Google Hummingbird: what it changed and how to rank today
Short answer Hummingbird was Google's 2013 rebuild that made Search read the meaning of a whole query instead of matching single words. Google now lists it as a retired system; newer AI systems such as RankBrain, neural matching and BERT do that job. There is nothing to optimize for by name. Write pages that answer customers' real questions in their words.
Want help applying this to your business? Talk it through on a free 30-minute call.
Book a callWhat was Google Hummingbird?
A 2013 overhaul of how Google reads a search. Before it, Google leaned on matching the words you typed to the words on a page. Hummingbird moved it toward the meaning of the whole query: what the person actually wants, even if a page phrases it differently.
Google's own one-line history:
"This was a major improvement to our overall ranking systems made in August 2013." (Google Search Central: ranking systems guide)
It was never a penalty. Unlike Panda and Penguin, which targeted thin content and spammy links, it didn't punish anyone. It changed how Google understood questions.
Is Hummingbird still part of Google?
No, not by that name. Google lists it under "Retired systems", with this note:
"Our ranking systems have continued to evolve since then, just as they had been evolving before." (Google Search Central: ranking systems guide)
The idea it stood for, reading intent instead of matching strings, is now handled by several AI systems.
What replaced Hummingbird?
| System | What Google says it does | What it means for your pages |
|---|---|---|
| RankBrain | "helps us understand how words are related to concepts" | A page can rank for words it never uses |
| Neural matching | understands "representations of concepts in queries and pages and match them to one another" | Google pairs a vague search with the page that fits it |
| BERT | lets Google "understand how combinations of words express different meanings and intent" | Small words like "to" or "for" can change results |
| MUM | "not currently used for general ranking in Search" | Ignore any "MUM optimization" offer |
| AI Overviews and AI Mode | may use "query fan-out", running related searches to build an answer | Same SEO rules; see below |
Quotes from the ranking systems guide and AI features page.
When BERT launched in October 2019, Google said it would help Search "better understand one in 10 searches in the U.S. in English" (Google, 2019). Its example: in "2019 brazil traveler to usa need a visa", the word "to" decides who needs the visa. Older keyword matching missed that.
Google's plain summary of the whole shift: its synonym system finds "relevant documents even if they don't contain the exact words you used" (Google: How Search works).
How do you rank today?
The same way Hummingbird pointed to, with fewer shortcuts left. What Google's own guides say:
- Use your customers' words, not your trade's. Google's example: some people search "charcuterie", others "cheese board". A plumber's customer types "water heater making popping noise", not "sediment buildup". Write for the words they use (SEO Starter Guide).
- Don't chase every phrasing. Google says its "language matching systems are sophisticated and can understand how your page relates to many queries, even if you don't explicitly use the exact terms in them" (SEO Starter Guide). One clear page per service beats ten near-copies. See on-page SEO for service pages.
- Never stuff keywords. Repeating words "is tiring for users, and keyword stuffing is against Google's spam policies" (SEO Starter Guide). More in keyword density.
- Answer the whole question, at the length it needs. Google asks site owners whether they write to a set word count, then answers: "(No, we don't.)" (Google: helpful content).
- Treat schema as a bonus, not a ranking lever. Structured data (code that labels your hours, prices or reviews for Google) "enables a feature to be present, it does not guarantee that it will be present" (Google: structured data guidelines).
- Do nothing special for AI answers. Google: "There are no additional requirements to appear in AI Overviews or AI Mode, nor other special optimizations necessary" (Google: AI features). More in AI Overviews for local businesses.
Hummingbird-era advice that no longer holds
| Old advice | What Google says now |
|---|---|
| Put the exact keyword in every heading | Google matches meaning; variations and synonyms are fine |
| Longer articles rank better | Google has no preferred word count |
| Add the keywords meta tag | "Google Search doesn't use the keywords meta tag" (SEO Starter Guide) |
| Schema markup boosts rankings | It makes rich results possible, never guaranteed |
| Optimize for Hummingbird (or BERT, or MUM) | These systems read pages; there is no setting to target |
Did Hummingbird cause my traffic drop?
No. It shipped in 2013 and is retired. A recent drop usually lines up with a core update, a site change, or a competitor. Check the dates in traffic dropped after a Google core update.
Example
A roofer rewrites one page
Their "Roof Repair Services" page lists the keyword six times and says little else. Customers search "roof leaking after heavy rain" and "missing shingles after storm". The roofer rewrites the page around those problems: what causes each, what a repair involves, a typical price range in USD, and how fast they come out. The page now matches what people mean, not one phrase.
A dental office drops its keyword pages
The office has separate pages for "teeth whitening", "tooth whitening" and "whiten teeth near me", each a near-copy. They answer one need and compete with each other. The office merges them into one full page and redirects the other two to it.
FAQ
When did Google Hummingbird come out?
Google dates it to August 2013 and lists it as retired in its ranking systems guide.
Is Hummingbird the same as RankBrain?
No. Hummingbird was Google's 2013 ranking overhaul, now retired. RankBrain is a separate AI system linking words to concepts, still active.
Did Hummingbird make voice search possible?
Google's own note on Hummingbird doesn't mention voice. For what matters now, see voice search for a local business.
Do I need to do anything for Hummingbird now?
No. Write clear pages that answer real customer questions, and the current systems handle the rest.
Different for you?
- Your pages repeat keywords: fix them with keyword density and on-page SEO for service pages.
- Your traffic dropped recently: see traffic dropped after a Google core update.
- You want to show up in AI answers: see Google AI Overviews for local businesses.
- You'd rather someone else write and maintain the pages: see SEO.