The short version: for most small businesses in 2026, short video ads and single image ads in the news feed outperform everything else on cost per result, carousel ads win for anything with multiple products or steps, and lead ads beat a landing page for local services. The format matters far less than most agencies tell you it does. Your offer, your creative, and how fast you stop a losing ad matter more.
Why “best format” is the wrong first question
Every year someone writes a post ranking Facebook ad formats like they’re football teams. I’ve run ads for retail clients, coaches, a wedding stationery business, and a Kent based furniture restorer, and the format was never the thing that made the difference. The offer was. A weak offer in a beautiful video still loses money. A brilliant offer in a scrappy single image ad still converts.
That said, format does affect cost per click, cost per result, and how forgiving Meta’s algorithm is with your budget, so it’s worth knowing where to start.
Short video ads: the current default for a reason
Meta has been pushing everyone towards video for two years now, and if you’re advertising to a cold audience, a 15 to 30 second video in Reels or feed placement will usually get you a lower cost per click than a static image, simply because Meta serves it more places by default.
I’ve seen cost per click on video ads for small service businesses sit between 30p and 70p in the UK, versus 60p to 1.20 pounds for a similar single image ad to the same audience. That’s not a universal rule, but it’s a pattern I’ve watched repeat across a dozen accounts.
- Film vertical, not landscape. Square or 4:5 works too, but vertical fills the phone screen.
- Put your point in the first three seconds. No logo intro, no slow zoom.
- Add captions. Around 85 per cent of Facebook video is watched with sound off.
You don’t need a videographer. A client of mine, a small pottery studio in Norfolk, films every ad on an iPhone propped against a mug. Her cost per lead sits around 4.50 pounds, which beats what the same money got her from a professionally shot ad two years ago.
Single image ads: still underrated, still cheap to test
Every marketer wants to talk about video because it feels current, but single image ads remain the fastest and cheapest way to test five different angles on an offer before you spend a penny on production. You can have ten variations live by lunchtime.
Use single image ads for:
- Testing headlines and hooks before you commit to video
- Retargeting people who already visited your site (a product shot with a clear price does the job)
- Local awareness campaigns where a simple, honest photo of the shop or product outperforms anything polished
The uncomfortable bit nobody likes admitting: a lot of “high performing” video ads only beat images because Meta’s delivery system favours video right now, not because video is inherently more persuasive. If Meta changes its weighting next year, and it has done this before, the ranking flips. Chase results, not the format that’s fashionable this quarter.
Carousel ads: the format small businesses underuse
Carousels get a fraction of the attention video does, but for anyone selling more than one product, or explaining a process in steps, they earn their place. Each card is its own image and headline, so you can show a before and after, a range of colours, or a three step service in one ad.
The furniture restorer I mentioned earlier ran a carousel showing a table before restoration, mid-strip, and finished, with the price for that service on the final card. That single carousel ran for four months, cost 380 pounds total, and generated 11 booked jobs averaging 340 pounds each. Her single image ads on the same budget never got past 4 or 5 enquiries a month.
Carousels also work well alongside organic content. If your Facebook page has been struggling with reach lately, worth reading up on why a Facebook page stops showing up for followers, because paid carousel ads can partly compensate for that drop while you fix the organic side.
Lead ads: the quiet workhorse for local and service businesses
If you’re a plumber, a therapist, a driving instructor, or anyone whose goal is “get someone’s phone number,” lead ads beat sending traffic to a website nine times out of ten. The form opens inside Facebook, pre-fills the person’s name and email, and takes two taps to submit. No page load, no faffing about on a slow mobile connection.
Step by step, a lead ad campaign I’d set up for a small business today looks like this:
- Objective: Leads, optimising for on-Facebook forms
- Budget: start at 10 pounds a day minimum, 15 to 20 pounds if you want data faster
- Audience: 1 to 3 mile radius for local, or a lookalike of your customer list for online
- Form: three questions maximum, name, phone, one qualifying question
- Creative: one video, one image, running at the same time so Meta can find the winner
- Review point: day 4, kill anything with a cost per lead more than double your best performer
The catch with lead ads is follow up speed. Leads from these forms go cold within about an hour. If you’re not calling or messaging back same-day, you’re paying for names on a spreadsheet, not customers.
Collection and product ads: only if you’re selling online
Collection ads, the ones with a hero video or image and a grid of products underneath, are built for ecommerce. If you’re a service business, skip them entirely, they’ll waste your budget on a format designed for browsing a catalogue, not booking a consultation. This is one of those places where a lot of guides give you a full list of every format available without telling you half of them aren’t for you.
If you do sell physical products, connect your catalogue and let Meta’s dynamic ads retarget people who viewed specific items. That’s a different setup to the ads discussed above, and it’s worth its own separate test budget rather than mixing it into your main campaign.
Where your ads show up matters as much as the format
One thing that trips up small business owners constantly: you set up a “Facebook ad” and it doesn’t just run on Facebook. By default it also serves across Instagram feed, Instagram Stories, Reels, Messenger, and the Audience Network, splitting your budget in ways you didn’t choose. I’ve written a full breakdown of what happens to your budget when Facebook ads run on Instagram too, and it’s worth reading before you launch anything, because a video that looks great in Facebook feed can look cropped and wrong in an Instagram Story.
My advice for a first campaign: leave automatic placements on, but check performance by placement after five days. If Reels is eating 40 per cent of budget and delivering nothing, you can exclude it and reallocate.
A simple test structure that finds your best format without guessing
Here’s the exact structure I use with new small business clients when we don’t yet know what works for them:
- Week 1: Run three ad sets, same audience and budget, one video, one single image, one carousel. Total spend around 150 to 200 pounds for the week.
- Day 7: Compare cost per result, not just cost per click. A cheap click that never converts is worse than an expensive one that does.
- Week 2: Kill the two weakest formats. Take the winner and build three creative variations within that format.
- Week 3 onwards: Refresh creative every 10 to 14 days. Ad fatigue sets in fast on small audiences, sometimes within a week if your radius is under five miles.
This is where most small businesses go wrong, not the format, but the refresh. I’ve watched a perfectly good carousel ad quietly decline from a 2.1 per cent click-through rate to 0.6 per cent over three weeks purely because the same 8,000 local people had seen it forty times.
What I’d tell a small business starting from scratch today
Start with one video ad and one single image ad, same offer, same audience, 10 pounds a day between them. Give it five days before you touch anything. Add a lead ad if your goal is enquiries rather than sales. Add carousel once you have more than one product, service tier, or step to show. Skip collection ads unless you run an online shop with a real catalogue.
And keep your organic side alive too. Ads work better when they’re reinforcing a page that already looks active, so don’t neglect things like using effective Facebook hashtags on your organic posts, or exploring whether it’s worth trying to monetise your Facebook page alongside your ad spend, since the two feed each other more than people expect.
Frequently asked questions
What is the cheapest Facebook ad format for a small business to start with?
Single image ads are the cheapest and fastest to test because they need no production time, you can launch ten variations in an afternoon, and cost per click typically sits lower than video for a cold audience with no prior brand awareness.
Do video ads always beat image ads on Facebook?
No. Video often gets a lower cost per click right now because Meta’s algorithm favours it in delivery, not because it’s inherently more persuasive, and single image ads regularly win on cost per result for retargeting and local awareness campaigns.
How much should a small business spend to test Facebook ad formats?
Around 150 to 200 pounds over one week, split across three ad sets with the same audience and offer, is enough to see a clear cost per result difference between formats before you commit a full month’s budget to one.
Are lead ads better than sending traffic to a website?
For local and service businesses, yes, because the form opens instantly inside Facebook without a page load, which lowers drop-off, but leads go cold within about an hour so they only pay off if you follow up the same day.
