Google Indexing of Twitter Appears to Be Declining

At Perficient Digital, we’ve been actively tracking Google’s indexation of Twitter since July of 2014. Today, I’m publishing our fourth update, and it offers up some fascinating insights. At the end of this post, you’ll find a handy slide deck summary of the whole study.
To start, here are two of the more interesting conclusions of this report:

1. Google’s indexation of Twitter has peaked. Back in July of 2015 we saw the highest indexation of Twitter ever. In fact, in all our updates since July of 2014, Twitter indexation was continually increasing, but as of this update, Twitter indexation is down overall.
2. In our prior updates, we showed that Follower count or Social Authority appears to make a difference, but today we’ll show that having a Verified Account has a very high correlation with the percentage of your tweets that get indexed. This a fresh confirmation that clear identity matters quite a bit to Google.

In this study (as with the other three that we’ve published), we looked at more than 90,000 tweets from over 900 accounts, across a wide range of follower counts and social authority. The other three studies we’ve published can be found here:

1. Twitter indexing as of July 2014
2. Twitter indexing as of January 2015
3. Twitter indexing as of July 2015 (right after Google’s use of the Twitter firehose was launched)

Overall Indexation Over Time

The following chart shows us the aggregate indexation of all the tweets looked at in each study:

Each looked at tweets from the same basic group of people on Twitter. The only adjustments occurred when people dropped off of Twitter, but otherwise the accounts tracked remained the same. Each study included more than 90,000 tweets.
So what we see here is that Google hit an indexation peak around July of 2015, and today indexes only about 64% of the tweets it was indexing back then. As a company, Google is constantly testing and measuring impact of different approaches to indexation and ranking. This drop off is a strong indication that the level of indexation they had in July of 2015 was higher than the optimum for their overall search results, and they’ve scaled it back a bit.
[Tweet “New study shows Google indexing of Tweets has declined over time. More at”]
What most people don’t realize is that Google’s indexation of Twitter is often a bit slow. We’ve shown that consistently through our four studies in the topic. Look at how many tweets are indexed after seven days:

However, as of October of 2016 (when the data I’m publishing today was collected) it’s the fastest it’s ever been, with 4.1% of tweets getting indeed within seven days. Still, this is less than half the tweets that Google eventually indexes, and that means that there are other factors that Google uses to decide that a tweet is worth indexing.

Impact of Favorites and Retweets on Indexation

Revolutionize Your Business With Generative AI

From product design and software development to virtual agents, content creation, and reporting, GenAI is transforming business. Our AI experts help you unlock GenAI’s full potential and drive growth.

Let’s Get Started

Here is the data on the impact of Favorites on indexation from the data for this fourth version of the study:

And, here is the corresponding data on the effect of Retweets on indexation

As you can see, tweets that garner a lot of Favorites and Retweets are indexed at a much higher rate. Our data seems to show a preference for Favorites over Retweets, though I’m still working on why that may be the case. It may be that data shows that Favorites occur somewhat more sparingly than Retweets, perhaps because of all the accounts out there that auto-Retweet all the tweets of any influential account.
[Tweet “Study shows tweets with more favorites or retweets have higher chance of showing in Google’s index. More at”]

Impact of Follower Count / Social Authority

Here is the data using the Follower count of the accounts tracked, showing indexation levels over time:

We also tracked indexation levels based on Followerwonk Social Authority, though we have that data only for two of the four studies we’ve done:

What Social Authority takes into account that Follower counts does not is the level of engagement with an account’s tweets, as well as the authority levels of those who engage with it, and this is why we include this view in our data.
As you can see, higher Follower counts, or higher Social Authority, impacts your level of indexation. This makes sense – more people are interested in what these people have to say. In addition, this data shows that Google has dropped the level of indexation since July 2015 regardless of Follower Count or Social Authority. Higher levels still get more than lower levels, but just not as much as they used to.
[Tweet “Study shows higher follower counts or authority increase chances tweets will be indexed by Google. More at”]

Impact of Having a Verified Account

Last, but not least, is a look at the impact of having a verified account. This shows a very powerful impact:

This is underscored even more when we look at the indexation of unverified accounts:

Notice how the level of Social Authority matters little for unverified accounts. In contrast, it’s a very big deal for verified accounts. The indexation level for low Social Authority accounts that are verified is not much better than it is for unverified accounts. It looks like you need both for maximum results!
[Tweet “Study shows verified Twitter accounts with high authority have the best chance of tweets showing up in Google.”]

Conclusions

Our data provides some very interesting insights into how Google looks at Twitter. Please bear in mind what follows is my interpretation of, and speculation about, the data.
1. Testing Processes at Google: The evolution of Twitter indexing over time is revealing. Shortly after the Twitter firehose deal was implemented by Google, it looks like indexation levels peaked. This makes sense. Google was conducting testing to see what the optimum levels for indexation were. They will continue to test and iterate those algorithms over time. For example, the continuing shift of consumer attention to engaging with the web via Smartphone devices may well continue to drive shifts in indexation levels.
2. Google Is Not in a Rush to Index a Tweet: Speed is NOT a primary driving factor in indexation. Remember, even after 7 days, less than half the tweets that Google will eventually index are indexed. That means that other factors play a very big role in what gets indexed, and what doesn’t.
3. Activity Levels a Big Factor: Our data suggests that the driving factor is engagement levels (Favorites and Retweets). One could imagine a sophisticated algo that has varying thresholds over time, where a certain rate of initial activity could cause something to get indexed quickly. On the other hand, a sustained level of engagement over time might cause a tweet to be viewed as more valuable evergreen content, worthy of inclusion because of that ongoing engagement.
4. Social is Not the Highest Priority Content: Over the past couple of years, we’ve shown not only in these Twitter studies, but also in a Facebook indexing study, that Google doesn’t make any attempt to index all social media posts. In fact, it’s fair to say that the vast majority of such posts are ignored. This underscores one of the reasons our Mark Traphagen gave for why social media is not a direct ranking signal for Google.
5. Identity Matters: The strong data for verified accounts was pretty revealing. Recall what Eric Schmidt said in his book The New Digital Age: “Within search results, information tied to verified online profiles will be ranked higher than content without such verification, which will result in most users naturally clicking on the top (verified) results. The true cost of remaining anonymous, then, might be irrelevance.” That’s a pretty powerful statement, and the data we have on Twitter indexing seems to back that up.

Slide Deck Version of This Study