Insights
8 FAQ Page Tips That Actually Reduce Tickets
Casey Rowland
TL;DR:
A good FAQ page deflects real ticket volume, but only if it's organized around how customers actually search, not your internal categories.
66% of customers try self-service before contacting a human, per Microsoft's own research. A weak FAQ page wastes that instinct.
Google retired the classic FAQ rich result from Search entirely on May 7, 2026. FAQPage schema is still valid, but it no longer gets you that dropdown in the SERP.
FAQ-formatted content still matters for AI search, ChatGPT, Perplexity, and Google's AI Overviews all favor clear question-and-answer pairs when pulling an answer, schema markup or not.
A static FAQ page and an AI agent solve different problems. One answers what's already documented. The other resolves what isn't.
FAQ's, we've all read them. Large language models (LLMs) love 'em. But that doesn't mean that they are actually useful for your customers. So how do you make FAQs actually useful and informative? In this post, we'll give 8 tips and tricks to writing FAQs that should help reduce the number of tickets your team has to tackle.
1. Organize FAQs by how customers search, not your internal categories
Organization is key when setting up your FAQ page. Where we see people get tripped up is they write for how the business thinks about the problem, not how customers think about the problem. They don't phrase the question like a customer, "where is my order", "can I return this item", "why was I charged XYZ). Match the customers language and not your own taxonomy.
For an ecommerce store specifically, most ticket volume clusters into a handful of categories:
Shipping and delivery: Processing times, carriers, tracking, lost or delayed packages.
Returns and exchanges: Policy windows, condition requirements, how refunds are issued.
Sizing and product fit: Especially for apparel, footwear, or anything with variant-specific fit issues.
Payment and billing: Accepted methods, failed charges, duplicate charges, promo codes.
Account and order management: Editing an order after purchase, canceling, changing an address.
Group questions into these buckets, not a single undifferentiated list. If any one category is large enough (returns policy detail, sizing charts by product line), give it its own page rather than burying it inside the general FAQ.
2. Use accordion dropdowns, not a long visible scroll
Accordion sections keep the page scannable without hiding content from search engines or AI crawlers reading the page, collapsed content in a standard accordion is still indexable. This also helps customers find their question quicker since they don't have to scan through a sea of information.
3. Keep answers to 2-3 sentences, then link out
If an answer takes more than 2 or 3 sentences, then it needs its own deep dive. Consider writing a blog post, or KB article that goes into depth on this topic, and then use the FAQ to link over to the article. This helps show search engines that the FAQ and the blog article are related, and valuable to customers.
4. Add a search bar once you're past 15-20 questions
FAQ pages can get long. Like, really long. Customers word questions many different ways, which can cause an FAQ page to expand greatly, without adding much incremental value. If it does start to get to an infinite scroll, consider adding a search bar. This helps the user experience immensly, and helps get your customers the information that they want.
5. Make it readable
Size does matter; font size that is. Too small of font makes it unreadable to your customers. Mobile traffic can't read your FAQ's if its written too small. Minimum 16px font, no answers that require pinch-to-zoom. Ecommerce traffic is mobile, and a FAQ page that's hard to read on a phone doesn't get used.
6. Surface it everywhere it's relevant, not just at /faq
Customers aren't going to search for an FAQ page for their specific answer. They want an answer as they are shopping. Embed the relevant few questions directly on product pages for pre-purchase concerns (sizing, materials, care), and surface it during checkout for shipping and return-policy questions specifically.
7. Assign ownership and review it on a real cadence
A FAQ page with out of date information, is worse than no FAQ page at all. Plus, its a liablitly to your brand. You're telling customers something that is false, and they'll hold your feet to the fire. If you FAQ says "free delivery with $50 minimum purchase" but your actual policy is $100 minimum, you're at risk for customer escalations. Review your FAQ on a regular basis.
8. Get the schema and AI-search question right, since most advice on this is outdated
This is the tip most likely to be wrong if you learned it more than a year ago. Here's what's actually true as of late 2026:
FAQPage schema no longer gets you a rich result in Google. Google retired the classic FAQ dropdown rich result from Search entirely on May 7, 2026, following a partial restriction back in August 2023 that had already limited it to a handful of authoritative government and health sites. As of the 2026 change, nobody gets that rich result anymore, not big sites, not small ones. If you're adding FAQPage schema expecting the expandable snippet in Google, that specific payoff is gone.
The markup itself is still valid and still parsed. Google has said FAQPage remains a legitimate Schema.org type and that Search still parses it to understand the page, it just doesn't render the visual rich result from it anymore. There's no real cost to leaving it in place.
Whether the schema itself helps AI citations is genuinely disputed. Some SEO-industry sources claim FAQ schema improves citation rates in AI Overviews specifically, through a knowledge-graph pathway distinct from how ChatGPT or Perplexity extract content. But a separate industry study (Search/Atlas, December 2024) reportedly found no correlation between schema markup coverage and citation rates across AI platforms. That's a real disagreement in the field, not a settled fact, so treat any confident claim that "FAQ schema boosts AI citations" with some skepticism.
What isn't disputed: the visible Q&A format itself matters for AI search, independent of whether schema is present. AI systems (ChatGPT, Perplexity, Claude, and Google's AI Overviews) tend to extract and cite content that's already structured as a clear question followed by a direct, self-contained answer, because that's the easiest shape to lift a clean answer out of. A FAQ page is naturally that shape already. The lever that reliably helps is the visible writing, not the invisible markup.
Practical takeaway: keep FAQPage schema on your pages if you already have it (it's harmless and still technically valid), but don't add it expecting a Google rich result, and don't treat it as your main AI-visibility strategy. Put the real effort into writing genuinely clear, direct, self-contained question-and-answer pairs. That's what both a human skimming the page and an AI system extracting from it actually need.
Where a static FAQ page stops being enough
A FAQ page answers questions you already anticipated and already wrote down. That's exactly its limit: it can't handle "where is my specific order" or "can I return this specific item I bought three weeks ago," because those require looking up real data, not serving a pre-written answer.
This is where the two tools do genuinely different jobs. A FAQ page deflects the generic version of a question. An AI agent that reads live order, return, and account data resolves the specific version of it. See our breakdown of what AI customer service actually is for how that resolution-first approach differs from a static self-service page.
If you want to know whether your current setup (FAQ, chat, or both) is actually working, track it the same way you'd track any other support channel: resolution rate, not just page views. Our guide to customer service metrics that actually matter covers how to measure that without fooling yourself with a vanity number.
Where Weav fits
We think a good FAQ page and a good AI Agent do different jobs, and a store needs both. Weav's AI Agents pick up exactly where a static FAQ runs out: reading your actual order, return, and subscription data to resolve the specific version of a question your FAQ page can only answer generically. If your FAQ is solid but your ticket queue is still full of "where's my specific order" questions, get started with Weav for free and see how it handles the tickets a FAQ page can't.
FAQ
What is a FAQ page? A FAQ page is a self-service page that answers a business's most common customer questions in a short, scannable format, so customers can find an answer without contacting support directly.
How do I create a FAQ page? Start by pulling your last 30-90 days of support tickets and grouping the repeat questions by topic. Write short, 2-3 sentence answers in the customer's own language, organize them into categories (for ecommerce: shipping, returns, sizing, payment, account), and use an accordion format with search if the list is long.
Does FAQPage schema still help SEO? It no longer produces Google's classic FAQ rich result, that was fully retired from Search on May 7, 2026. The markup is still valid and still parsed by Google, and it's harmless to keep, but it shouldn't be your main SEO or AI-visibility strategy anymore.
Does AI search actually favor FAQ pages? The visible question-and-answer format does, AI systems tend to extract and cite content structured as a direct question followed by a clear, self-contained answer. Whether the FAQPage schema markup itself adds a citation boost on top of that is genuinely disputed in the industry, evidence is mixed rather than settled.
Can a FAQ page replace live support? No. A FAQ page only answers questions you've already anticipated and written down. Anything requiring a look-up against real customer or order data, like a specific order status or a specific return, needs either a human or an AI agent that can read that data directly.
Related Articles
Casey Rowland
