> **Methodology in brief.** We separate the **measured** from the **estimated**. Measured: our first-party LinkHub data (**657,722 real comments** with impressions) — a comment's raw reach, its likes and replies. Estimated / third-party: the relative weight of the 3 signals (reaction, comment, share) in LinkedIn's scoring. **Important: for the share, we have no first-party data** — its weight is handled via dated third-party consensus, flagged as such. We never derive a first-party stat from an estimate.

## Key takeaways

- **The consensus ranking: share ≈ comment > reaction.** The comment and the share are the two strong signals; the reaction (like) is the weak one. *(van der Blom 2025, AuthoredUp 2025)*
- **The comment is the signal we can measure.** Our data: **179 average impressions** (35 median), **0.90 like** and **0.71 reply** per comment → broad real reach, concrete conversational effect. *(LinkHub, n = 657,722)*
- **The reaction (like) is the weakest signal**: passive, a click, ~199 measured impressions for a single like triggered by a comment.
- **The share: no first-party data on our side.** Consensus ranks it high (it exposes content to a new audience), but public reshare has declined sharply; sharing via DM is valued as a strong-interest signal. *(third-party, not first-party verified)*
- **Sisters to read**: [comments vs likes](/en/blog/commentaires-vs-likes-linkedin) (the comment/like duel) and [comment vs like weight](/en/blog/poids-commentaire-vs-like-algorithme) (the multiplier debate). Here, we **add the share** and rank all **three**.

## 1. The ranking of the 3 signals (consensus)

On the hierarchy of LinkedIn engagement signals, sources converge on an order — not on exact figures. Here's the 2025-2026 consensus ranking:

| Rank | Signal | Engagement type | Data status |
|---|---|---|---|
| 🥇 | **Comment** | active, conversational | **measured** (LinkHub) + consensus |
| 🥇 | **Share / repost** | exposes to a new audience | third-party consensus (no first-party) |
| 🥉 | **Reaction (like)** | passive, a click | measured indirectly + consensus |

The comment and the share contend for first place depending on the source; the **reaction is unanimously the weakest signal** ([van der Blom, 2025](https://www.linkedin.com/posts/richardvanderblom_chapter-1-algorithm-insights-report-2025-activity-7322514599126130688-Q895); [AuthoredUp, 2025](https://authoredup.com/blog/linkedin-algorithm)). Let's break down each.

## 2. The comment: the only signal we can quantify

It's the signal for which we have **solid first-party data**. Across **657,722 real comments** whose impressions we recorded:

| Metric per comment | Average | Median |
|---|---|---|
| **Impressions** | **179** | **35** |
| Likes received | 0.90 | — |
| Replies received | 0.71 | — |

**Reading.** The average comment is seen **179 times** for just **0.90 like** — that's ~199 impressions for a single like. And it earns **0.71 reply**, almost as many as likes: it's a conversational back-and-forth, exactly the kind of interaction the algorithm values most. It's measured, at scale, and unambiguous: the comment exposes you to a wide audience. Detail in the sister study [comments vs likes](/en/blog/commentaires-vs-likes-linkedin).

## 3. The reaction (like): the weak signal, and it's measurable

The reaction sits at the bottom of the hierarchy — everyone agrees on this. It's a **passive** signal: a click, no thought, no content. In our data, a comment generates 179 impressions for 0.90 like: in other words, the like is rare and cheap as a signal. The algorithm prioritizes relevant comments over vanity reactions ([van der Blom, 2025](https://www.linkedin.com/posts/richardvanderblom_chapter-1-algorithm-insights-report-2025-activity-7322514599126130688-Q895)).

The debate over **how much** the comment weighs vs the like (the famous "2x to 15x" multiplier) is unverifiable from the outside — we settle it honestly in [comment vs like weight](/en/blog/poids-commentaire-vs-like-algorithme). What's certain: **reaction = weakest of the three.**

## 4. The share: the signal we CANNOT measure (honestly)

Let's be transparent: **we have no first-party data on the share.** Our dataset covers comments, not reposts. Everything below about the share is **third-party consensus**, flagged as such.

What the sources say:

- **The share exposes content to a new audience.** Unlike a like, resharing pushes content beyond the original audience — hence its high value in most rankings ([Botdog, 2025](https://www.botdog.co/blog-posts/linkedin-algorithm-2025)).
- **Sharing via DM is valued.** Sending a post in a private message reads as a **strong-interest** signal (the content is worth a one-to-one share) — often ranked above the public repost ([AuthoredUp, 2025](https://authoredup.com/blog/linkedin-algorithm)).
- **The "naked" public repost has declined.** A repost with no added comment contributes little; LinkedIn reduced its impact, favoring reposts with added thought. *(third-party, disputed)*

**Our honest stance:** the share is probably a strong signal *when it's qualitative* (DM, or commented repost), but we cannot quantify it the way we quantify the comment. Be wary of rankings that give a precise "share weight": none has access to LinkedIn's scoring.

## 5. Which signal should you aim for when short on time?

Practical answer: **the comment.** Not because it "wins" in absolute terms, but because it's:

- **the only one you control at will** (you can't force others to share you, but you can comment under whomever you want);
- **the only one we measure** (179 impressions/comment);
- **the most efficient for borrowed reach**: a comment posted early under the right creator exposes you to their entire audience *(see [when to comment](/en/blog/quand-commenter-sur-linkedin) and [who to comment under](/en/blog/sur-qui-commenter-linkedin))*.

The like is free but nearly useless as a signal. The share is powerful but depends on others (or requires you to repost with a relevant addition). To steer your visibility actively, the targeted comment remains lever #1.

To put your comments in front of the right audience at the right time, gather your targets in [personalized feeds](/en/features/feeds-personnalises) and let AI spot the best hosts via [AI profile recommendation](/en/features/ia-recommandation-profils).

## FAQ

**What's the strongest signal on LinkedIn: like, comment or share?**
The comment and the share are the two strong signals; the reaction (like) is the weakest. The comment is the only one we measure first-party: 179 average impressions, 0.90 like and 0.71 reply per comment (n = 657,722).

**Does the share count more than the comment?**
According to sources, it's neck and neck — the share exposes to a new audience, the comment creates conversation. Honestly, we have no first-party data on the share; we can't settle it with certainty. *(third-party consensus)*

**Why is the like the weakest signal?**
Because it's passive: a click, with no content or thought. The algorithm prioritizes active interactions (comments, conversations) over vanity reactions ([van der Blom, 2025](https://www.linkedin.com/posts/richardvanderblom_chapter-1-algorithm-insights-report-2025-activity-7322514599126130688-Q895)).

**What's the difference with your "comments vs likes" and "comment vs like weight" articles?**
[Comments vs likes](/en/blog/commentaires-vs-likes-linkedin) measures the comment's underestimated value against the like. [Comment vs like weight](/en/blog/poids-commentaire-vs-like-algorithme) settles the multiplier debate. Here, we **add the share** and rank all **three** signals.

**Which signal should I focus on if short on time?**
The targeted comment: it's the only one you control at will, the only measured one, and the most efficient for borrowed reach. See [when to comment](/en/blog/quand-commenter-sur-linkedin).

## Sources & methodology

- **LinkHub dataset** — **657,722 real comments** with impressions: average 179 / median 35, 0.90 like and 0.71 reply per comment. **The share is not measured in our data** — its relative weight is handled via third-party consensus only.
- [van der Blom — Algorithm InSights Report 2025](https://www.linkedin.com/posts/richardvanderblom_chapter-1-algorithm-insights-report-2025-activity-7322514599126130688-Q895) (comments prioritized over reactions) · [AuthoredUp (2025)](https://authoredup.com/blog/linkedin-algorithm) (signal hierarchy, DM share) · [Botdog (2025)](https://www.botdog.co/blog-posts/linkedin-algorithm-2025) (share = new audience).
- Relative signal weights and share value: **unverifiable third-party estimates**, cited as directional consensus.
- Sister studies: [comments vs likes](/en/blog/commentaires-vs-likes-linkedin) · [comment vs like weight](/en/blog/poids-commentaire-vs-like-algorithme).
- Find all our studies on the [LinkHub blog](/en/blog).