Twitter Following Ratio: What the 0.6 Rule Means

Jump to calculating the ratio, the 0.6 threshold, effects on reach, follow limits or reviewing your list.

How to calculate your Twitter follower to following ratio

Use the two counts on your X profile. Followers are the accounts that follow you. Following is the number of accounts you follow. State the direction whenever you compare results.

This article uses following divided by followers when discussing the 0.6 threshold. Lower is better in the historical adjustment. In the reciprocal version, higher is better. The direction changes; the account does not.

A 0.6 following ratio means following six accounts for every ten followers. It does not mean you should keep only 60% of your present following list. Your target depends on both counts.

With zero followers, the ordinary following ratio has no finite value. With zero following, the reciprocal ratio has no finite value. Do not treat a calculator's placeholder as an account reputation score.

Where the 0.6 threshold comes from

Twitter's 2023 UserMass calculation reduces account weight when following exceeds 500 and its ratio exceeds 0.6. The code uses (following + 1) / (followers + 1), then divides weight by exp(5 × (ratio − 0.6)). The adjustment also applies after the verified account branch.

The arithmetic below isolates that one historical formula. These percentages are remaining account weight, not remaining views.

Calculated from the published formula, assuming its following count condition is met. Ratios are inputs to the smoothed formula above, rather than rounded profile ratios.

The added ones matter at a boundary. A displayed 600 / 1,000 equals 0.6, while 601 / 1,001 is slightly higher. A rounded profile ratio and an exact internal comparison therefore need not agree. Leave room around a target instead of treating one follow as an algorithm trick.

There is a second reputation adjustment

A separate file, Reputation.scala, adjusts a PageRank based reputation input when following exceeds 2,500. It also references 0.6, but its divisor depends on following count and is capped at 50. Below the ratio threshold, the divisor is clamped so it does not increase the input.

For example, substituting 3,000 following and 3,000 followers into that formula gives a divisor of about 12.14. That leaves roughly 8.24% of the input to this adjustment. It does not establish that the account receives 8.24% of its previous impressions.

The two files explain why articles cite different following counts. One threshold is 500; the other is 2,500. They belong to different calculations. Neither number is the total number of accounts X allows you to follow.

Does your following ratio affect reach on X?

In the published historical system, an excessive ratio reduces reputation inputs under the conditions above. That is a concrete algorithmic effect. It is not evidence that every account above 0.6 suffers a fixed percentage loss of impressions today.

The Tweepcred documentation describes a network influence calculation using PageRank. This is about relationships within a network, rather than counting every follower as an identical unit of authority.

X also publishes a newer For You feed repository. It describes recommendation models, filtering and account signals, but the older UserMass file alone cannot establish which formula and parameters a particular production request uses today.

Our practical recommendation is to treat 0.6 as a network review target with a documented historical rationale. It is especially useful when a following list grew through indiscriminate reciprocal follows and no longer reflects what the account is about.

Consider two hypothetical accounts with the same ratio. One follows researchers, customers and peers whose work it reads every day. The other follows thousands of unrelated accounts in the hope of getting follows back. The ratio describes the shape of those networks; it cannot tell you their value by itself.

Keep useful relationships. A carefully chosen network gives you better material to read and more relevant conversations to join. Removing it for a number defeats the purpose of being on the platform.

The 5,000 following limit is a different rule

X's current follow guidance says that once an account follows 5,000 others, further following depends on its individual ratio of followers to following. X does not publish one universal ratio for that limit.

That is an operational limit on following more accounts. The 0.6 historical reputation threshold is a different mechanism. Reaching one target does not prove that the other condition is satisfied.

If X says you cannot follow more people, read the actual message and compare it with the follow guidance. A daily limit, fast following behavior and an account ratio limit require different responses. Do not assume a low reach article explains every following error.

Work out a useful target for your own account

For a simple profile ratio of 0.6 or lower, multiply followers by 0.6 and round down to a whole account. This is ordinary arithmetic for planning a list review, not a reproduction of every internal reputation rule.

These examples are not automatic instructions to remove that many people. They show the size of the difference. The historical formulas still have their own count thresholds and smoothed ratios.

You can also improve a ratio by attracting followers while keeping the following count steady. At 900 following, reaching 1,500 followers produces a simple ratio of 0.6. That may be a better direction than deleting 300 useful connections from an account with 1,000 followers.

Improve the list without damaging the network

Start with why you followed someone. Do you read their work, learn from their experience, serve their industry or have a relationship worth maintaining? Those are good reasons even if the person never follows you back.

Look first for accounts you no longer recognize, topics you have stopped following and reciprocal follows that never became useful. Review profiles before deciding. Someone who posts rarely can still be a valuable contact.

Make deliberate changes over time. X's authenticity policy prohibits follow churn and indiscriminate following or unfollowing in large numbers over a short period. A ratio target is not a reason to turn a normal list review into bulk account manipulation.

Do not buy followers to improve the denominator. A larger public count does not create readers who care about your work. It also leaves the underlying reason for the weak network unresolved.

Put more effort into the part you want to keep: answer relevant questions, publish things those readers will use and develop conversations with people in the same field. Our guide to getting followers on Twitter covers audience development, while the reply reach guide explains where those conversations get discovered.

Common Twitter ratio questions

What is a good Twitter follower ratio?

First specify the direction. A following to follower ratio of 0.6 corresponds to about 1.67 followers per followed account. It is a useful review target derived from historical code, not a universal definition of a good account.

Is a 1:1 ratio bad?

A 1:1 ratio means the two counts are equal. In the historical adjustments, following count also determines whether the formula applies. For a small personal network, the number alone does not demonstrate a reputation penalty.

Does a blue checkmark cancel the ratio adjustment?

The historical UserMass adjustment happens after its verified branch. A badge and a following ratio therefore address different parts of that calculation. For current subscription benefits, use the separate X Premium comparison.

How soon will unfollowing people improve impressions?

There is no supported timetable that turns a list edit into a guaranteed view increase. Judge the review by whether your network becomes more useful, then examine reach across comparable posts and conversations.

Research checked October 22, 2026. Historical formula results and profile examples are calculated from the linked code or stated counts. They are not measured changes in current X impressions. Current follow restrictions are documented separately from the 2023 reputation implementation.