre-empts the founder question about why the service mix includes work that Google itself says is unnecessary.

How does this audit change for local or ecommerce Webflow sites?

For local Webflow sites, the audit adds two steps: verify Google Business Profile completeness and verify Local Business schema markup on relevant pages. For ecommerce Webflow sites, the audit adds two different steps: verify Google Merchant Center feed completeness and verify Product schema markup on product detail pages. Both additions are referenced in Google's May 15 guide as relevant for AI features that include shopping or local results.

For Phoenix Studio's audit pattern, the local and ecommerce additions are conditional on the client's audience type. A B2B SaaS marketing site without local or ecommerce surface skips both. A client with hybrid local-and-B2B audience (for example, a regional accounting SaaS) runs the local additions. A client with a Shopify storefront alongside their Webflow marketing site coordinates the ecommerce checks with the Shopify side rather than running them on Webflow.

How do I track AI Overview impressions for my Webflow site?

To track AI Overview impressions for a Webflow site, open Google Search Console, navigate to Performance, and look for any new metric or filter that surfaces AI Overview impression data. As of May 2026, Google has not exposed AI Overview impressions as a separate metric in Search Console for most accounts, so the tracking pattern is to observe overall impression and CTR trends and infer AI Overview impact from changes in the long-tail query mix.

The Phoenix Studio tracking pattern uses a monthly Search Console export to a spreadsheet, with a query-mix analysis that flags queries with rising impressions and falling CTR (a common AI Overview signature, since AI Overviews satisfy the query without a click). The analysis produces a list of pages that warrant rewrites to capture the AI Overview citation surface. The tracking takes 30 minutes per month per client and is durable signal.

What's the right cadence to re-run this audit?

The right cadence to re-run this audit is quarterly for retainer clients and at major Google announcement points for project clients. The quarterly cadence catches drift in content quality, schema validity, and Search Console signal patterns. The announcement-driven cadence catches changes when Google updates AI Search optimization guidance (which the May 15 guide is the most recent example of).

For Phoenix Studio specifically, the audit cadence aligns with the retainer billing cycle, which makes the work easy to schedule and easy to report. The total audit time per client is roughly 90 to 120 minutes per quarter, which fits comfortably inside the retainer scope. The pattern compounds because each audit produces a small prioritization list that becomes the content production queue for the following quarter, which keeps the work focused on durable improvement rather than reactive churn.

If you want to run this audit on your own Webflow B2B SaaS marketing site and want a second pair of eyes on the results, drop me a line and tell me which two or three high-priority pages you want to audit first. I will share the Phoenix Studio audit spreadsheet template and the actual CMS query I run in the Webflow Designer to surface non-commodity content gaps. Let's chat.