In the matchup of shares and saves vs likes, shares and saves win, and it isn't close. A like is a reflex that costs the viewer nothing. A share is a referral, someone spending a little of their reputation to put your video in a friend's messages. A save is a bookmark with intent, a plan to come back and use what you said. Platforms reward the expensive actions, and shares and saves track reach far better than likes do.
What each action says about your viewer
A like is a thumb twitch. It happens mid-scroll, takes half a second, and carries almost no information beyond "I saw this and didn't hate it." People like plenty of things they would never watch twice and would never mention to a friend.
A share is a different animal. When someone shares your video, they attach their name to it in a group chat or a DM thread. They spent social capital on you. That makes shares the closest thing social platforms have to word of mouth, and word of mouth is how accounts grow past their existing audience.
A save is the most honest metric on the platform, because nobody sees it except the person who tapped it. There's no performance in a save. It means someone plans to use your content later, and if you sell services, that person deserves your attention. Someone who saves "questions to ask before you waive inspection" is probably buying a house this year.
Comments sit somewhere in the middle. A thoughtful question is a gift, and I'll take ten of those over a hundred likes, but comment counts are easy to inflate with bait, so read them for their content instead of their volume.
Why likes lie to you
Likes reward agreeable content. Pleasant, familiar, easy to nod at. The trouble is that agreeable content rarely gets sent to anyone, so you end up optimizing for applause from people who already follow you while making nothing worth passing along.
Likes also flatter you. A 2,000-like post feels like momentum even when it brought in zero new followers, so you make five more just like it. That loop can run for a year before you notice your reach never moved.
A like says I saw this, and a share says I'd put my name on it.
Why do platforms reward shares and saves?
Think about what each action gives the platform. A like gives it a data point. A share hands it a brand new viewer, someone who opened the app or a message thread because of your video, and new sessions are the thing every platform is built to collect. A save promises a return visit, since the viewer plans to come back for the content later.
Recommendation systems favor the behavior that grows the platform, so content that earns shares and saves gets shown to more people. You don't need the exact formula, and anyone claiming to have it is selling something. The observable pattern is enough. Across accounts, shares and saves track reach far better than likes, which makes them the right numbers to steer by.
What happens when you re-rank your posts by shares and saves?
The leaderboard flips, and the flip is usually the most useful thing you'll learn about your account all quarter.
A realtor I'll call Trina had 41 posts over two months. Ranked by likes, her number one was a luxury house tour: 4,100 likes, 19 shares, 12 saves. Ranked by shares plus saves, her number one was a plain talking video called "three questions to ask before you waive inspection": 512 likes, 131 shares, 89 saves. Same account, two different verdicts. The tour collected likes from other agents and daydreamers. The advice video got forwarded into group chats of people buying houses right now. Trina shifted from two tours a week to one tour and two advice videos, and her non-follower reach and follower growth both picked up over the next six weeks. Her like counts dropped. She stopped caring.
Here's how to run the same flip on your account. I call it the 20-minute re-rank:
- List your last 30 to 60 posts in a spreadsheet. On TikTok, start from the analytics file, which I walk through in how to export TikTok analytics.
- Add one column: shares plus saves, combined into a single number. If saves are missing from your export, pull them by hand from each video's own analytics screen.
- Sort by that column, descending.
- Compare the new top ten to your old top ten by likes, and write down what changed.
- Circle any topic or format that appears at least twice in the new top ten. That circle is next month's content plan.
This weighting is also the core of how Batchly scores an analytics upload. Shares and saves count most, likes count least, and the winners get clustered into pillars automatically.
Do likes matter at all?
Some. On a brand new account, likes are the first sign that anything is landing, and they work fine as a tiebreaker between two posts with similar shares. A post that pulls strong shares and strong likes at the same time is your best work, worth studying line by line. Just never let likes make a decision alone when a better number sits one column over.
Once you have your re-ranked top ten, the next move is grouping those winners into three to five lanes. That's the whole subject of content pillars explained, and it's where a pile of numbers turns into a plan.
Do this next
Tonight, take your last 30 posts and rank them by shares plus saves. Twenty minutes, one spreadsheet, and you'll probably meet a top ten you didn't know you had. Make your next three posts sequels of the new number one. If you'd rather skip the manual math, Batchly runs the scoring from a single analytics upload and hands back pillars plus ready-to-film scripts.