Platforms
Written for someone who has never opened Ads Manager. What a Reels ad is, how to build one without guessing, and what the numbers mean once it starts running.
A Reels ad is a vertical video that appears inside the Reels feed, between the videos people are already scrolling. It plays full screen, it plays automatically, and it looks like the content around it apart from a sponsored label and a button near the bottom.
Because Instagram is part of Meta, the same ad system also fills the main Instagram feed, Stories, Explore and the equivalent surfaces on Facebook. That matters for one practical reason: Reels is a placement — one of several destinations for the same campaign — not a separate product with its own login.
Ads Manager nests everything in this order. Most beginner confusion is something set at the wrong layer.
Holds your objective — the single thing you are asking Meta to optimize for. Traffic, leads, sales, installs. Set once, and it governs everything underneath.
Holds budget, schedule, audience and placements — how much you spend, over what period, to whom, and whether Reels is one of the surfaces you appear on.
Holds the creative — video, text, button and link. One ad set can hold several ads, which is how you test more than one video against the same audience.
Both spend your money in the same auction. They do not give you the same control.
| Boost from the app | Build in Ads Manager | |
|---|---|---|
| Setup effort | Two minutes | Half an hour, the first time |
| Objective choice | A short simplified list | The full set, including sales and leads |
| Placement control | Largely decided for you | Yours to set |
| Testing two videos | Not really possible | Separate ads in one ad set |
| Reporting depth | Headline numbers only | Breakdowns by placement, age, device |
| Good for | A post already doing well organically | Anything you intend to measure |
Meta renames tools and moves controls frequently, and the boost flow changes often. Meta's own Business Help Center is the authority on what any given screen does today.
Boosting is not a scam. It is a shortcut with the decisions taken away from you, and if a post is already doing well organically it is a fine way to spend fifty dollars. The problem starts when you want to know whether the money worked — boosting gives you thin reporting and little control over what was optimized for, so the result is hard to attribute and impossible to repeat deliberately. If you are running UGC video to find out which message sells, you need two videos against one audience. That is an Ads Manager job.
Ads Manager is free and lives at business.facebook.com. It needs a Facebook Page connected to your Instagram account, plus a payment method. Do that before you start — a missing Page link is the usual reason a first-time advertiser cannot select Instagram placements at all.
Described by what each step decides rather than by button labels, because the labels move.
The objective tells Meta which action to hunt for. Pick what you genuinely want — sales if you want sales. Choose engagement because the numbers look bigger and the system will faithfully deliver cheap engagement instead of customers.
A daily budget is easier to reason about than a lifetime one. Set what you are willing to lose while learning, not what you hope to spend when it works.
Country, age range and language are enough to start. A narrow interest stack on a small budget starves the system of data and makes every result harder to read.
Meta defaults to placing your ad automatically across all its surfaces, which generally delivers more cheaply. There is a manual option if you want Reels only — but that trades reach for control.
At the ad level you say which Page and Instagram account the ad appears from. That handle is the byline viewers see, and if the account is not selectable, the Page connection is what to fix.
Add the vertical file, the text, the button and the destination link. Meta will offer to auto-crop the same video for other placements — check every one of those previews.
The preview panel shows the ad as it will appear in each placement. This is where you catch text sitting behind the caption or the button.
New ad sets go through a learning period while Meta works out who responds, and editing budget, audience or creative restarts it. Give it a few days before judging anything.
Meta's interface changes constantly — panels get renamed, steps merge, options move between the ad set and ad levels. The sequence of decisions above has been stable for years; the screens have not. Meta's own documentation is the authority on where a control currently lives.
Instagram draws its own interface on top of your video. Anything under it is invisible to the viewer even though it looks fine in your editing timeline.
There is a second way to run UGC on Instagram, worth knowing even if you do not use it immediately. Instead of uploading a creator's video as your own ad, you can run their actual post as the ad — appearing from their handle, with your brand credited alongside it.
Meta has called this branded content ads and, more recently, partnership ads. The mechanic has stayed the same: the creator marks the post as a paid partnership with your business and grants your ad account permission to promote it. You never get access to their account, and they never see your ad account.
What that buys is a byline belonging to a real person. What it costs is a dependency — a specific creator has to agree, post, and keep the post up. It is the right tool when the creator's credibility is the point; for testing which message works, your own uploaded creative is far cheaper per test.
That is also the honest limit of AI-generated video. Yuugc delivers the video, the script and full commercial rights — not a creator's account, their following, or a genuine verified customer testimonial. An AI video runs as your own creative from your own handle. It cannot appear as a partnership ad from a creator who never posted it.
Ads Manager shows dozens of columns. On a first campaign, four of them matter.
Your spend divided by the thing your objective optimized for. The only number that answers whether the ad worked; everything else explains why.
Meta reports video plays at several length thresholds. Comparing the earliest to impressions tells you whether your first seconds are doing their job. A weak opener cannot be rescued further down.
Whether the video persuaded anyone to act. Watching without clicking usually means the video entertains but the offer is not landing.
How many times the average person has seen it. When it climbs and cost per result climbs with it, the audience is tired of the creative — bring in a new video.
Meta's delivery system needs conversions to learn from. The commonly cited threshold is around 50 optimized events per ad set per week to leave the learning phase and stabilize. Below that, results swing hard day to day and comparing two ads honestly gets difficult.
On a small budget that has a blunt implication: do not split it across five ad sets. One ad set, one audience, two or three videos inside it. If your product is expensive enough that 50 purchases a week is unreachable, optimize for an earlier event — an add to cart, a lead — and accept you are measuring a proxy.
Expect the first campaign to teach you more about your creative than about your audience. That is the normal outcome, and the useful one. If you run the same footage on TikTok too, it needs different treatment there — UGC for Meta ads covers what changes and why.
In practice, yes. Ads Manager ties your Instagram account to a Facebook Page, and the account often will not be selectable at the ad level until that link exists. Sort it before you build a campaign rather than halfway through.
Enough that your ad set collects a meaningful number of optimized events per week, and no more than you can afford to lose while learning. There is no universal figure — it depends on what an event costs in your category. Judge it on cost per result, not total spend.
Automatic usually delivers more cheaply, because Meta can place the ad wherever it is cheapest to win the auction. Choose Reels only when you have a specific reason — a creative built purely for full-screen vertical that cropping would damage.
Two or three inside one ad set — enough to see a difference, few enough that each gets real delivery. Ten videos on a small budget means none of them is properly tested.
Film them yourself, buy them from creators, or generate them. Yuugc writes the script from your brief and delivers finished vertical video in 48 hours with full commercial rights, at around $50 a video. See how it works for what a brief needs to contain.
Tell Yuugc what you sell and who it is for. Script writing, an AI creator matched to your audience, caption variants and one revision round are included with every video.
Get a quote