← Writing

How to analyze a competitor's Instagram Reels

26 July 2026 · Joonas Soininen

Your competitor posted a Reel last Tuesday. It did 38,000 views. Yours, same day, did 200. You watched theirs five times. Someone talking to camera with text overlays. Nothing fancy. So why did it work?

Most competitor analysis advice tells you to track likes, count hashtags, and monitor posting frequency. That is benchmarking, and it tells you that something worked. It does not tell you why, and the why is the only part you can reuse.

First, what Instagram actually rewards

Skip the algorithm folklore. Here is what Meta and Instagram have said themselves, with dates, because most of what circulates on this topic is unsourced.

Watch time, likes, and sends are the top three ranking signals. Adam Mosseri said this in January 2025. His hedge matters: sends are "slightly more important for unconnected content", meaning for reaching people who do not follow you. Slightly more important than likes. Not a magic lever.

Original content is now most of what gets recommended. On Meta's Q4 2025 earnings call in January 2026, CFO Susan Li said Instagram "grew the prevalence of original content in the US by 10 percentage points in Q4, with 75% of recommendations now coming from original posts." Instagram's own policy, from April 2024, is blunter: "when we find two or more identical pieces of content on Instagram, we will only recommend the original one." In April 2026 that rule was extended from Reels to photos and carousels.

Recency got dramatically more important. On the Q1 2026 call, Li said "same-day posts now represent more than 30% of recommended Reels on both Instagram and Facebook, more than double the levels one year ago." If you have been studying competitors' evergreen hits from eight months ago, that is a real shift in what the feed wants.

Reels are roughly half of Instagram. Sensor Tower measured Reels at 46% of US Instagram time in 2025, up from 37% in 2024.

Notice what is not on that list: hashtag counts, posting times, follower counts, and the precise-sounding thresholds that get repeated across marketing blogs. Take "a 3-second hold rate above 60% gets 5-10x more reach." We went looking for the source. It originates in a single post by a competing video tool, supported by the sentence "I've analyzed performance data across multiple accounts, and the correlation is undeniable." No dataset, no sample size. It also sets a threshold on "3-second hold rate", which is a Meta Ads term, not something organic creators can even see.

If a number about the algorithm does not come with a date and a source, assume someone selling something invented it.

Step 1: Pick competitors you can learn from

Look for five to eight accounts in your niche or an adjacent one, posting Reels several times a week, with a few Reels that clearly beat their own average.

That last criterion is the one people skip, and it is the important one. Rank by how far a post beat that account's own typical performance, not by raw views. A 500k-view Reel from an account that averages 400k is a normal Tuesday. A 50k-view Reel from an account that averages 5k is a 10x outlier, and it is telling you something specific.

Sorting by raw views just re-discovers which accounts are big. Sorting by outlier strength surfaces the individual videos that did something unusual, which is the only thing you can actually learn from.

Step 2: Break down the hook

Instagram's own guidance on hooks is qualitative. Mosseri talks about "the first few seconds." There is no official number, and anyone quoting a precise attention-span figure attributed to "Instagram's data" is making it up. The practical point stands anyway: if nobody stays past the opening, nothing else matters.

For each outlier Reel, name the hook type:

Write down the exact opening line, whether there is on-screen text, and the emotion it creates. Then the honest question: would you have stopped scrolling?

After 15 or 20 hooks, you will notice most creators in a niche lean on two or three types repeatedly. That is not laziness. It is a formula that works on their audience.

Step 3: Map the structure

After the hook, high performers are rarely random. A common shape:

As you watch, record where the energy shifts, whether text overlays mirror the audio or add a second layer, the cut rhythm, and whether the audio is original, trending, or voiceover. You are not looking for a magic cut interval. You are looking for whether the cuts land where attention would otherwise drop.

Step 4: Work out why it got sent

Since sends carry real weight for unconnected reach, this deserves its own pass. The question is not "did people like this?" It is "why would someone send this to a specific person?"

Step 5: Find the gap

The most valuable output is not a list of things to copy. It is the thing nobody covers. After 15 to 20 Reels, look for topics your audience asks about that no competitor addresses, formats nobody uses, depth everyone skips, and audiences everyone ignores. A good session ends with three specific things you can film this week.

Step 6: Track it over time

One analysis is a snapshot. The pattern is the strategy. Analyse three to five new Reels a week and note which hook types recur.

Two caveats. Given the originality rule, this is for understanding why things work, not for reposting. And given that same-day posts are now more than 30% of recommended Reels, patterns go stale faster than they used to.

The honest version of the shortcut

Everything above takes hours. That is the real reason most people track vanity metrics instead: the deep version does not scale by hand.

That is what we built Silma to do. It watches the video frame by frame, listens to the audio, and breaks down hook, structure, pacing and text overlays. It tracks the competitors you choose and ranks their posts by how far each beat that account's own median, so you see genuine outliers rather than big accounts having a normal day.

The part we care about more: before you post your own video, Silma checks it and tells you what to fix. Analysing other people's content only matters if it changes what you ship.

The free plan is 25 credits a month and needs no card. A pre-check costs 5. Start free, or see what a credit buys.

FAQ

How often should I analyze competitor Reels?
Three to five a week, with a fuller pass monthly. Since same-day posts are now more than 30% of recommended Reels, older patterns date faster than they used to.

Can I use the same hooks my competitors use?
Hook types are universal. Applying a curiosity gap to your own topic is not copying. Reposting their actual content is, and only the original gets recommended.

Should I sort competitor posts by views?
No. Sort by how far a post beat that account's own average. Raw views mostly tell you which accounts are big.

Do I need a tool for this?
No. A spreadsheet and this guide work. A tool earns its place when manual analysis stops scaling.

Sources