Resources for deciding what to film
Posts and definitions for people who run a brand’s short-form video: what to film next, how to read the accounts around you, and whether it worked.
From the blog
People swipe past your videos in the first seconds. What to change
What a normal Reels skip rate looks like, what teams try when people swipe away early, and four ways to rewrite one opening that you can copy.
The brief that gets back the video you wanted
A creator or a colleague filmed it and it isn’t what anyone pictured. What people send before a shoot, where each one breaks, and a one-page brief to copy.
A video got far fewer views than usual. How to find out why
A video you expected to do well got a fraction of your usual views. Three possible reasons, how to tell which one it is, and what to tell your manager.
They post three times as often. Should you?
What the large posting-frequency studies found, what it means when a rival out-posts you, and how to post more without filler.
How to justify a content plan to someone who does not make content
Lead with a measured gap and your own baseline, put one reason under every item, and a content plan survives a meeting with someone who does not make content.
Why not just ask a chatbot what to post?
What ChatGPT is genuinely good at in a content team, where it runs out when the question is what to film next, and when a research tool is worth paying for.
What is a good engagement rate? It depends on whose formula
The 2026 benchmark figures for Instagram, TikTok and YouTube, why they disagree, and which comparison to use with your manager and which for your plan.
Reading the accounts around you without copying them
Social media competitor analysis in one pass: what to measure on another account, what to ignore, and how to turn a gap into something you can film this week.
How to evaluate a content research tool in one week
Content tools look alike in a demo. Three questions that separate research from a list of ideas, and how to answer them on your own account in a week.
How to decide what to film next
Brainstorms, trend lists, AI assistants, your own best posts: where teams get next week’s video ideas, what each source is good for, and how to combine them.
Is your content working? How to actually find out
Five ways teams answer "is our content working?", what each one can and can’t tell you, and a weekly review template that turns results into next week’s plan.
Glossary
Audience overlap
The share of people in more than one targeting group at once, such as two competitors' followings. High overlap means one audience; low can mean a new segment.
Baseline
An account's own recent median, used as the line every new post is measured against.
Brief
A short document, written before a shoot, stating what a video is for, who it is for, and what it needs to say.
Cadence
How often an account posts, held steady enough that an audience and a platform can learn what to expect.
Call to action
The specific instruction a piece of content gives a viewer at the moment it wants them to act.
Caption
The text posted alongside a video or photo, setting up what is about to play or carrying the call to action.
Closed loop
A content process where every result and every piece of feedback feeds the next plan, rather than disappearing after it posts.
Competitor set
A competitor set is the group of accounts worth measuring against, chosen by what your audience actually compares you to, not by who sells something similar.
Completion rate
The share of viewers who watched a video to the end, or to a set point, out of everyone who started it.
Content pillar
One of the handful of themes an account consistently returns to across its posts.
Coverage gap
Rankfield's own term for a format or angle an account's audience already responds to elsewhere that the account has never posted.
Direction
Rankfield's own term for one of a handful of plain-language options proposed before the deeper research pass runs.
Engagement rate
How actively an audience interacts with a post — likes, comments, shares, saves — measured against how many people saw it.
Evidence
Rankfield's own term for the specific numbers, posts or observations attached to a recommendation.
First pass
Rankfield's own term for a plan built from the brief and an account's own posts, before any deeper research.
Format
The shape a piece of content takes — a talking-head clip, a process video, a list to camera — independent of its topic.
Hold rate
How well a video keeps the viewers its hook already won, measured well into the video rather than at the opening.
Hook
The opening seconds of a short-form video, built to make someone stop scrolling and keep watching.
Hook rate
The share of viewers who kept watching past the first few seconds of a video, out of everyone who saw it.
Impressions
The number of times a post was displayed on screen, counting every repeat view by the same account.
Median
The middle value in a set of numbers once they are sorted from smallest to largest.
Reach
The number of unique accounts that saw a post at least once.
Retention curve
A retention curve shows what share of viewers are still watching at each moment of a video. A sharp drop points at one cut to fix; an even slope, at the pacing.
Saturation
The point where a format, hook or topic has reached most of the audience that was ever going to respond to it.
Save rate
The share of viewers who bookmarked a post to return to later, out of everyone who saw it.
Scenario
Rankfield's own term for one ranked, ready-to-film idea, complete with a hook, a shot list and its evidence.
Shot list
A shot list lays out every shot a video needs, in order, before filming starts: specific enough that someone else can hold the camera and get it without asking.
Watch time
Watch time is the total time viewers spend watching a video or a channel, added up across every view. It is a volume, so length alone does not raise it.