Every social media manager has been there: you post a video, the views roll in, and for a brief moment, it feels like you’ve cracked the code. Then the next video flops, and you’re back to square one, staring at your analytics dashboard wondering what the hell went wrong.
The problem isn’t your content. It’s the metrics you’re using to measure it.
View counts have become the metric equivalent of vanity sizing. They make you feel good about yourself, but they don’t tell you anything useful. Some platforms count a scroll-past as a view. Others count a split-second glance as engagement. And yet, brands everywhere still treat view counts like they’re the scoreboard that matters.
Here’s the uncomfortable truth: the numbers that actually move the needle are the ones you’re probably not paying enough attention to.
The Problem With Looking Up
Let’s be clear: I’m not saying views are useless. They’re not. If you’re running a brand awareness campaign and your goal is reach, view counts matter. Hootsuite’s Former Social Media Strategist Eileen Kwok puts it well: from an awareness standpoint, you care about how widely a video is shared and whether the algorithm favored it.
But here’s what most people miss. Instagram recently made views the primary metric across all content, not just Reels. On the surface, that sounds like a win for transparency. Dig a little deeper, and you’ll notice something troubling. When a platform shifts its primary metric, it’s usually because that metric benefits the platform more than it benefits you.
Think about it. More views mean more ad inventory. More engagement signals mean the algorithm keeps people scrolling. The platform wins either way.
What matters far more than how many people saw your video is whether they stayed. That’s where average watch time comes in, and it’s one of the most underutilized metrics in social video. If you post a one-minute video and the average watch time is three seconds, your hook is broken. If you post a three-second video and people watch it five times, you’ve done something right.
This is the kind of insight that transforms your strategy. View counts tell you what happened. Average watch time tells you why.
The Engagement Layer Nobody Talks About
Likes are fine. Comments are better. But saves? Saves are where the real story lives.
When someone saves your content, they’re signaling something powerful: this was useful or memorable enough to come back to later. That takes effort. That takes intent. A like is a split-second impulse. A save is a deliberate choice, and the algorithm knows it.
Instagram’s Head, Adam Mosseri, has been pretty direct about this. Engagement metrics rank among the most important to track, and saves carry particular weight because they indicate long-term value, not just momentary approval.
The same logic applies to shares, though in a different way. Shares are word-of-mouth marketing in digital form. When someone sends your video to a friend or posts it to their story, they’re vouching for it. That kind of organic amplification is what algorithms reward, and it’s why Kwok calls shares a key metric from a consideration standpoint.
So here’s my take: stop celebrating likes as if they’re validation. Start looking at saves and shares. They’re harder to earn, but they actually mean something.
Retention Tells You the Whole Story
If average watch time is the headline, retention is the full article.
Retention data shows you exactly when viewers lose interest and leave. Most platforms surface this information in one form or another, though they might call it different things. YouTube calls it audience retention. TikTok shows it differently. The insight is the same.
Mosseri has confirmed that Instagram’s algorithm pays close attention to total seconds viewed, not just completion percentage. His reasoning is worth understanding: a viewer who watches 10 seconds of a one-minute video is providing the same signal as someone who watches 10 seconds of a 10-second video. The platform doesn’t want to punish longer content, so it rewards total engagement time instead.
This changes how you should think about video length. There’s no universal right answer. A 15-second video that gets watched in full will outperform a 60-second video where people drop off at the five-second mark. But here’s what most people miss: you can actually make longer videos work if you earn every second.
The data is clear. Audience retention drops significantly after the 60-second mark unless your storytelling is strong. If you’re making videos that are four or five minutes long, you’d better have a damn good reason, and the retention graph should prove it.
What You Should Actually Track
Here’s where it gets practical. The metrics you track should depend on what you’re trying to achieve.
For brand awareness, views and reach matter, but don’t stop there. Add shares into the mix because that’s how your content actually spreads beyond your existing audience.
For community building, engagement rate is your north star. We’re talking likes, comments, saves, and shares. But don’t just look at the numbers. Look at the quality. A hundred comments from real people discussing your content is worth more than a thousand likes from bots.
For conversions, you need to trace the path from view to action. That’s where conversion rate comes in. Divide your conversions by your video views or clicks, depending on how you’re tracking. If your video has a clear call-to-action and people aren’t following through, the problem isn’t the metric. It’s the CTA itself, or the landing experience after the click.
I’m going to say something that might be controversial. Most brands pick too many metrics and track none of them well. Pick two or three KPIs that align with your actual goal, and ignore the rest. The dashboard overload is costing you more than you think.
The Human Element
Mariam Ordubadi, Head of Marketing at Aequilibrium, put it simply: her team obsesses over hooks. The first few seconds decide everything. One of their most reliable formulas is the problem-solution hook, something specific and urgent that promises immediate value. Think “don’t have time to make dinner but need to finish your groceries? This hack will save you time and food rot.”
That works because it’s specific. It doesn’t try to be everything to everyone.
She also runs what she calls the “mute test.” If your first five seconds can’t stop someone from scrolling without sound, you’ve already lost them. The visual has to do the work.
This is where I think a lot of video strategy goes wrong. People treat metrics like they’re solving an equation. But video is emotional. It’s visual. The numbers are symptoms, not the disease. If your retention graph shows people dropping off at 45 seconds, that’s a data point. What you do with it is a creative decision.
The Real Talk
Here’s what I’ve learned from watching thousands of videos across every platform: the numbers that matter most are the ones that tell you whether you’re connecting with actual humans or just generating noise.
View counts can be inflated and meaningless. Likes are vanity. But when someone saves your video, shares it with a friend, or watches it all the way through, something real happened. That’s not just a metric. That’s proof that your content added value to someone’s day.
The rest is just math.
If you’re serious about video, stop optimizing for the algorithm and start optimizing for attention. The algorithm will follow.
