How to find out which of your tweets drove the most new followers

Your most liked tweet is probably not the one that grew your account.

Those are two different populations doing two different things. Likes come overwhelmingly from people who already follow you, which is why a good joke to your existing audience can clear a thousand likes and bring in nobody. New followers come from strangers, and a stranger has to do something a follower never does: leave the timeline, open your profile, read it, and decide.

Which means the metric that predicts follower growth is not likes. It is profile clicks.

The frustrating part is that X will not connect the two for you. There is no "this tweet earned you 14 followers" number anywhere in the product, and there never has been. You get a follower count that moves, and a pile of individual tweets, and no line drawn between them.

So the answer has to be assembled. With Circleboom Twitter, you can rank your entire tweet history by profile clicks, then line those posts up against a daily follower growth chart and see which ones actually moved the number.

What Is Circleboom Twitter?

Circleboom is an Official X Enterprise Developer, pulling tweet and account data through X's official APIs rather than scraping the interface.

• Ranks your full tweet history by any engagement metric, including profile clicks, impressions, bookmarks, and link clicks
• Tracks follower count as a daily time series with the net change calculated for each day
• Shows the accounts that recently followed you with enough profile data to judge whether they are worth having
• Sorts by any column across the full accessible history, which X's own analytics cannot do
• Exports the results so you can keep a record and compare periods later

Two datasets, one question. Neither answers it alone.

How to Find the Tweets That Drove the Most Followers

1. Connect your X account to Circleboom Twitter and open Post Engagement Analytics. Your tweets load as a sortable table, one row per post, with impressions, engagements, engagement rate, likes, reposts, replies, profile clicks, URL clicks, and video metrics as columns.
2. Sort by profile clicks, descending. Do this before you look at anything else. This is the closest thing X gives you to follower attribution, because almost nobody follows an account without visiting the profile first. The tweets at the top of this sort are the ones that made strangers curious enough to check who you are.
3. Compare that list against your likes sort. Sort by likes descending and put the two lists side by side. The overlap is usually smaller than people expect, and the tweets that appear high on both are your genuine anchors. The ones that rank high on likes and low on profile clicks are entertaining your existing audience without recruiting anyone.
4. Calculate the click through rate rather than trusting raw numbers. Profile clicks divided by impressions is more useful than profile clicks alone. A post with 400 profile clicks from 90,000 impressions did worse per person reached than one with 60 from 3,000. The second tweet is the one to study, because it converted at a far higher rate and simply needs more reach.
5. Open Followers' Growth and find the daily change chart. This tracks your follower count over time and calculates the net change for each day, which is the second half of the evidence.
6. Line up the dates. Take your top profile click tweets, note the date each was published, and look at what the follower change did on that day and the one after. A tweet that generated a lot of profile clicks and sits on a visible growth spike is confirmed. One that generated clicks with no matching bump attracted attention but did not convert it.
7. Read the size of the spike, not just its presence. Compare each spike against your normal daily baseline. An account gaining 5 a day that gained 40 on a Tuesday has found something. An account gaining 200 a day that gained 210 has not.
8. Check what the spike actually brought in. Open the recently followed accounts view and look at who arrived. A viral post that delivers 300 dormant or bot accounts is not the growth it appears to be, and the follower count will quietly give most of it back.
9. Look for the pattern, not the winner. One tweet is an anecdote. Take your top ten by profile clicks and look for what they share: format, topic, opening line, thread versus single post, whether they had media. That shared property is the finding. The individual tweet is not repeatable; the property is.
10. Check the profile the traffic landed on. Every one of those profile clicks arrived at your bio, your pinned post, and your recent timeline. If the conversion rate from clicks to follows is poor across all your best posts, the tweets are working and the profile is not, and that is a much easier fix.
11. Export the ranked table and keep it. Comparing this quarter's top ten to last quarter's is where the real learning is, and neither X nor memory will hold that for you.
12. Schedule more of what worked, at hours when people are awake to see it. A high converting format published into a dead window converts nobody. Feed the finding back into the schedule rather than just noting it.

Why Profile Clicks Is the Metric That Matters

A follow is the end of a short funnel that most people never look at.

Someone sees the post. Something in it makes them want to know who wrote it. They tap the name. They read the bio, glance at the pinned tweet, scroll a few recent posts. Then they decide.

Likes measure step one and stop there. Profile clicks measure step two, which is the only step where a stranger does something a follower would not. That is why the metric predicts growth and likes do not. What a profile click actually means on X is worth reading if the number has never been on your radar.

📌 A tweet that converts well is one where the content raises a question that your profile answers. Advice implying expertise, a strong claim that makes people check your credentials, a result that makes them wonder what else you do. Posts that are self contained and complete leave nothing to investigate, which is why they get liked and forgotten.

This also explains a pattern most accounts eventually notice: the tweet that went viral often brought in fewer followers than a quieter one from the same week. Reach and recruitment are not the same thing, and going viral is only useful if the traffic converts. Regenerating your viral tweets is the follow up worth running once you know which ones actually earned something.

For the wider metric picture, the complete guide to X analytics metrics and the performance metrics worth tracking both cover what each number is actually telling you. Whether likes count as impressions and seeing your total impressions clear up two of the most commonly confused ones.

What X Does Not Give You, and Why This Is Correlation

Being straight about the method matters, because the confidence you place in the answer should match how the answer was produced.

X does not attribute followers to posts. No tool can conjure a metric the platform does not expose. What you are doing here is correlating two real datasets: a per tweet profile click count and a daily follower change. When a high click post sits on a clear spike, the inference is strong. When several posts went out the same day, or something happened off platform, attribution gets blurry and you should treat it as blurry.

Two limits worth knowing up front.

The tweet table works on the accessible history window, which is up to the 3,200 most recent posts for standard access. Anything older than that falls outside what the API returns, so a long running account is analysing its recent era rather than its whole life.

⚠️ Follower growth history begins when the account is connected to Circleboom, not retroactively. There is no way to reconstruct last year's daily changes after the fact, which is a good argument for connecting the account before you need the data rather than when you do. Free ways to track follower growth over time and tracking followers daily, weekly, and monthly cover why a continuous record beats a snapshot.

Native analytics will not close the gap either. Checking analytics without Premium and why you sometimes cannot see tweet analytics at all describe the ceiling, and exporting analytics is how you keep anything across periods.

Reading the Growth Chart Without Fooling Yourself

A spike with no matching post. Someone with reach mentioned you, or you appeared in a recommendation surface. Why you are suddenly getting new followers covers the usual causes, and it is worth finding out because it is often repeatable.

A spike that decays within a week. Classic viral churn. Attention arrived, curiosity was satisfied, people left. Why people follow then unfollow explains the mechanism, and tools for monitoring unfollowers tell you how much of a gain survived.

A spike made of bots. Sudden growth with no matching engagement is usually not real people. How many of your followers are bots and a bot follower check settle it quickly.

Flat growth despite strong engagement. The posts are performing and the profile is not converting. Bio, pinned post, and the top of your timeline are doing the work at that point.

Slow steady growth with no spikes at all. Often the healthiest pattern there is. Organic follower growth, gaining followers organically, and going from zero to ten thousand all describe accumulation rather than events.

Missed gains you never noticed. X does not reliably show new follower notifications, so monitoring new followers without relying on notifications is worth setting up.

Turning the Finding Into More Followers

Do the property, not the post. If eight of your top ten profile click tweets are single posts making a specific claim, that is your format. Repeating the exact tweet is not the lesson.

Fix the landing page. Profile clicks that do not convert are a bio problem. This is the highest leverage change available and the one nobody makes.

Expand the converters. A tweet that made people investigate you is a thread waiting to happen, and deciding what to post from past analytics, identifying your most engaging tweet, and sorting tweets by likes all feed the same queue.

Give converters better hours. Best time to post on X matters more for a recruiting post than an entertaining one, because you need strangers present, not just regulars.

Keep teaching. Educational posts build authority and pull follows more reliably than most formats, and ten minutes a day of deliberate activity compounds faster than occasional bursts. Improving engagement rates and increasing impressions are the supporting work.

FAQ

Does X tell you how many followers a tweet brought in?

No. There is no per tweet follower attribution anywhere in X's analytics. The workable method is to rank posts by profile clicks and correlate the top ones against your daily follower change.

Why profile clicks instead of likes or impressions?

Because a follow requires a profile visit and a like does not. Likes mostly come from existing followers, so they measure how well you serve the audience you have. Profile clicks measure how many strangers wanted to know more, which is the step immediately before a follow.

How far back can I analyse?

Post data covers up to the 3,200 most recent tweets for standard API access. Follower growth history starts from when the account was connected to Circleboom and cannot be backfilled, so connect it before you need the record.

What if several tweets went out on a high growth day?

Then attribution is genuinely ambiguous and you should say so. Use profile clicks to rank the candidates on that day, and look for the pattern across many days rather than trying to settle a single one.

My biggest tweet brought almost no followers. Is that normal?

Very. Reach and recruitment are different outcomes. A widely shared post that is complete in itself gives nobody a reason to investigate the author. Check its profile click rate against its impressions and you will usually find it converted poorly despite the volume.

Should I judge by raw profile clicks or by rate?

Rate, for learning. Profile clicks divided by impressions tells you which content converts, independent of how much reach it happened to get. Raw counts tell you which posts delivered the most total traffic, which matters for a different question.

The spike disappeared a week later. What happened?

Viral churn, usually. People followed out of curiosity, the curiosity was satisfied, they left. Judge a post's contribution by the followers still there a month on, not by the peak.

Can I see exactly who followed me after a specific tweet?

Not with timestamps tied to the post, but you can see the accounts that recently followed you with enough profile detail to judge quality, and checking someone's new followers covers the equivalent view for other accounts.

Final Thoughts

The tweet you are proudest of is rarely the one that built the account. That is uncomfortable, and it is the most useful thing the data will tell you.

Finding the tweets that drove the most followers on Twitter means giving up on a metric X never built and assembling the answer from two it did: profile clicks per post, and follower change per day. Where those two line up, you have your answer. Where they do not, you have learned something almost as valuable about the difference between attention and recruitment.

With Circleboom Twitter, you can rank every post in your accessible history by profile clicks, track your daily follower change alongside it, and find out which posts on X actually earned you an audience rather than just applause.

https://circleboom.com/twitter-management-tool/twitter-post-analytics