How to Track SERP Features in 2026 (Free + API Methods)

The blue links aren’t the SERP anymore. They’re maybe a third of it.

Google now stuffs the result page with AI Overviews, featured snippets, People Also Ask, People Also Search For, related searches, and knowledge panels. Each one is a separate surface that can steal or feed your traffic. Ahrefs’ December 2025 analysis of 300,000 keywords found that the presence of an AI Overview alone cuts the top-ranking page’s click-through rate by 58%. If you only track rank position, you’re optimizing for a page that doesn’t exist anymore.

So the real question isn’t “what’s my rank?” — it’s how do you track SERP features one surface at a time. This guide covers the seven that matter and how each moves clicks, then the two ways to actually track them: the free methods you can start today, and the API method that scales SERP features tracking to millions of keywords. “We’re #1 on Google” is no longer a meaningful claim on its own.

What are SERP features

A SERP feature is any element on a Google search results page that isn’t a standard organic blue link. Google has been adding them since the Knowledge Graph launched in May 2012. Expansion accelerated in 2024 when AI Overviews rolled out globally, and again at Google I/O 2025 when AI Mode started replacing the classic SERP entirely for a subset of queries. By May 2026, AI Mode had reached 1 billion monthly users across 180+ countries.

The result: a typical commercial query in 2026 renders 4-6 distinct features above the first organic result. Each is ranked, optimized, and tracked separately. They share a query, not a leaderboard. That fragmentation is exactly why SERP features tracking has become a discipline of its own, distinct from classic rank tracking.

Why this matters:

• Ranking #1 organically while losing the AI Overview cuts CTR roughly in half on average, and as much as 79% on the worst-affected queries per Authoritas’ research. Same position, different traffic.
• Owning the featured snippet on an informational query lifts CTR to 35.1%, versus 23.3% for the #1 organic result when a snippet is present. Position zero is worth more than position one.
• PAA blocks expand the surface area of every query. A keyword you thought was settled has 4-8 sub-queries attached to it, each its own opportunity.
• Brands cited inside AI Overviews earn 35% more organic and 91% more paid clicks than non-cited brands on the same query, per a Seer Interactive study of 25M impressions.

If you treat the SERP as a single ranked list, you’ll miss all of this. Every SERP is a collection of independent surfaces. Your visibility is a vector, not a scalar.

The 7 SERP features that matter

Here’s the canonical list, ranked by how much they currently move traffic in commercial categories:

Feature	What it is	Why it matters	Tracking priority
AI Overview	AI-generated answer block at the top of the SERP	Pushes organic below the fold; cites 3-8 sources	Critical
Organic results	The classic 10 blue links	Still the largest single source of traffic	Critical
Featured snippet	Highlighted answer pulled from a single source	Position zero — high CTR when present	High
People Also Ask (PAA)	Expandable related questions	Reveals intent fan-out; rewards FAQ schema	High
Knowledge panel	Structured entity card (right rail or top on mobile)	Defines how Google understands the entity	Medium
People Also Search For (PASF)	Alternative queries shown after a click-back	Maps refinement intent	Medium
Related searches	Bottom-of-page query suggestions	Cheapest source of long-tail keyword ideas	Low

The rest of this section breaks each one down.

1. AI Overview

AI Overview is Google’s generative answer box, rendered above all organic results. It synthesizes 3-8 cited sources into a 2-4 paragraph answer.

Independent studies converge on the same direction with different magnitudes: Ahrefs measured a 58% drop in top-result CTR when AIO is present, Seer Interactive saw 49-65% organic CTR decline, and Authoritas reported up to 79%. The variance depends on query category, AIO length, and citation count. The trend is settled: this is the single biggest reason SERP features tracking now leads with AI Overview presence.

What to track:

• Presence: does this query trigger an AI Overview today?
• Citations: which domains are cited, in what order?
• Text: the actual generated answer (it drifts daily)

AI Overview tracking is the biggest measurement shift since featured snippets. We’ve covered the tooling landscape in our AI Overview tracking tools comparison, and the mechanics of scraping the surface in how to scrape Google AI Overview.

2. Organic results

The blue links. Still the largest source of clicks in absolute terms, but with a shrinking share of the page.

Track the standard fields per result: position, URL, title, snippet, displayed domain — cloro’s Google organic results API returns each of them per listing. For the position-tracking side specifically, see our Google rank tracking API guide, or build it on the rank tracking API use case.

Watch for sitelinks, FAQ rich results, video thumbnails, and review stars. Google attaches these to organic results when structured data matches.

3. Featured snippet

A single source promoted to “position zero”. Google pulls a paragraph, list, or table directly from the page and renders it above the organic results. Featured snippets are zero-sum: one URL wins, everyone else loses.

They’re also volatile. A SearchPilot × Moz split-test measured a ~12% traffic drop within days when Moz’s pages were forced to give up their snippet. When held, a snippet pulls a materially higher click-through rate than the 23.3% the #1 organic gets in the same SERP — one reason featured-snippet ownership is a first-class metric in any SERP features tracking setup. The exact answer length, schema, and format-matching pattern that wins them is covered in featured snippet optimization.

Track the snippet URL, the snippet text, and the snippet type (paragraph / list / table / video). Diff over time to catch ownership changes.

4. People Also Ask (PAA)

PAA is a block of 3-4 expandable questions related to the original query. Click any one and 3-4 more questions appear underneath it. The block fans out indefinitely. Each question has its own cited source.

PAA matters for two reasons. First, it’s the most visible artifact of Google’s intent-fan-out model: it reveals what users actually want to know around your query, which is content-strategy gold.

Second, it’s a citation surface. If your page is cited as the answer to a PAA question, you get a click even when you don’t rank in the top 10 organic. Pairing answers with FAQPage structured data is how you signal eligibility.

Track the questions shown, the cited URLs, and the answer text — the People Also Ask API returns all three per question. The question list rotates between refreshes, so a single snapshot understates real coverage. Track over a window of 5-10 fetches and union the results.

5. Knowledge panel

A structured card about an entity: a brand, a person, a place, a product. Pulled primarily from Wikipedia, Wikidata, and the Google Knowledge Graph, with reinforcement from on-site schema markup (Organization, Person, Product).

Knowledge panels only appear for entity-named queries. When they do appear for your brand, they’re the single highest-real-estate piece of the SERP. Track presence, the panel’s title and description, the linked social profiles, and the image — all fields the Google Knowledge Graph API returns per query. Misinformation in a knowledge panel is a brand-safety issue, not an SEO one.

Schema markup is how you reinforce the panel. We cover the playbook in schema markup for AI.

6. People Also Search For (PASF)

PASF is the easiest SERP feature to confuse with PAA, and the easiest to underestimate. It appears after a user clicks an organic result and returns to the SERP. Google injects a “People also search for” block beneath the clicked result, showing alternative queries.

Where PAA reveals intent broadening (“what else might this user want to know?”), PASF reveals intent refinement (“what did the previous click fail to answer?”). For commercial queries, PASF is often a direct competitor map. Click on “best CRM” and bounce back, and you see “best CRM for small business”, “Salesforce alternatives”, “free CRM tools”.

Track the PASF queries shown beneath each organic result. The set is per-result, not per-query. Different clicked URLs surface different PASF lists.

7. Related searches

The list of 8-10 queries shown at the bottom of the SERP. Lower commercial value per query than PAA or PASF, but the cheapest long-tail keyword discovery surface available. Google is literally telling you what the next searches are. Track the full list via the related searches API and feed it into keyword research.

Beyond the core seven: other SERP features to track

The seven above move the most traffic, but Google renders a long tail of specialized features that dominate specific verticals. Mature SERP features tracking widens to these whenever your queries trigger them.

Local pack

The three-result map block that appears for local-intent queries (“dentist near me”, “coffee shop”). It’s the dominant surface for any location-based business, ranked by proximity, prominence, and relevance rather than classic organic signals. Track its presence, the three ranked businesses, and their star ratings — a discipline that overlaps heavily with local SEO tooling.

Shopping pack

Product listing ads and organic product cards render for commercial-product queries, often above organic results with images, prices, and merchant names. For retail, this is frequently the single highest-converting surface on the page. cloro’s Google Shopping results API returns each product tile — title, price, merchant, rating — as structured JSON.

Video and Top Stories

Video carousels (usually YouTube) surface for how-to and entertainment intent. Top Stories carousels appear for newsworthy, time-sensitive queries and rotate within minutes. Both are worth tracking whenever your category spikes into the news cycle or leans visual, since presence flips far faster than organic rank.

Sitelinks and rich results

Sitelinks are the indented sub-links Google attaches beneath a dominant result — usually a brand’s own navigation — and signal strong entity authority. Rich results (review stars, FAQ drop-downs, recipe cards) attach to organic listings when structured data matches. Both change the click-through math of a position without changing the position itself, which is exactly the kind of shift rank-only tracking misses.

Ads and “People are saying”

Paid results sit above and below the organic block and shift the true fold; cloro exposes them through the Google Ads results API. Google’s newer “People are saying” panel surfaces quotes pulled from forums and reviews — a citation surface that rewards Reddit and community presence.

How to track SERP features for free

You don’t need a paid tool to start. For a handful of keywords, three free methods cover most of what you need — the catch is every one of them stops scaling at exactly the point real tracking begins.

1. Google Search Console. GSC is the only free source of your own SERP-feature data. In the Performance report, filter by Search Appearance and you’ll see impressions and clicks broken out by feature type — AI Overview, videos, and the rich results (FAQ, review snippets, etc.) your pages already win. It tells you which features you’re eligible for and how their CTR compares, but only for queries where you already rank, and only after Google reports the data on its own delay.

2. A manual incognito search. Open an incognito window, search your target keyword, and read the page top to bottom: is there an AI Overview, who does it cite, is there a featured snippet, which questions are in the PAA block? This is the ground truth — it’s what a real user sees today — and it’s free. It’s also unrepeatable at scale: it’s one keyword, one location, one moment, and it captures nothing over time.

3. Free-tier trackers and browser extensions. Free plans on rank trackers (Wincher’s free tier, Ahrefs Webmaster Tools) and SERP-feature browser extensions will flag features on a small fixed keyword set. Fine for a side project; the free ceilings (keyword caps, weekly refreshes, one location) are hit fast.

Free works until you need many keywords, more than one location, or a daily snapshot you can diff over time — which is to say, until you’re actually tracking SERP features rather than spot-checking them. That’s the line where the manual labor costs more than an API, and the method below takes over.

How to track SERP features at scale (the API method)

The mechanics of SERP features tracking are the same regardless of which feature you’re tracking:

1. Issue a search at a controlled location, device, and language.
2. Parse the response into structured fields per feature type.
3. Snapshot to storage, keyed by (keyword, location, device, timestamp).
4. Diff snapshots to detect feature presence changes, citation changes, and rank changes.

The hard parts are the search and the parse. Google aggressively blocks scrapers and rotates its DOM, so DIY parsing breaks every few weeks. A purpose-built API is faster and cheaper to operate than a self-rolled stack.

cloro’s SERP API returns all seven features in a single structured JSON response per query: organic, PAA, PASF, AI Overview text and citations, featured snippet, knowledge panel, related searches. One call, all surfaces, parsed.

Sample workflow:

1. Define a tracked-query list (the keywords you care about per market).
2. Schedule daily fetches against the API with location and device set per market.
3. Write each response into a time-series store keyed by query.
4. Run diffs nightly: feature_appeared, feature_disappeared, citation_added, citation_removed, snippet_owner_changed.
5. Route the diffs into alerting (Slack, email) and a dashboard.

For brand-tracking (“how often is my brand cited in AI Overviews and PAA for queries I care about?”), the same pipeline becomes a share of voice tracker with a different aggregation layer on top. The dedicated tools layer that sits on top of feeds like this is what we benchmark in LLM visibility tracking tools.

SERP features tracking tools vs. a raw API

Dedicated SERP features tracking tools — the rank-tracker category, including Semrush, Ahrefs, and Wincher — bundle scraping, storage, and dashboards. That’s convenient for a fixed keyword set but expensive and rigid once you scale.

A raw SERP API hands you the parsed features and lets you own the storage and diffing layer: cheaper per query, and far more flexible when you need custom aggregations or millions of keywords. Most teams start with a tool and graduate to an API once their SERP features tracking outgrows a dashboard.

Optimizing for SERP features

Each surface rewards different content:

• AI Overview: citation-worthy, entity-rich, fact-dense content. Named statistics, clear definitions, authoritative sources. The mental model is the same as GEO: make your content the easiest thing for an LLM to lift.
• Featured snippet: a 40-60 word definitional answer placed immediately under an H2 that matches the query verbatim. Lists and tables also work; Google picks the format that matches the question type.
• PAA: FAQ blocks with explicit question-shaped H2s/H3s, paired with FAQPage schema markup. The questions you want to win are the ones already appearing in PAA blocks for your target query. Start by mining them.
• Knowledge panel: Wikidata entry, consistent Organization schema across your site, consistent NAP across the web, linked social profiles. This is a 6-12 month project, not a quick win.
• PASF: won by being the page that satisfies the refined intent. If “X alternatives” appears in PASF for your category, that’s your next page.

Optimization without SERP features tracking is hope. Track the surfaces you’re trying to win, weekly at minimum.

Common mistakes

The most common SERP features tracking failures we see teams make:

• Tracking rank without tracking features. “We rank #2 but traffic dropped 40%” usually means an AI Overview appeared. The rank didn’t change. The page did.
• Treating PAA and PASF as the same feature. Different surfaces, different intent, different optimization. Confusing them produces incoherent content briefs.
• Snapshotting once and moving on. PAA, related searches, and AI Overview text all drift between refreshes. A single snapshot is a sample, not a measurement. Aggregate over a window.
• DIY scraping at scale. Google’s anti-bot stack will defeat a homegrown scraper within weeks. The labor cost of maintaining your own infrastructure dwarfs API costs at any volume that matters.
• Optimizing for AI Overview citation without authority. Citation in AI Overviews tracks domain authority and entity recognition. A new domain with perfect schema still won’t get cited. The compounding work is AEO and brand authority. Schema is the multiplier, not the foundation.

---

Ready to put SERP features tracking into production?

cloro’s SERP API returns AI Overviews, PAA, PASF, featured snippets, knowledge panels, related searches, and organic results in a single structured response across 250+ locations and every major device profile. Stop guessing why your traffic moved.