For several years, FAQ and HowTo rich results were the easiest visible win in SEO. Add some markup, get an expanded listing occupying more of the results page, watch click-through improve. Entire content formats were designed around them.
Google withdrew both. FAQ rich results were narrowed to a small set of well-known authoritative government and health sites, and HowTo rich results were removed from general display. Most sites still carry the markup, because nothing broke and nobody was told to remove it.
So the practical question is what to do with it now, and the strategic question is what the episode should teach you.
Does leaving it in place hurt?
No. Markup for a feature that no longer displays is ignored rather than penalised. If your site has FAQ markup, it is not costing you rankings.
It is costing you two smaller things. It is dead weight in your page, which is trivial. And more importantly, it is a maintenance liability: a block of structured data that must stay consistent with your visible content, that somebody will eventually update in one place and not the other, and that produces validation warnings nobody acts on because the feature is gone anyway.
Unmaintained structured data is how sites end up with markup that contradicts the page. That is a real problem when it happens on markup that still matters.
The decision, per case
Three situations and three answers.
If the FAQ content is real and visible on the page, keep the markup. It describes the page accurately, it costs nothing, and accurate machine-readable structure has value beyond Google’s rich results. Other systems read structured data too, and a question-and-answer pair expressed as data is easier to extract than one expressed as styled text.
If the markup exists without matching visible content, remove it. This happens more than people expect, usually because a plugin generated it from a field nobody fills in any more. Markup describing content that is not on the page is wrong regardless of whether the feature displays.
If the FAQ section was created solely to earn the rich result, reconsider the section rather than the markup. A block of six invented questions appended to the bottom of every page was always content padding. Now that it earns nothing visible, it is only padding. Remove it and put the answers where they belong in the body.
For HowTo markup, the answer is simpler. Remove it. There is no feature consuming it and the step structure in your visible content does the work.
What still gets you extracted
The useful thing to understand is that the rich result and the extraction are different mechanisms, and only one of them went away.
An assistant composing an answer reads your page. What it needs is a clearly stated answer, close to the question, in plain text. That worked before the markup existed, it worked while the rich result existed, and it works now.
So the format still matters enormously. A page that asks a question as a heading and answers it in the first sentence underneath is highly extractable. A page that buries the same answer in the fourth paragraph of a section called Our Approach is not. None of that depends on structured data.
We set out how to write those sections in writing FAQ sections that work for AEO, and the mechanics of how sources get selected in how AI search engines choose what to cite.
What to do with your how-to content
The content was never the problem. Step-by-step instructional content is among the most useful material a business can publish, and the removal of a display feature does not change that.
What changes is that you can stop contorting it to fit a markup schema. No more forcing a nuanced process into numbered steps with a tool list and a time estimate because the specification wanted those fields.
Write it the way the task actually works. Use an ordered list where the order genuinely matters. Use prose where the decision is a judgement rather than a step. Put the prerequisites at the top because readers need them, not because a field existed for them.
Keep clear step numbering in the visible content. That is what makes instructions scannable for a reader and liftable by an assistant, and it costs nothing.
Which markup is still worth maintaining
Worth being clear about where the effort should go instead.
- Organisation and site identity. Who you are, your identifiers, your profiles elsewhere. Underpins how you are understood as an entity.
- LocalBusiness for anywhere with a premises. Well supported and genuinely useful.
- Product, where you sell things. This one has direct commercial consequence.
- Article and author. Attribution to a real named person with credentials, which matters more as content provenance becomes a live question.
- Breadcrumb. Small, stable, and it improves how your listing reads.
- Event, Job posting, Recipe, Video, where they apply to what you actually publish.
Generate all of it from the same source as the page rather than maintaining it separately, and validate it on deployment. The approach is in automating structured data from your CMS.
The strategic lesson, which is the valuable part
A visible search feature that many businesses built a content strategy around was withdrawn with limited notice, and the sites most affected were the ones that had optimised hardest for it.
That is not unusual. It is the normal lifecycle of a search feature. Features appear, get exploited, get abused, and get narrowed or removed. Anybody who has worked in this field for long has watched it happen repeatedly.
Two conclusions follow, and they are worth holding on to.
Do not build content formats around display features. Build them around what readers need. If a feature happens to reward that, good. When the feature disappears, your content is unaffected because it was never the reason the content existed.
Treat an easy visible win as temporary by default. If a tactic is cheap, obvious and widely adopted, it is on a clock. That is not a reason to avoid it. It is a reason not to depend on it, and not to measure your programme by it.
The same reasoning applies to everything currently being sold as an AI search tactic. Ask what remains valuable if the feature changes. If the answer is nothing, it is a trade rather than an asset.
What is a waste of effort
Auditing your whole site to strip FAQ markup urgently. It is not hurting you. Remove it as you touch pages for other reasons.
Adding FAQ markup now in the hope the feature returns. It will not, and if something similar appears it will have different requirements.
Removing genuinely useful question-and-answer content because the rich result went away. The content was serving readers. That has not changed.
Chasing whichever rich result is currently available as a strategy. You will be writing this same article about that one in two years.
Believing a tool that reports your markup as an error requiring immediate action. Most of those warnings are about features that no longer exist.
A sensible order of work
- Inventory where FAQ and HowTo markup currently exists across your site.
- Remove HowTo markup. Keep the visible steps.
- For FAQ markup, check whether matching visible content exists. Remove the orphans.
- Identify FAQ sections that were created purely for the feature and fold their answers into the body.
- Keep accurate FAQ markup where the content is real, and stop thinking about it.
- Redirect the effort to Organisation, LocalBusiness, Product and author markup, generated from source and validated on deployment.
- Review your instructional content and rewrite anything that was shaped by a schema rather than by the task.
If you are deciding what to do with a large library of old how-to content, the update, merge or delete question is covered in content refresh. The broader honest position on what markup achieves is in does schema markup help you get cited.
If you want an audit of what your site currently publishes and what is worth keeping, get in touch.