Let Your Own Post Data Choose the Aspect Ratio

"Make it vertical" has become the default instruction for almost any piece of video content, to the point where it barely gets questioned anymore. It's not a bad default — a huge share of social video is watched full-screen on a phone, and 9:16 wins that context more often than not. But it's still a default, not a finding. It was true in aggregate for a lot of accounts, at some point, and it got repeated until it stopped sounding like a recommendation and started sounding like a rule. The account it's actually true for is a specific question, and most creators never ask it about their own numbers before reformatting everything they make.

Where the rule came in ahead of the evidence

The vertical-first instinct exists because platforms built their default surfaces — Reels, TikTok's For You feed, Shorts — around a full-screen phone frame, and that changed what got the most inventory and the easiest discovery. That's a real structural advantage for 9:16 in those specific surfaces. What it isn't is a guarantee that every format on every platform, for every account, performs worse outside that shape. A landscape explainer that lives on YouTube search, a square carousel that performs well in a feed context, a widescreen clip embedded on a landing page — none of those are "wrong" just because the vertical-first rule of thumb exists. They're a different format serving a different surface, and the only way to know whether your account's version of that format is actually underperforming is to look at what it's already done.

What your own numbers actually contain

Versely's agent has a tool built for exactly this read: get_content_insights breaks your recent posts down by platform and by format — image, video, carousel, text — and returns post counts, average and median views and engagement for each combination, your top three posts, and your best posting hour and day where the timestamps allow it. That's not a generic best-practice table. It's a summary of what your specific account, posting to your specific audience, has actually produced.

The detail worth noticing in how that tool is built: its own contract says to use it before creating content aimed at your channels, specifically to recommend which format and platform performs best — and it explicitly says it's not a mandatory gate. It's not inserted as a blocking step you have to clear before you're allowed to post. It's a read you can take when you're planning, get an answer from in seconds, and either act on or set aside. That framing matters because it removes the excuse for skipping it — checking doesn't cost you a workflow step, it costs you the time to ask a question you were probably about to guess the answer to anyway.

Reading the split without fooling yourself

A platform-and-format breakdown is only useful if you read it correctly, and the most common way to misread it is trusting an average that one outlier post is quietly carrying. If square posts on one platform show a strong average but a much lower median, that's a tell — one post did enormously well and the rest of the format's history is unremarkable, which is a different situation than "square consistently outperforms" and calls for a different decision. Median next to average, in the same breakdown, is what makes that distinction visible instead of hidden inside one number.

The top-three-posts detail in the same read exists for the same reason. If your top three posts across every format all happen to be one shape, that's a much stronger signal than a format-level average, because it's telling you what actually got the engagement rather than what a spreadsheet formula produced. And the best-hour-and-day detail, when your timestamps support it, answers a question that's easy to conflate with format: sometimes a format looks like it underperforms when it was actually just consistently posted at a worse time than everything else, which is a scheduling fix, not a reformatting one.

None of this replaces judgment — a small sample of posts in one format can't tell you as much as a large one, and a new platform you just started posting to won't have enough history yet to draw a real conclusion from. But it's a real read on your own account, which is a categorically better starting point than a rule of thumb built for accounts in general.

From the read to the reframe

Once the data points at a format, the actual reformatting is a separate, simple step. Change a video's aspect ratio walks through the general version of that job on Versely — reframing an existing clip to 9:16, 1:1, or 16:9 without regenerating it from scratch, which matters because aspect ratio is a composition decision baked in at generation time, not just a crop applied afterward. A clip generated wide and cropped down to vertical after the fact usually loses the framing that made the wide version work in the first place — the subject was placed for the frame it was actually made for. That's the argument for checking the format decision early, before generating, rather than generating in whatever shape felt default and hoping a reframe fixes it later.

Versely's guide to picking the right format per platform covers that fuller tool chain in more depth — reading the insight, sanity-checking it against individual outlier posts, then reframing toward the winner — if the strategy question above is settled and the next step is execution.

Walkthrough: checking before you commit to a shape

1. Ask Versely's agent something like: "How are my posts performing lately — is video, image or carousel doing better, and does that differ by platform?" That routes to get_content_insights, which returns the platform-and-format breakdown described above.
2. Look at median next to average for the format you were about to default to. If they're close, the format's performance is broad-based. If the average is well above the median, one post is likely doing most of the work — check whether it shares anything with your other posts in that format before trusting the average alone.
3. Cross-check against your top three posts overall. If none of them share the format you were about to commit to, that's worth weighing against the averages, not ignoring in favor of them.
4. If timestamps support it, check whether the format you're deprioritizing was simply posted at worse times than the one that's winning — that's a scheduling question, not a format one, and conflating them leads to the wrong fix.
5. Only then reframe. If the data points to a shape different from what you'd have defaulted to, reformat the clip to that shape — or better, generate the next one at that ratio from the start, since a clip composed for its target frame from the beginning holds up better than one cropped into it after the fact.

The point isn't that vertical is wrong, or that any other shape is secretly better. It's that "everything should be vertical" is a claim about accounts in aggregate, and the only account your next post actually has to perform for is yours. Read the format breakdown first; let it, not the rule of thumb, make the call.