YouTube · Free tool guide · 6 min read · updated 2026-09-05

How to Use Search Suggest Scraper (2026 Guide)

Scrape YouTube autocomplete suggestions to discover real search patterns.

Open Search Suggest Scraper →

YouTube's autocomplete box is the closest thing creators have to live search-demand data. Every suggestion that appears when you type "meal prep" exists because enough people have searched that exact phrase recently. The Search Suggest Scraper pulls those suggestions for your seed keyword — including the A-to-Z and 0-to-9 expansions most people never type manually — and organises them into themes with an opportunity score for each.

This guide covers how the scrape works, how to read the intent classification, and how to turn a list of 150 suggestions into three video ideas you can actually publish this month.

How the scrape works

For your seed keyword the tool queries YouTube autocomplete for the base term, then for the base term followed by each letter a–z and each digit 0–9 ("meal prep a", "meal prep b" …), then for common question prefixes ("how to meal prep", "why meal prep", "meal prep vs"). The union is de-duplicated and typically comes to 80–200 real phrases.

Each phrase is then classified by search intent — informational (how, what, why), comparison (vs, best, review), transactional (buy, price, cheap) or navigational (brand or creator names) — and grouped into topic clusters so you can see, for example, that 22 of the suggestions are about "meal prep for weight loss" while only 4 are about "meal prep containers".

What the opportunity score means

The score combines how specific the phrase is (long-tail phrases are easier to rank for), how many other suggestions sit in the same cluster (cluster size is a proxy for demand), and whether the phrase includes a modifier that signals a fresh angle ("2026", "beginner", "budget", "without"). Scores above 70 are phrases where a well-packaged video from a small channel can realistically reach the first row of results.

From suggestions to a publishing plan

Take the top 10 underserved keywords the report lists at the end. For each, search it on YouTube in an incognito window and look at the top five results: if two or more are older than 18 months or have under 10,000 views, the phrase is open. Pick the three with the largest cluster behind them and write the exact phrase into the first five words of your title.

Tips

  • Use a two-word seed. One word ("cooking") returns suggestions that are too broad; four words ("easy healthy meal prep") returns too few.
  • Run the same seed once a month and diff the results. New suggestions appearing are the earliest signal of a trend you can still get ahead of.
  • The comparison-intent cluster ("X vs Y") is consistently the least competitive and converts best to watch time, because the viewer has already decided to spend time deciding.

Common mistakes

  • Chasing the highest-volume phrase — The base term is always the biggest cluster and always the most competitive. Small channels win on the modifiers.
  • Ignoring the question cluster — Question phrases map directly to video titles and often to Google video carousels too.
  • Using suggestions as tags only — A suggestion in your tags does nothing if it is not also in your title or first two description lines.

Worked example

Input: Seed keyword: "home espresso".

What the report returned: 142 suggestions in 9 clusters. Largest cluster: "home espresso setup" (31 phrases). Highest-opportunity phrase: "home espresso under 500" (score 84, comparison intent, top results 2+ years old). Recommended title: "Home Espresso Under $500: The Setup I'd Buy in 2026".

Frequently asked questions

Is this real YouTube autocomplete data?

Yes. The suggestions are pulled live from YouTube's autocomplete endpoint at the moment you run the tool, so they reflect current search behaviour rather than a cached database.

Why do I get different results from what I see in YouTube?

YouTube personalises autocomplete based on your history and region. The scraper uses a neutral, non-personalised request, which is closer to what a new viewer sees.

Can I export the list?

Yes — every report has a copy button per section and a full-report export. The keyword list copies as plain text, one phrase per line, ready for a spreadsheet.

Does it work for non-English keywords?

Yes. Autocomplete is returned in the language and script of the seed keyword. Intent classification is strongest in English, Spanish, Portuguese, German, French, Hindi and Indonesian.

Related tools

← All guides · YouTube tools