We queried Ahrefs Keywords Explorer for 50 keyword pairs comparing "Twitter ___" vs "X ___" search terms. Same feature, same intent, just the brand name swapped. Twitter wins 42 of the 50 matchups. In some cases by 3:1. In others by 45:1. The full dataset is below.
TL;DR
- "Twitter" = 7.5M US monthly searches. "X" can't be measured cleanly because "x" is also a letter, a variable, a movie franchise, etc.
- Twitter wins 42 out of 50 keyword matchups. Margins range from 1.5:1 to 45:1.
- X wins 8 terms, all navigational or transactional (login, app, .com). None cultural.
- Legacy vocabulary persists: "retweet" gets 70x more searches than "repost x." "Quote post x" = zero.
- "Twitter video downloader" alone gets 303K/mo. The X version? 6,700. Ratio: 45:1.
The Raw Numbers
124,000 people search "twitter x" every month. They need both names to find the platform. That alone tells the story.
Head-to-Head: 50 Keyword Pairs
Twitter Wins (42 terms)
Sorted by ratio, highest first. Every row is the same feature, just the brand name swapped.
| Keyword | "Twitter ___" | "X ___" | Ratio |
|---|---|---|---|
| video downloader | 303,000 | 6,700 | 45:1 |
| verification | 4,100 | 100 | 41:1 |
| search | 58,000 | 1,900 | 30:1 |
| spaces | 2,500 | <100 | 25:1+ |
| archive | 3,400 | 150 | 23:1 |
| memes | 3,400 | 150 | 23:1 |
| alternatives | 4,200 | 200 | 21:1 |
| image size | 1,200 | 60 | 20:1 |
| bio ideas | 350 | 20 | 18:1 |
| thread maker | 150 | 10 | 15:1 |
| sensitive content | 1,300 | 100 | 13:1 |
| shadowban | 3,100 | 250 | 12:1 |
| trending | 15,000 | 1,400 | 11:1 |
| banner size | 6,000 | 700 | 9:1 |
| hashtag | 150 | 20 | 8:1 |
| download | 8,100 | 1,100 | 7:1 |
| analytics | 6,500 | 900 | 7:1 |
| bot | 1,000 | 150 | 7:1 |
| name generator | 250 | 40 | 6:1 |
| handle | 1,500 | 250 | 6:1 |
| embed | 500 | 90 | 6:1 |
| lists | 600 | 100 | 6:1 |
| developer | 500 | 90 | 6:1 |
| header size | 1,100 | 200 | 6:1 |
| report | 500 | 100 | 5:1 |
| video limit | 250 | 50 | 5:1 |
| tips | 100 | 20 | 5:1 |
| profile | 1,000 | 200 | 5:1 |
| advanced search | 25,000 | 5,600 | 4:1 |
| dm | 400 | 90 | 4:1 |
| for business | 600 | 150 | 4:1 |
| character limit | 4,200 | 1,200 | 4:1 |
| engagement rate | 80 | 20 | 4:1 |
| schedule post | 60 | 20 | 3:1 |
| impressions | 200 | 60 | 3:1 |
| down | 40,000 | 12,000 | 3:1 |
| ads | 1,800 | 600 | 3:1 |
| marketing | 900 | 300 | 3:1 |
| delete account | 4,100 | 1,500 | 3:1 |
| followers | 1,300 | 500 | 3:1 |
| is ___ down | 49,000 | 23,000 | 2:1 |
| communities | 600 | 400 | 1.5:1 |
X Wins (8 terms)
| Keyword | "Twitter ___" | "X ___" | Ratio |
|---|---|---|---|
| .com login | 600 | 7,300 | 12:1 X |
| buy followers | 2,500 | 9,100 | 3.6:1 X |
| account | 1,600 | 4,400 | 2.8:1 X |
| stock | 16,000 | 42,000 | 2.6:1 X |
| api pricing | 350 | 700 | 2:1 X |
| app | 6,500 | 12,000 | 1.8:1 X |
| login | 34,000 | 45,000 | 1.3:1 X |
| blue / premium | 2,200 | 1,500 | * |
*Ahrefs Keywords Explorer, US data, February 2026. "Blue / premium" counted as X win because "X Premium" is the current product name, replacing "Twitter Blue." All volumes are 12-month rolling averages.
Look at the X wins list. Every term is either a URL-bar action (login, .com, app, account), a financial instrument (stock), or an infrastructure query (api pricing). Not one of them is how people describe what they do on the platform.
Nobody says "I saw a great x post." They say "I saw a great tweet." The search data matches.
Where Each Name Wins
The same person, two different search bars
The Vocabulary Test
X replaced "tweets" with "posts," "retweets" with "reposts," and "quote tweets" with "quote posts." We tested whether the new vocabulary shows up in search.
| Old Term | Volume | New Term | Volume | Ratio |
|---|---|---|---|---|
| tweet | 6,200 | x post | 700 | 9:1 |
| tweets | 3,400 | x posts | 400 | 9:1 |
| retweet | 700 | repost x | 10 | 70:1 |
| quote tweet | 300 | quote post x | 0 | ∞ |
"Quote post x" has zero monthly searches. The replacement term does not exist in search behavior. "Repost x" gets 10 searches per month. "Retweet" gets 700. The vocabulary rebrand failed harder than the name rebrand.
This is consistent with external data points. A 2024 YouGov survey found 55% of daily US users still say "Twitter," and only 19% say "X." The AP Stylebook still recommends "X, formerly known as Twitter." But the search data makes the point on its own: people search the way they talk, and they talk in Twitter vocabulary.
Limitations and Counterarguments
This is keyword search data, not a controlled experiment. Some caveats worth noting.
twitter.com still redirects to x.com. Some "twitter" searches may come from people who type "twitter" in the address bar and get redirected. This likely inflates Twitter-branded navigational terms. It does not explain the gap in non-navigational terms like "twitter memes" (23:1) or "twitter shadowban" (12:1). Nobody types those into an address bar.
"X ___" terms have ambient noise. "X down" could mean Xbox down. "X account" could reference any X. We tried to exclude pairs with obvious non-platform meanings, but some leakage is possible. This noise would actually inflate X volumes, making the real gap larger than what we report.
Search volume ≠ usage. People can use X daily while still searching for "Twitter." The data measures brand recall in search behavior, not platform activity. These are different things.
Ahrefs volumes are estimates. All SEO tools use clickstream models to estimate search volume. The absolute numbers have margins of error. The ratios between paired terms, which use the same estimation model for both sides, are more reliable than the raw numbers.
US-centric. We pulled US data. The pattern may differ in markets where Twitter had less cultural penetration pre-rebrand.
Some pairs may not be perfectly clean. "Twitter poll" vs "x poll" is clean. "Twitter growth" vs "x growth" is muddier. We included pairs only where the platform interpretation was the dominant search intent for both versions, but judgment calls were involved.
The Bottom Line
The rebrand moved the URL. It didn't move the name. After 2.5 years, Twitter-branded keywords win 42 out of 50 matchups we tested. The old vocabulary persists at ratios of 9:1 to infinity. X wins only where infrastructure forced the switch: login pages, app stores, stock tickers.
The interesting question isn't whether the rebrand failed by this measure. It clearly did. The interesting question is whether it matters. If 100 million people use x.com daily but search "twitter trending" when they want to find what's popular there, what does the brand actually own?
The data says: "X" owns the login page. "Twitter" owns the mental model. And in search, the mental model is what generates traffic.
Asaf Fybish
Building organic growth engines for startups | SEO, AI Search & Relevance Engineering. Co-Founder & GM at Growtika.
Related Articles
ChatGPT SEO: How to Get Your Company Cited in AI Answers (2026)
The complete guide to getting cited in ChatGPT, Claude, Perplexity, and Google AI.
OpinionI'm Breaking SEO Best Practices. And It Feels Good.
I created something called the LLM Sitemap. It's not a best practice. SEO tools are throwing errors at me.
OpinionSEO After LLMs: Where Competitive Advantage Lives Now
When content generation becomes trivial, differentiation must come from something other than optimization.