SERP Features Cheat Sheet 2026: Ranked by CTR Lift

A SERP feature is any result element beyond the classic blue link — featured snippets, People Also Ask boxes, carousels, the local pack, star ratings, knowledge panels, AI Overviews. Each one changes the click math for the pages that win it.

This is every SERP feature Google serves today, ranked by CTR lift and how to earn each one. Use it as a checklist, not a bucket list: chasing all of them at once is how teams produce schema soup and win nothing. Before publishing any page targeting these features, preview exactly how it renders in search with the SERP Preview tool.

On the numbers: the CTR ranges are aggregated from audit experience and published industry studies. Treat them as directional — lift depends on position, device, and query intent. The eligibility requirements, though, are documented and testable. That is what the how-to-earn lines below are built on.

Key takeaways

• The biggest lift is local: the local 3-pack earns +25–40% CTR on local queries — and is irrelevant if you are not a local business.
• The best eligibility-to-effort ratio is the featured snippet: ~8% lift for a 40–60 word answer placed above the fold.
• Reviews and stars (+10–17%) are gated: markup restricted to third-party review sites means self-serving the schema gets nothing.
• AI Overviews change the math, not the game: your page is cited if you rank in the top 10. No direct CTR lift, but zero visibility loss either.
• Prioritize structurally: snippet first, then PAA, then sitelinks and breadcrumbs — the low-effort, architectural wins.
• One trap kills campaigns: match the feature to the query intent. A how-to post should optimize for the HowTo rich result, not PAA boxes.

What a SERP feature is

A SERP feature is any non-link element Google renders in search results. The features split into two families, and conflating them causes wasted work. The first is rendered automatically when the algorithm decides a feature helps the query — featured snippets, PAA, image packs, knowledge panels, AI Overviews. For these you write. The second is rich results you request through structured data — FAQ and HowTo extensions, breadcrumbs, video displays. For these you add schema. Every section below says which game you are playing.

How to earn the featured snippet (~8%)

The featured snippet is Google’s extracted answer box — roughly an 8% CTR lift when you hold it. You earn it by answering the question above the fold, in 40–60 words, in a clean paragraph or list.

Google prefers answer blocks that can be lifted without surgery: a 40–60 word paragraph beginning with the answer, directly under a question-formatted H2. Lists win list-shaped queries; paragraphs win “what is” definitions. Practical sequence: find queries where you rank 1–10 and a snippet box exists, rewrite the top-of-page answer to match the box’s format, re-check after the next recrawl. The keyword density checker tells you whether your target terms appear at the right frequency without triggering over-optimisation penalties.

How to earn People Also Ask (+5–15%)

People Also Ask is the accordion stack of related questions, driving +5–15% total SERP CTR where it appears. You earn it by including related questions with concise answers in your content.

Each PAA box is a real query with a real searcher behind it. Create a section per question, phrase the heading the way the box phrases it, answer in a short block under the heading, and you have given Google a clean lift candidate. Expect displacement too — PAA boxes often sit above your position even when you do not hold them, so treat the questions as a content-gap report, not a badge. The same question-shaped blocks also feed AI answer systems, which makes PAA formatting an investment in both the classic SERP and the layer on top of it.

How to win the local 3-pack (+25–40%)

The local pack — the map-and-three-listings block on local queries — carries the largest effect on this sheet: +25–40% CTR for local queries. You earn it with an optimized GBP profile.

“Optimized” in practice: every Google Business Profile field complete, category correct, hours accurate, photos current; reviews accumulating steadily rather than in bursts; name-address-phone identical across every citation. Relevance, distance, and prominence are the ranking inputs — you control relevance and prominence. If you serve local customers, this feature outranks everything below in priority. The setup order is in our local SEO guide.

How to earn review stars (+10–17%)

Star ratings render beneath your title and lift CTR roughly +10–17% where eligible. The critical constraint: reviews markup is restricted to third-party review sites. Self-serving stars on your own pages are parsed and ignored.

The logic is trust separation — the rater cannot be the seller. Aggregators and comparison publishers earn the display because their opinions are structurally independent. If you sell products, your realistic star presence lives on retailer and platform listings. If you publish reviews, the Product and AggregateRating path is yours to earn honestly.

How to earn Top Stories (~12%)

Top Stories is the fresh-results carousel on trending queries, with about a 12% CTR lift when you appear. You earn it with Article schema and the news-publisher requirements.

The bar is real: news-cadence freshness, a publishing footprint Google treats as news-worthy, and clean Article markup so headline, dates, and author parse without guessing. For most commercial blogs the honest read is that this feature is not for you, and approximating it wastes the effort you owe to the features that are.

How to earn sitelinks and breadcrumbs (+5–10%)

Sitelinks are the sub-URL rows Google attaches beneath a strong result, worth +5–10% CTR. You earn them by clear site architecture and breadcrumbs schema.

The mechanism is inference, not request: Google adds sitelinks when it is confident your site is the query’s destination and four to six internal pages are obviously distinct. BreadcrumbList markup hands it the hierarchy explicitly; a clean structure lets it read the hierarchy anyway. Neither works on a flat site where every page is a sibling — this feature rewards the architecture you should already have.

How to earn the image and video features (+3–7, +4–8)

Two visual features, two earn-lines. Image pack placements add +3–7% CTR on image-friendly queries — earned by including original images with descriptive alt text and ImageObject schema. Video carousel entries add +4–8% CTR on queries where intent matches — earned with VideoObject schema and embedded YouTube videos.

Images: originality is the hard filter — stock photography is already indexed everywhere, so a duplicate offers Google nothing to surface. Shoot or generate distinctive assets, name the file for what it shows, write alt text that describes the image in context, compress for speed, and pair ImageObject markup so dimensions and captions parse. Video: a YouTube-hosted video with a descriptive title and chaptered content ranks in its own vertical, while an embedded video with VideoObject markup attaches player metadata to your listing. The carousel is strongest for tutorial, review, and entertainment queries — confirm it actually renders on your target SERP before investing.

How to earn FAQ and HowTo rich results (+2–5, +3–7)

FAQ rich results expand Q&A pairs beneath your listing at +2–5% CTR — earned with FAQPage schema and concise Q&A pairs. HowTo rich results display step summaries at +3–7% — earned with HowTo schema and a step-by-step format.

Honest current state: Google has narrowed FAQ display in the main SERP toward government and health sites, so many commercial pages mark up cleanly and never render — the schema still feeds answer systems, and eligibility stays conditional rather than closed. HowTo remains visible on genuinely procedural queries. Both are cheap once your content already answers in pairs and numbered steps. Write for the questions first, mark up second; the Schema Generator outputs valid markup for both FAQPage and HowTo types.

How to earn the knowledge panel

The knowledge panel is the entity card — logo, facts, description — on branded queries. Its payoff is branded trust: searchers read the panel as verification that the result is the real one.

You earn it slowly: Organization schema so your entity parses machine-readably, a Wikipedia presence where notability genuinely supports one, and Wikidata so the entity’s facts match across sources. The panel is assembled from the entity graph, which means corroboration decides arrival — wide, consistent brand coverage, not a request.

How AI Overviews affect your traffic

AI Overviews are Google’s synthesized answer blocks on qualifying queries. The honest read of the data: an AI Overview cites your page if you rank in the top 10, there is no direct CTR lift, and the alternative is zero visibility loss — not the catastrophe the headlines claim.

The consequence is an eligibility rule: you cannot optimize into an AI Overview from position 25. Top-10 ranking is the ticket, so the layer rewards the same fundamentals the rest of this sheet assumes — answer-first blocks, question-matched headings, clean schema. Track whether cited pages appear on your top queries; absence tells you the answer structure is beatable.

The prioritization order

Sequence by eligibility-to-effort, not lift size. Start with featured snippet — it has the highest eligibility-to-effort ratio on the sheet. Move to PAA next. Then site links and breadcrumbs, which are low effort because they are structural. Local 3-pack only if you have a local business; otherwise the biggest number on the sheet is not even in your game.

The markup family — FAQ, HowTo, video — sits after the automatic features: cheap once, worth adding when the content shape already fits, never worth bending content to fit a schema type. Our SERP features rankings guide tracks which features are winning for your target queries over time.

Chase the feature the query already shows. A feature Google never renders for your keywords is a task to delete, not optimize.

The trap: chasing every feature

Don’t chase every feature. Pick the one that matches your query intent. A how-to post should optimize for the HowTo rich result, not PAA boxes.

The failure mode is architectural: a team applies the checklist uniformly, publishes definition blocks on procedural queries and step lists on definitional ones, and the SERPs ignore both. The discipline runs the other way — inspect the live results for the exact target query first. If the snippet box is there, format the answer for it. If a carousel dominates, that is the whole job. If nothing renders, rank the blue link properly and stop dressing it up.

How to measure feature performance

1. Baseline visibility. For a tracked set of target queries, record which features render and whether you hold them — a weekly rank-tracker column does it.
2. Read the Search Console filters. Appearance and search-type filters isolate image and video performance; our walkthrough of the hidden Search Console reports shows where those views actually live.
3. Compare CTR at stable position. A page holding a snippet against a matched page without one is your in-house version of every range on this sheet.
4. Re-check quarterly. Rendering shifts with algorithm updates — this sheet is a 2026 snapshot, not permanent law.

Frequently asked questions

Which SERP feature gives the biggest CTR lift?

For local queries, the local 3-pack by a wide margin: +25–40%. For everyone else, reviews and stars at +10–17% — but that markup is restricted to third-party review sites, which makes the featured snippet’s ~8% the biggest lift most pages can actually earn on their own content.

Do SERP features directly improve rankings?

No. Features change presentation and click share, not position. The indirect case mirrors meta descriptions: a presentation that wins the click compounds through engagement signals over time, but the lift shows up on CTR first and possibly nowhere else.

Can I request or buy a SERP feature?

You can request eligibility for markup-driven rich results with valid structured data; you cannot request the algorithmic ones. Featured snippets, PAA, image packs, knowledge panels, and AI Overviews are Google’s rendering decisions — content structure earns them.

Does the featured snippet steal clicks from position one?

The research has gone both directions and depends on query type. Our audits and the broader ranges say holding the snippet still nets roughly +8% versus not holding it. The honest answer: format for it because it is nearly free, then check the effect on your own pages with a stable-position CTR comparison.

How often should I re-check which features my queries show?

Quarterly at minimum, and immediately after major Google announcements. Feature rendering changes with algorithm updates and AI-layer expansion — what renders on your queries in 2026 is not guaranteed in 2027.

Bottom line

Rank by lift, act by eligibility. The local 3-pack is the biggest number and the narrowest audience; the featured snippet is the best effort-to-reward trade and the first thing to fix; PAA and the structural features follow; the markup family is cheap housekeeping once content fits; AI Overviews reward the fundamentals you needed anyway.

Build the answer-first muscle with the meta description study — one feature, the query’s feature, at a time.