Why Your Facebook Video Ads Aren't Working

Four creative problems account for most underperforming Facebook video ads, and none of them are targeting. The first three seconds don't stop anyone. The video needs sound to make sense. The offer shows up at the end. The file is the wrong shape for the placement it's running in. Check those four before you touch the audience.

One disclosure first, because it matters. Media buying and creative are two different jobs. Runna Media doesn't manage ad accounts and we don't run PPC, so we're not going to tell you your cost per lead is a campaign structure problem. We make the video. This covers the creative half, which is the half most people skip on the way to blaming the audience.

If almost nobody makes it past your first three seconds, no amount of audience tinkering will save the campaign. Meta will spend your budget showing exactly the right people in Orland Park a video they don't watch.

Is this a targeting problem or a creative problem?

There's a fast way to tell, and it doesn't take a media buyer. Open the ad, find where people drop off, match the symptom to the cause.

One row on that list belongs to your media buyer. That isn't a statistic, it's just where the levers sit: the audience gets set once and then runs, while the video has to earn its keep on every single impression.

Why do people scroll past in the first three seconds?

Because the first three seconds are usually the logo. Or a drone shot of the building. Or the owner saying hi, I'm Mike, and I've been serving the southwest suburbs since 1998.

Facebook autoplays your video muted, mid-scroll, in a feed that also holds a friend's baby photo and an argument about the Bears. Nothing there owes you attention. The opening frame has one job, which is making a specific person feel the video is about them.

Openers that earn the next two seconds: the problem stated in five words, a face already mid-sentence, the finished result shown before the process. Openers that don't: a logo animation, a slow fade from black, a drone pass over a strip mall on LaGrange Road.

If you cut the first three seconds off your ad and it plays better, you just found your problem.

The first frame is not an introduction. It is the entire ad. Everything after it exists only for the people that frame already convinced.

Does the ad still make sense with the sound off?

Assume sound off. The feed is built for muted viewing, so a video that needs a voiceover to land its point is a video most people never actually understand. They see a mouth moving and keep going.

Sound-off design means the message is legible without audio. Captions burned into the file, not platform auto-captions you don't control. Text on screen that says the thing. Visual proof instead of narrated proof: the before and after, the hands doing the work, the truck pulling into a New Lenox driveway.

A spot that works silent and gets better with sound is the standard we build to. It's the same discipline a connected TV spot needs, for the opposite reason.

Where should the offer go in the video?

Not at the end. That's where it usually is, and that's why nobody sees it.

A thirty second ad where the offer lands at second twenty-six only converts the people who watched all the way through. Say what you want them to do early, then say it again at the close. Repeating it inside a fifteen second cut feels heavy-handed in the edit bay and reads as clear in the feed.

Be specific about it, too. Learn more is not an offer. Free estimate, we're based in Mokena, we can be out this week is an offer. If you sell a trade or a home service, name the service, name the area, and name the next step early enough that a partial viewer still catches all three. There's more on how that plays locally on our Orland Park marketing page.

Why does my footage look like stock?

Because it either is stock, or it was shot like stock.

We're not knocking stock footage. There are shots you can't practically get, and licensed b-roll fills real gaps in an edit. But an ad built entirely from generic clips of people smiling in a bright open-plan office tells a Chicago buyer nothing, and hints that you might not be local.

The fix is specificity, and it's cheaper than people expect. Shoot in your actual shop. Put your actual crew on camera. Show the job site in Frankfort, the counter in Tinley Park, the bay door open in Lockport. A slightly rough shot of a real place beats a perfectly lit shot of nowhere. A real place is also the only thing a competitor can't buy from the same stock library you did. Most of what's on our portfolio is that, and none of it is accidental.

What aspect ratio should a Facebook video ad be?

Uploading one master file across every placement is a common technical mistake, and one of the easiest to fix. Meta crops or letterboxes your video to fit the slot, and it doesn't care that the crop cut your phone number off the bottom of the frame.

• Feed, Facebook and Instagram: 4:5 vertical. Fills more of the phone than square without getting trimmed.
• Reels and Stories: 9:16 full screen. Keep captions, logos, and pricing out of the top and bottom strips where the interface sits.
• In-stream and video feeds: 16:9. The one placement where assuming sound-on is fair.
• Explore and marketplace slots: 1:1 square still holds up when you don't want a fourth cut.

Same shoot, different exports. If the frames were composed on set with those crops in mind, the extra versions cost you nothing. If they weren't, you're rebuilding the ad in post and losing the top of every shot.

How often do you need new ad creative?

Sooner than you want to. Even in a metro this size, a tight geo-target hits the same households over and over, and once frequency climbs the ad stops working without a single thing about it changing. In the dashboard that looks exactly like a targeting problem. Costs up, click-through down, audience untouched.

Swapping the headline doesn't reset it. Same video with new text is still the same video. What resets performance is a genuinely different concept: different opening, different person, different angle on the same offer.

The same shoot day should generate at least three to five usable assets when it's planned correctly, which gives you a rotation instead of a single hero spot carrying the whole account. It's the same argument we make in our short-form video strategy post.

What does it cost to fix the creative?

Most foundational video production packages run $8,000 to $30,000 depending on scope, and a paid social package sits toward the lower end because the deliverables are short and the shoot is usually one day. Ongoing content retainers run $4,000 to $12,000 a month if you want a rotation instead of a one-time build.

None of that is media spend. Media spend goes to Meta and has nothing to do with us. If you already have a buyer you trust, keep them and bring us the creative. The full breakdown of why production ranges are this wide is in the video production cost guide.

Why is this harder than it used to be?

Facebook and Instagram CPMs in the Chicago metro sit higher than they did in 2022. You're paying more for the same impression than you were when this channel was cheap, which means the creative has to carry weight it never used to carry. A mediocre ad survived on volume once. It doesn't now.

So the question isn't why the ads stopped working. It's whether the video you're running right now would stop you.

Frequently Asked Questions

Why are my Facebook video ads not working?

Usually the creative, not the targeting. The common causes are an opening three seconds that doesn't stop the scroll, a video that only makes sense with sound on, an offer that doesn't appear until the end, footage that looks generic, the wrong aspect ratio for the placement, and creative fatigue from running one ad too long. Check those before you rebuild the audience.

How long should a Facebook video ad be?

For cold audiences on Feed and Reels, fifteen seconds is a working default and thirty is the ceiling. Shorter cuts get more people to the end, and a finished view is worth more to you than a long one nobody completes. Longer cuts can work for warm retargeting where the viewer already knows you. Length matters less than whether the offer appears early enough for a partial viewer to catch it.

Do Facebook video ads need captions?

Yes. Facebook and Instagram autoplay video muted, so an ad that depends on a voiceover loses most of its audience. Burn the captions into the file rather than relying on platform auto-captions, which you don't control and which can render differently by placement. Design the ad to be understood silently, then treat the audio mix as an upgrade for people who turn sound on.

What is creative fatigue and how do I know if I have it?

Creative fatigue is what happens when the same households have seen the same ad too many times. In the dashboard it looks like rising costs and falling click-through on an ad that used to perform, with no change to the audience or the budget. New headlines don't fix it. A genuinely different concept does: different opening, different person, different angle on the same offer.

What aspect ratio should I use for Facebook and Instagram video ads?

Export a cut per placement. Feed video performs best at 4:5, Reels and Stories need full-screen 9:16, in-stream placements are 16:9, and 1:1 square still covers Explore and marketplace slots. Keep captions, logos, and pricing out of the top and bottom strips of vertical cuts, because the interface sits there and will cover them.

Does Runna Media manage Facebook ad campaigns?

No. Runna Media produces the creative. We don't manage ad accounts, run PPC, or buy media, and we won't pretend those are the same job as ours. We build the video, the cutdowns, and the placement-specific exports, then hand them to whoever runs your campaigns. If you don't have a media buyer yet, we'll tell you that plainly instead of selling you something we don't do.