The short version: on a small budget, Website Card ads and native video ads targeted at competitor followers beat everything else on X (formerly Twitter), while Promoted Trends and Follower campaigns are the two formats that quietly burn through cash with nothing to show for it. Spend your first test money on follower lookalike targeting before you touch anything else, and expect to need at least ten days of data before you can trust any of it.
Why most small business owners get X ads wrong from day one
I ran X ads for a client, a small B2B software company selling scheduling tools to trades businesses, back when their marketing budget was £400 a month total across every channel. Their instinct, like almost everyone’s, was to go straight for a Follower campaign. It felt logical: more followers, more reach, more sales eventually. We spent £120 over ten days and picked up 340 new followers at roughly 35p each. Six months later I checked the account. Engagement on their organic tweets hadn’t moved. Not one of those followers had ever clicked through to the website, based on the UTM data we had tracking every link.
That’s the uncomfortable bit nobody selling you a “how to advertise on Twitter” guide wants to say out loud: buying followers on X, even through the official ad product, does almost nothing for a small business. It inflates a vanity number and that’s it. The follower who was paid for doesn’t come back to buy your product because they were never looking for it. They just happened to be in the audience you targeted when the algorithm needed to hit its delivery quota.
The formats worth your money
If you’re working with £5 to £30 a day, which is where most small businesses sit on X, here’s where I’ve seen results.
Website Card ads (or “Website Traffic” objective)
This is a single image or short video with a headline and a call-to-action button that takes people straight to a page on your site. For small budgets it’s the workhorse format because you’re not paying for engagement or follows, you’re paying per click through to your website, and the cost per click on X regularly sits between £0.30 and £1.10 in most Western markets, which is competitive against Meta and often cheaper than Google Search for the same intent level.
With the scheduling software client we ran a Website Card campaign at £15 a day for two weeks, targeting people who followed three named competitor accounts. Cost per click came in at £0.62 average, and out of roughly 340 clicks, 11 people booked a demo. Compare that to the Follower campaign spend of £120 for zero demos. The maths isn’t close.
Native video ads under 15 seconds
X’s own data has said for years that video gets more engagement than static image tweets, and in my testing that holds on the ad side too. A 12 to 15 second vertical or square video, filmed on a phone, talking directly to the camera about the problem you solve, will outperform a stock photo carousel almost every time on a small budget. You don’t need production value. You need a hook in the first two seconds and captions burned in, because most people scroll with sound off.
Promoted single tweets that already have organic traction
This is the cheapest win most people skip. Instead of building a brand new ad creative from scratch, look at your last 20 organic tweets and find the one or two that got the best engagement rate without any spend behind them. Put £5 to £10 a day behind that existing tweet rather than a fresh ad unit. You’re paying to amplify something the algorithm has already told you people like, rather than gambling on a cold creative. This is a technique I use across every social platform, not just X, and it’s covered in more depth in my guide to what moves the needle on social media if you want the wider version of the strategy.
The targeting option that punches above its weight
The single most underused targeting feature on X, especially for small budgets, is follower lookalike targeting: you tell the ad platform to show your ad to the followers of specific named accounts. Most other platforms make you build broad interest categories that are guesswork at best. X lets you say “show this to people who follow @CompanyA, @CompanyB, and @IndustryPublication” and that’s it. That level of precision on a small spend is rare.
Steps that worked for us:
- List 8 to 12 accounts your ideal customer already follows, mixing direct competitors with industry publications and well-known individuals in the niche
- Start the campaign at the platform minimum, which is usually £5 to £10 a day depending on your account
- Run it for a minimum of 7 days before touching anything, because X’s algorithm needs a learning period and pausing or editing resets that clock
- Pull the cost per click and cost per result numbers on day 7 and 10, not day 2, because early days are always more expensive while the system is still learning who converts
- Cut anything with a cost per click more than double your best performer, then reallocate that budget to the winner
On day 2 of that scheduling software test, cost per click was £1.40. By day 9 it had dropped to £0.58 on the same creative and same targeting, purely because the algorithm had learned who was clicking. If you judge a campaign on day 2 or 3, which is what most nervous small business owners do, you’ll kill winning campaigns before they’ve had a chance to work.
What to skip entirely on a small budget
Promoted Trends
These sit at the top of the trending list and cost a flat daily fee that runs into five figures in most markets. This is a big brand awareness play for companies with television budgets. There is no version of this that makes sense below a five-figure monthly ad spend, so don’t let anyone quote you a “package” that includes it.
Follower campaigns
Covered above, but worth repeating because it’s the format the ad rep on the phone will push hardest, since it’s an easy number to sell you on. More followers sounds like progress. It rarely is.
Conversation ads
These invite people to reply with pre-set hashtags, which is a nice idea for a big brand campaign with a community team ready to respond to hundreds of replies, but overkill and expensive for a business running £15 a day.
Budget splits that work
If someone handed me £300 a month to spend on X ads for a small business today, here’s the split I’d run:
- 60 percent (£180) on Website Card ads using follower lookalike targeting, tested against 2 to 3 different competitor account lists
- 25 percent (£75) on boosting your best-performing organic tweet each week, rotating to whichever post got the strongest engagement rate that week
- 15 percent (£45) held back as a testing pot for one new video creative a month
Nothing goes to Follower ads, nothing goes to Promoted Trends, and nothing goes to broad “everyone in the UK aged 25 to 55” interest targeting, which on X specifically tends to be more expensive and less accurate than it is on Meta because the interest categories aren’t as tightly maintained.
The part that annoys most business owners
Here’s the truth that most guides on this topic avoid saying plainly: X ads on a small budget rarely work as a cold prospecting tool for building brand new demand. Where they work is retargeting people who already know you, warming up an audience that follows a competitor, or amplifying content that’s already proven itself organically. If your whole plan is “spend £10 a day and get strangers who’ve never heard of us to buy,” on X specifically that’s a harder ask than on Meta or TikTok because the platform’s core use case is still text-first conversation and news, not product discovery. People are on there to see what’s happening, not to shop. Build your ad strategy around that reality rather than fighting it.
That’s also why pairing X ads with a wider content plan matters more here than on almost any other platform. If you’re short on hands to create that content consistently, it’s worth reading through what works right now in freelance writing and content support for small teams, because a good writer producing the tweets you then boost is often a better spend than a bigger ad budget on weak creative.
Tools that make the small-budget version manageable
Running this without a dedicated ad manager on staff means you need decent scheduling and reporting tools, otherwise you’ll lose track of which creative is which by week three. I keep a running rundown of what’s worth paying for in my guide to the best social media management tools, and for creating the still images that go with Website Card ads without hiring a designer, I’ve switched almost entirely to AI image generation, which I wrote about in my ChatGPT Images review after cancelling my Canva subscription off the back of it.
If your budget covers ads on X but you’re also juggling five other channels with no time to manage any of them well, that’s usually the sign you need a wider strategy rather than another ad platform to learn from scratch. My guide to marketing strategies for cash-strapped small businesses covers how to sequence spend across channels so X ads support the rest of your marketing instead of competing with it for the same £300.
What I’d tell you before you spend your first pound
Set the budget you can afford to lose, because your first ten days are a learning cost, not a results cost. Pick one format, Website Card ads with lookalike targeting, and one clear conversion goal, whether that’s a demo booking, an email signup, or a purchase. Resist judging the campaign before day 7. And ignore any account rep who tells you Promoted Trends or a Follower campaign is the fastest way to grow, because for a small budget it’s the fastest way to spend money and learn nothing.
Frequently asked questions
What’s the minimum daily budget for X (Twitter) ads?
Most campaign types allow a minimum of around £5 to £10 a day, but realistically you need at least £10 a day for 7 to 10 days, roughly £70 to £100 total, before you have enough data to judge whether a format or targeting choice is working.
Are X ads cheaper than Facebook or Instagram ads?
Cost per click on X often lands between £0.30 and £1.10, which is competitive with or sometimes cheaper than Meta, but conversion rates tend to be lower because people use X for conversation and news rather than shopping, so treat it as a warming and retargeting channel rather than a direct-response one.
Should a small business ever run a Follower campaign on X?
Almost never. Follower campaigns buy accounts that follow you but rarely engage or buy, and the money is better spent on Website Card ads or boosting organic tweets that already have proven engagement.
How long should I run an X ad before deciding if it works?
At least 7 days, ideally 10. Cost per click and cost per result typically drop significantly between day 2 and day 9 as the platform’s delivery system learns who responds, so judging early kills campaigns that would have improved.
Related reading: What Makes A Twitter Chat Useful For Growing Your Audience and How to Write Sales Objection Email Templates That Convert.
Useful references
One underused tactic worth adding: how to create Twitter Moments, which groups related tweets into a single curated highlight your audience can browse in one place.
