What the Platforms Say About Ranking, and What Follows From It

One platform, several ranking systems

Instagram is explicit that its parts are ranked separately: Feed, Stories, Explore, Reels and Search each use their own system, tuned to how people use that surface. Advice that treats "the Instagram algorithm" as one thing is describing something that does not exist, and a tactic that lifts Reels reach can be irrelevant to Feed.²

The same split holds on YouTube, where search results are ranked on how well a title, description and the video itself match a query, while the home feed is built from watch history and what similar viewers watched.⁴

What each platform says it uses

Two things follow. First, the follower graph has not been abolished: it still supplies inventory on Instagram Feed and it still brings an audience with it everywhere. What is true is narrower and better documented, which is that on the recommendation surfaces the follow relationship is not a direct ranking input.¹²

Second, completion is weighted rather than binary. TikTok describes a strong indicator of interest, such as a viewer finishing a longer video, as carrying more weight than a weak one. That is an argument for making the length match the idea, not for making everything short.¹

Organic and paid are not the same problem

Everything above is organic distribution. Paid placement is bought, so the question is not whether the system will show the video but what the impression costs and what it returns. The two are often reported together and should not be.

• OrganicJudged on whether the content earned distribution. Reach, watch-through and saves are the diagnostics; leads are the outcome.
• PaidJudged on cost per result. Creative still decides most of the outcome, but the lever is budget allocation across tested concepts rather than whether a recommendation system picked the video up.
• The connection between themA post that earned attention organically is a tested concept. Promoting it is cheaper than writing an ad from nothing. That is the real reason to run both, and it is an argument about testing cost, not about reach.

What this means for a brief

1. 01Write per surface, not per platformDecide which surface a piece is for before it is made. A Reel built for the recommendation feed and a Feed post for existing followers are different briefs with different success conditions.
2. 02State a creative hypothesisWrite down what you expect to happen and why before publishing. Without it, a result is only a number, and a month of posting produces no learning.
3. 03Match length to the ideaCompletion is weighted, so a short video that finishes beats a long one that does not, and a long one that holds beats a short one that says nothing. There is no universal correct duration and we are not going to invent one.
4. 04Test production value rather than assuming itWhether polished or unpolished material performs better is an empirical question that differs by audience and category. Run both against each other in your own account and let your numbers decide.
5. 05Measure against leadsTag links, feed conversions into your analytics or CRM, and report on what the content produced. Reach and engagement rate are diagnostics that explain the outcome, not the outcome.

Related