IFA Website FAQ Schema: What Changed on 7 May 2026

Your IFA website FAQ schema lost its most visible SERP feature on 7 May 2026, when Google removed FAQ rich results from Search for every site. The expandable Q-and-A dropdown that used to sit beneath your blue-link listing is no longer rendered, even on government and health domains that kept the feature after the August 2023 restriction.

That sounds like a downgrade, and on the SERP it is. The wider picture is more interesting: the same FAQPage markup that just lost a Google snippet is being lifted directly into AI Overviews, ChatGPT Search and Perplexity answers. The structural job of an FAQ section on an IFA site has shifted from "win a snippet" to "get cited by an AI engine".

What changed for FAQ schema on 7 May 2026?

On 7 May 2026 Google retired the FAQ rich result, so the expandable accordion that used to appear under organic listings no longer shows in any search. The change applies universally, including the government and health domains that retained the feature after Google's August 2023 restriction. The FAQPage schema type itself has not been deprecated.

This is the end of a long taper. Google first restricted FAQ rich results to "well-known authoritative" government and health sites in August 2023 (Search Engine Land, 2023), citing low utility for general users. The May 2026 announcement closed the remaining carve-out and rolled the change to every domain on the same day.

The visible result on a typical IFA SERP listing: where you previously had a title, URL, meta description and a few expandable FAQ rows, you now have only the first three. Average vertical real estate per top-three listing has shrunk by around 25 to 40 per cent on queries that triggered FAQ accordions, depending on how many FAQ rows your page had.

Should you delete FAQ schema from your IFA website?

No. Google has explicitly confirmed that it still uses FAQPage structured data to understand pages, even though the rich result is gone, and unused structured data does not harm rankings (Google Search Central, 2026). Stripping it out costs you the AI citation signal the markup now provides without giving you anything back in exchange.

The rule of thumb: leave the JSON-LD block in place. If your CMS or theme strips schema you no longer use, override that. The cost of keeping FAQPage markup is zero.

The cost of removing it is invisible until you notice your firm has stopped appearing in AI Overview citations on the questions you used to rank for.

What you can drop is the dependency on FAQ-specific Search Console reports. Those endpoints will be gone by August 2026, so any monthly dashboard, Looker Studio panel or scheduled export that pulls from searchAppearance=FAQ should be re-pointed before then.

Why does FAQ schema still matter for AI search?

Answer engines including Google's AI Overviews, ChatGPT Search, Perplexity and Gemini all parse structured Q-and-A blocks when deciding which sentence to lift as a cited answer. For an IFA, the same FAQ section that lost a SERP feature is now an extraction surface for AI engines, especially on educational queries.

The contrast with regular page text matters. Unstructured content forces the model to guess where a question ends and an answer begins; FAQPage markup tells the engine explicitly. On finance topics where Google applies stricter "Your Money or Your Life" verification, structured signals carry more weight than they used to.

The shape of finance AI Overview coverage is uneven, and FAQ schema helps most where coverage is highest. Educational queries like "what is an ISA" trigger an AI Overview on roughly 91 per cent of searches; rate or planning queries on 67 per cent; real-time price queries on just 7 per cent (BrightEdge, 2026). The first two categories are exactly the questions visitors run before they enquire with an IFA, so the markup pays off where it matters.

This is where the broader answer engine optimisation play for IFA visibility overlaps with old-school structured data work. The same JSON-LD block that used to win a SERP accordion is now a quiet plumbing layer that helps AI engines extract you cleanly.

What should you do with your IFA website FAQ schema now?

Treat the FAQ as a citation product, not a snippet trigger. Expand coverage to the questions visitors actually ask before they enquire, write each answer in 40 to 70 words so it can be lifted whole, and name verifiable entities such as your FCA Firm Reference Number and CII Diploma. Schema stays; the editorial brief shifts.

Five concrete moves that change citation outcomes after May 2026:

1. Audit existing FAQ pages for AI-citable shape. Each answer should open with a direct sentence, not "great question, here at our firm we believe...". The opening sentence is what the engine lifts.
2. Widen the FAQ to cover the long tail. A typical IFA FAQ page has 5 to 8 questions; the AI-extraction sweet spot is 12 to 20, including the awkward ones (cost, minimums, complaints process).
3. Name entities precisely. "Chartered Financial Planner (CII)" beats "qualified", "FCA Firm Reference Number 123456" beats "regulated", and "CII Level 4 Diploma in Regulated Financial Planning" beats "experienced".
4. Add primary-source outbound links. Linking to the FCA Register or gov.uk inside an answer signals upstream verifiability, which YMYL-trained engines reward.
5. Keep the FAQPage JSON-LD block. Validate it against Google's Structured Data Markup docs for FAQPage. Even though the rich result is gone, the markup is still parsed.

This is the same content surface that powers an on-site chat widget. If your FAQ section is already shaped for AI extraction, it is also already shaped for chatbot lead generation on your IFA website, because both surfaces lift from the same answer-first content.

What does this mean for IFA lead capture?

Fewer visitors will skim a SERP FAQ accordion and never click. With AI Overviews now appearing on 48 per cent of all Google searches (ALM Corp, 2026), the visitors who do reach your site arrive expecting the same direct answer the AI gave them. An on-site answer surface, whether a tightly written FAQ section or a chat widget that reuses your content, is now the front line of conversion.

For an IFA practice, this also reframes which pages deserve investment. The classic "Services" pages laid out as marketing copy are now the wrong shape for both Google and AI extraction.

A Q-shaped page (one client question, a 50-word direct answer, then context, schema and an entity-rich about block) is doing two jobs at once: it is the page that ranks, and it is the page an AI engine quotes. The same restructure is what makes your site visible to AI search in the first place.

The conversion implication follows naturally. When the SERP no longer hands a visitor a snippet, you have a brief moment after they click in which they are scanning for the answer they came for. If the page surfaces it within the first 60 words, you keep them; if not, they leave.

A widget that answers the same question conversationally extends that window from seconds to minutes, and turns the answer into a contact request when intent is highest.

FAQ

Will Google penalise sites that keep FAQ schema after May 2026?

No. Google has stated that unused structured data does not cause problems for Search and that it will continue to use FAQPage markup to understand pages even though the rich result is gone (Google Search Central, 2026). Removing the JSON-LD block is unnecessary work, and on an IFA site it removes a signal that AI engines still use to extract answers.

Do AI tools like ChatGPT and Perplexity actually parse FAQPage schema?

Yes. ChatGPT Search, Perplexity and Google's AI Overviews all use structured data as one input among several when extracting Q-and-A pairs from a page. Structured FAQ blocks are easier to lift than free text because the markup explicitly delimits each question and answer, which is why answer engines tend to favour pages that include them on educational queries.

What schema should IFA websites use alongside FAQPage now?

Add FinancialService to declare your firm name, FCA Firm Reference Number, address and services, and Person schema for each named adviser with their qualifications. These give AI engines the verifiable entities they cross-check on YMYL topics. The combination is what turns a generic "qualified adviser" page into a page that an engine treats as authoritative.

When does Search Console stop reporting on FAQ schema?

The FAQ search appearance, the rich result report and Rich Results Test support are removed in June 2026, and Search Console API support for FAQ rich result data ends in August 2026 (Google Search Central, 2026). Any reporting pipeline that pulls from those endpoints should be migrated to a generic structured data validation flow before the August cut-off.

Can FAQ pages still rank on Google after the May 2026 update?

Yes, on the same SEO basis as any other page: relevance, quality, internal linking, page experience and topical authority. Losing the FAQ rich result removes a SERP feature, not a ranking signal. An FAQ page with strong direct answers and named entities can still rank competitively, especially on long-tail questions visitors actually type into IFA websites.

Does this change how I should write FAQ answers for an IFA website?

Yes, mildly. The optimal length is 40 to 70 words per answer, opening with a direct sentence that resolves the question. Avoid throat-clearing ("great question") and lead with the verb. The same answer that gets lifted by an AI engine also reads cleanly to a human visitor, so this is one of the few rewrites with no trade-off between the two audiences.

If you want a faster way to convert the visitors who arrive after losing the SERP snippet, the ChatIFA widget at chatifa.co.uk reuses your existing FAQ and site content to answer the same questions in a conversation, and captures lead details when intent is highest. The free tier covers 25 messages a month with no payment details required, which is enough to test whether on-site answers convert better than a static FAQ page for your traffic.