Social signals are the likes, shares, comments, followers and hashtag counts that a page or profile collects on social networks, and the SEO idea is that a lot of them lift a page in Google. That idea is a myth. Google staff have said for years that these counts are not a ranking factor, and no service that sells “social signals” changes that.
Where does the social signals idea come from?
The idea has three sources:
- Correlation studies. Pages that rank well also collect more shares. Both follow from the same cause: a page that people find useful gets shared and gets found. The shares are a result of quality and visibility, and nothing says they caused the ranking.
- Google’s own history. In the early 2010s Google showed social results from people in your network, and Matt Cutts said in 2010 that links from Twitter and Facebook were treated like other links. Search Engine Journal’s summary of his later statements says that in 2014 he stated Google did not use follower or like counts.
- A market for “signals”. Sellers of followers, likes and shares need a reason for you to buy.
What does Google actually say?
The statements are consistent, and they come from Google’s own staff:
| Who | What was said |
|---|---|
| Matt Cutts, 2014 | Google does not use follower or like counts in its algorithm. |
| John Mueller, 2016 | Use social links “not in the hope that they improve rankings”. |
| Gary Illyes, 2017 | Most social media links count for as little as a single drop in an ocean. |
| John Mueller, 2021 | “Sorry, we don’t use likes as a ranking factor.” |
Search Engine Journal collects these with the original videos and tweets. Two technical facts back them up. Links on social networks are mostly marked nofollow or similar, and Google’s documentation recommends rel="ugc" or nofollow for user-generated content, which tells Google not to pass ranking credit. And buying signals in bulk belongs with the manipulation that Google’s spam policies are written against.
What does social media do for SEO instead?
Social has an indirect effect, and it comes through things Google does count:
- It gets content in front of people who link. Journalists, bloggers and newsletter writers see your post, and some write about it. The link they publish is the ranking signal, the post was the delivery. See digital PR and linkable assets.
- It builds branded search. People who meet your brand on LinkedIn or Instagram later search for it by name.
- It fills your brand results. Profiles rank for your name (see social profiles).
- It ranks on its own. Posts and videos from social networks appear in search results, so a post can capture a query your site does not.
- It brings traffic and feedback. Visitors from social are real readers who tell you what to write next.
What should you do instead?
- Stop paying for signals. Followers, likes and shares you buy bring no ranking value, no buyers and a risk to your reputation.
- Share what deserves links. Post research, tools, templates and strong opinions, not only announcements. Start from your linkable assets.
- Distribute each article on purpose. Plan which network, which hook and which day for every piece, as described in content distribution.
- Use hashtags for discovery inside the network. Hashtags help people find your post on that platform. Google’s ranking does not read them as a signal, so use two or three relevant ones and drop the rest.
- Treat social proof as conversion help. Customer logos, reviews and testimonials on your pages convince visitors. They help sales, not rankings, and reviews matter for local results through their own channel.
- Measure the real effects. Track links earned after a post in a backlink tool, branded queries in Google Search Console and social referrals in GA4 with UTM tags.
- Join communities where your buyers talk. Forums and subreddits send traffic and ideas (see Reddit and forums).
What mistakes should you avoid?
- Buying followers or “social signals” packages. They are bots, so there is nothing to convert.
- Adding share buttons to rank. Buttons make sharing easier for visitors. They do nothing for Google.
- Reading correlation as cause. A page with many shares that ranks well proves the page is good.
- Dropping social entirely. The indirect effects are real, so keep it as a distribution channel with a clear goal.
- Judging social by likes. Count links, branded searches and leads.
Where can you learn more?
The free social media module of the Fe/male Switch startup school covers choosing networks and building an audience without buying one. Next steps on SEO100: set up your social profiles and read backlink monitoring to see which posts turn into links.