X (Twitter) Ads Guide: Formats, Targeting & Results

It’s 8:47 in the morning, you’ve got a coffee going cold beside you, and someone upstairs just asked whether you can “throw a little budget at that post to get it in front of more people.” Sounds simple. Then you open Ads Manager, and suddenly there are objectives and bid strategies and audience layers and three different things all called some flavor of “promoted,” and you close the tab and tell yourself you’ll figure it out later.

Let’s figure it out now instead. This X (Twitter) ads guide is the walkthrough I wish someone had handed me the first time I ran a campaign on the platform. We’ll cover every ad format on X (formerly Twitter), how Ads Manager is actually structured, how targeting really works, how to set budgets without lighting money on fire, what makes creative land, and, most importantly, how to measure results honestly so you know what to do more of. No invented statistics, no “this one trick,” just a real system you can start using today.

First, why advertise on X at all?

X is a strange and wonderful place for paid media. It’s built around the moment, live reactions, breaking news, sports, launches, jokes that age in hours. That real-time energy means ads on X can ride cultural momentum in a way that’s harder on more evergreen, polished platforms. It’s a place where a brand with a point of view and a fast hand can genuinely punch above its budget.

The flip side: X moves fast and scrolls faster, so your ad has a heartbeat’s worth of time to earn attention. That’s not a reason to avoid it. It’s a reason to be deliberate, which is exactly what the rest of this guide is about.

One reframe before we dive in: an ad on X is really just a post with a budget behind it. Everything you already know about writing a good post, the hook, the clarity, the reason to care, still applies. Paid distribution amplifies whatever you feed it. Amplify something mediocre and you’ve bought reach for a shrug. Amplify something genuinely good and the money multiplies momentum you’d already have earned organically.

The ad formats on X, in plain English

X’s ad products can feel like alphabet soup until you realize they fall into two buckets: the core ad units (what your ad looks like in-feed) and the premium placements (where an ad shows up, often reserved for bigger spends and bigger moments). Here’s the lay of the land.

Promoted Ads (your everyday workhorse)

Formerly called Promoted Tweets, these are regular posts, boosted to appear in the timelines of people who don’t follow you, marked with a small “Ad” or “Promoted” label. They look and behave like native posts: people can like, repost, reply, and quote them. This is the format you’ll use most, and it comes in several creative flavors:

• Single image or single video — the classic. One clear visual, one message.
• Carousel — multiple swipeable images or videos, great for showing several products, steps, or angles in one unit.
• Text ads — words only, leaning on copy and the conversation itself.
• Moment ads — a curated collection of posts stitched into a longer narrative, useful for storytelling that needs more room.

Follower Ads (formerly Promoted Accounts)

Instead of promoting a single post, you promote your whole account, encouraging people to follow you. These appear in the timeline, in Who to Follow suggestions, and in search. Follower Ads are a slower, foundational play: you’re building an owned audience you can reach for free later, rather than chasing an immediate click or sale.

Amplify (pre-roll video)

Amplify lets your brand run short video ads alongside content from publishers and creators, think of it as running before or around premium video that people are already leaning in to watch. There are two flavors: Amplify Pre-roll, where you choose broad content categories to align with, and Amplify Sponsorships, where you pair one-to-one with a specific publisher for a set period. Amplify is where you go when you want your video wrapped in the credibility of content people already trust.

Takeover placements (the big, loud ones)

These are premium, high-visibility placements usually booked for launches and tentpole moments:

• Timeline Takeover — owns the first ad slot at the top of the timeline for a full day, so you’re the first promoted thing a huge slice of the audience sees when they open the app.
• Trend Takeover — places your promoted trend near the top of the What’s Happening / Explore list, right where people go to find out what everyone’s talking about. Trend Takeover+ adds a visual creative at the top of the trends tab.

Takeovers are reach-and-awareness cannons. They’re not where you start, but it’s worth knowing they exist so that when someone asks “how do we make a splash for the launch,” you know what to reach for.

Objectives: tell X what “success” means

Before you pick a format, X asks you to pick an objective, and this is the single most important choice in the whole setup. The objective tells X’s delivery system what outcome to optimize toward, and the algorithm will relentlessly chase whatever you name. Choose the objective that matches the actual business result you want, not the one that sounds most impressive.

The objectives group neatly into the classic marketing funnel:

• Awareness — optimize for reach, getting your message in front of as many relevant people as possible. Use this when the goal is simply that more people know a thing exists (a launch, a rebrand, an event).
• Consideration — this is the busy middle, and it splits into several sub-objectives: video views (people watching your video), engagements (likes, reposts, replies on your post), followers (growing your account), website traffic / clicks (sending people to a page), and app installs / re-engagement (getting people into your app).
• Conversion — optimize for actions on your site, purchases, sign-ups, leads, using the X conversion tracking pixel to teach the system who converts. This is the bottom of the funnel where you’re chasing measurable business outcomes.

Here’s the trap to avoid: don’t pick a top-of-funnel objective and then judge it by bottom-of-funnel results. If you optimize for awareness and then complain there were no sales, that’s like ordering a salad and being upset it isn’t a steak. Match the objective to the metric you’ll grade it on, and set that expectation with anyone watching the numbers.

How Ads Manager is structured

Once you’ve got an ads account set up at ads.x.com (you’ll add a payment method and confirm your account details first), everything lives inside a three-level hierarchy. Understanding this structure is what turns Ads Manager from a maze into a map.

• Campaign — the top level. Here you set your objective and your overall daily and/or total budget. Think of the campaign as the mission.
• Ad group — the middle level, where the real strategy lives. Each campaign can hold multiple ad groups, and each ad group has its own audience targeting, placements, bid strategy, and schedule. Ad groups are how you test different audiences against each other cleanly.
• Ads — the bottom level: the actual posts (creative) that run. You can put several ads in one ad group and let the system favor the ones performing best.

The practical mental model: campaign = your goal and money, ad group = who you’re targeting and how you’re bidding, ad = what they actually see. When something’s working or failing, this structure tells you where to look. Weak results across every audience? Probably a creative problem, fix the ad. One audience crushing it and another flatlining? That’s an ad-group targeting insight, shift budget.

Targeting: finding the right people

X’s targeting is genuinely one of its strengths, because it can target around interest and conversation in ways that suit how people actually use the platform. You’ll generally combine a few of these layers inside an ad group. A word of caution first: every layer you add narrows your audience. Stack too many and you’ll strangle delivery down to almost nobody. Start broader than feels comfortable and tighten based on what the data shows.

The main targeting levers

• Demographics — location, language, age, gender, device. The basics, and often the ones worth setting loosely.
• Keywords — target people based on words in the posts they engage with or search for. This is X at its most X: you’re reaching people by the conversations they’re literally having right now. Powerful for tapping into intent and moments.
• Interests — broad categories X assigns based on behavior (sports, tech, cooking, gaming). Good for scale and top-of-funnel reach.
• Follower look-alikes — target people who behave like the followers of specific accounts. Instead of guessing at interests, you name accounts your ideal customer probably follows (competitors, relevant media, adjacent brands) and reach audiences resembling their followers.
• Conversation topics — a curated set of topics X maintains, letting you reach people engaged in specific ongoing conversations.
• Events — target around specific happenings (a sports final, an awards show, a holiday) where you know attention will spike.

Custom and tailored audiences

The highest-intent targeting uses your own data. Custom audiences let you upload a list (like an email list, matched securely) or retarget people who’ve visited your site (via the X pixel) or engaged with your previous posts and videos. Retargeting warm audiences, people who already watched your video or visited your pricing page, almost always outperforms cold targeting, because you’re talking to people who’ve already raised a hand. If you do nothing else sophisticated, install the pixel and retarget site visitors.

Budgets and bidding without the anxiety

This is where most people freeze, so let’s make it concrete. You control spend at two levels, and you choose how aggressively X bids on your behalf.

Setting the budget

• Daily budget — the maximum you’ll spend per day per campaign. This is your safety rail; X won’t blow past it.
• Total budget — an optional overall cap for the whole campaign’s lifetime, letting X pace spending across your flight.

How much? I won’t invent a magic number, because the right budget depends entirely on your goal, your market, and how much a result is worth to you. Instead, here’s the method: decide what a single desired outcome (a click, a lead, a follower, a sale) is genuinely worth to your business, then start with a budget large enough to buy a meaningful sample of those outcomes, enough data to learn from, not so much that a bad first week hurts. You’re buying information as much as results in the early days.

Choosing a bid strategy

X gives you a few ways to bid into the auction:

• Autobid — you hand X the reins and it tries to get you the most results for your budget. This is the right default when you’re starting out and don’t yet have your own cost benchmarks. Let it gather data first.
• Maximum bid — you set the most you’re willing to pay per action, giving you tight control at the cost of potentially limiting delivery.
• Target cost — you tell X the average cost per result you’re aiming for and it works to keep you near it.

My honest advice: begin with Autobid, let the campaign run long enough to establish what a result actually costs you, then, once you know your numbers, graduate to maximum or target bidding to control costs. You can’t set a smart cost target before you know what “normal” looks like for your account. Data first, control second.

Give it room to learn

Every ad platform, X included, needs a learning period, usually the first several days, where delivery is uneven and costs are unstable while the system figures out who responds. Resist the urge to yank things around on day one. Constantly editing budgets, audiences, and creative resets that learning and keeps you permanently in the noisy, expensive phase. Set it thoughtfully, then let it breathe before judging.

Creative that actually stops the scroll

You can nail targeting and budgets and still get nowhere if the creative doesn’t earn a half-second of attention. On a platform this fast, the creative is doing most of the heavy lifting. Here’s what consistently works.

• Lead with the hook, immediately. The first line of copy and the first frame of video have to work before anyone decides to scroll on. Front-load the payoff, don’t build to it.
• Design for sound-off, then reward sound-on. Most video is watched muted at first. Use captions and on-screen text so the message lands silently, and make the audio a bonus rather than a requirement.
• Keep video short and shapely. The tighter the better; every second you ask for is a second someone can leave. Say one thing well.
• Write like a human, not a billboard. X rewards a voice. Copy that sounds like a real person having a real thought outperforms polished corporate-speak almost every time.
• Make the ask unmistakable. Use a clear call to action and, where the format allows, a website card or button so the next step is obvious and tappable. Don’t make people hunt for what to do.
• Match the message to the moment. One of X’s superpowers is relevance to what’s happening right now. Creative that nods to a current conversation or event feels native rather than intrusive.

Above all, test more than one version. Put two or three different creatives in the same ad group and let real behavior, not your gut, pick the winner. You will be wrong about which one wins more often than you’d like, and that’s exactly why you test. If you want a deeper well of creative concepts to draw from, our roundup of short-form video ideas translates neatly to X video ads, and the principles in what actually makes content spread apply whether a post is paid or organic.

Measuring results honestly

Here’s the part that separates people who waste money on X from people who compound their results: a real measurement habit. Ads Manager gives you a wall of metrics, and the skill is knowing which ones matter for your objective and ignoring the vanity noise.

Match the metric to the objective

Come back to the objective you chose, and grade the campaign on the metric it was built to move:

• Awareness campaigns → look at impressions, reach, and cost per thousand impressions (CPM). Did you get in front of enough of the right people affordably?
• Video view campaigns → look at views, view rate, and cost per view, plus how far into the video people actually watched (completion). A cheap view that ends after one second isn’t worth much.
• Engagement campaigns → look at engagement rate and cost per engagement.
• Traffic campaigns → look at link clicks, click-through rate (CTR), and cost per click. And crucially, check what happens after the click, do those visitors do anything, or bounce?
• Conversion campaigns → look at conversions, cost per conversion, and ultimately return on ad spend. This is the number that pays the bills.

Set up conversion tracking before you spend

If you care about anything beyond awareness, install the X conversion tracking pixel (or the mobile SDK for app campaigns) before your first dollar goes out. Without it, X can’t optimize toward conversions and you can’t see what actually drove business results, you’re flying blind and telling yourself it’s fine. The pixel is the difference between “we got a lot of clicks” and “we got fourteen sign-ups at a cost we can live with.”

Read results at the right level

Remember the campaign structure? Use it when you analyze. Break performance down by ad group to see which audiences respond, and by ad to see which creative wins. That’s how you turn a fuzzy “the campaign did okay” into specific, fundable decisions: pause the audience that isn’t working, pour budget into the one that is, kill the tired creative, and make three more like the winner.

Keep a simple log

You don’t need a fancy dashboard to start, but you do need a record. Keep a running note (a spreadsheet is plenty) of each campaign’s objective, audience, creative, budget, and the one metric that mattered. Over a few campaigns, patterns emerge, which audiences convert cheapest, which hooks land, what day of the week performs. That accumulated, account-specific knowledge is worth more than any generic benchmark you’ll read online, because it’s true for your audience. If you want a fuller framework for deciding what’s worth tracking in the first place, our guide to the social media metrics that actually matter is a good companion to this one.

A starter workflow you can run this week

Enough theory. Here’s a concrete sequence to get a first campaign live and learning without overreaching.

• 1. Set up the account and pixel. Create your ads account, add billing, and install the conversion pixel on your site, even if your first campaign is awareness. You’ll want the data flowing from day one.
• 2. Pick one clear objective. Choose the single result that matters most right now. Just one. Don’t try to do awareness, traffic, and conversions in your first campaign.
• 3. Build one campaign with two or three ad groups. Give each ad group a different audience, for example, keyword targeting vs. follower look-alikes vs. a retargeting audience. Same objective, different people, so you can see who responds.
• 4. Put two or three creatives in each ad group. Vary the hook or the format. Let the platform find the winner.
• 5. Start on Autobid with a modest daily budget. Enough to buy a real sample of results, small enough that a rough first week doesn’t sting.
• 6. Wait. Let it learn. Give it several days without meddling before you draw conclusions.
• 7. Read, then reallocate. Cut the weakest audience and creative, shift budget to the strongest, and make more of what’s winning. Repeat.

That loop, launch, learn, cut, double down, is the entire game. Every good advertiser is just running it faster and with better instincts than they had last quarter.

Common mistakes to sidestep

A few pitfalls trip up nearly everyone the first time, so learn them the cheap way, by reading about them:

• Over-targeting. Stacking so many audience layers that almost no one qualifies. Start broad, then narrow.
• Judging by the wrong metric. Grading an awareness campaign on sales, or a traffic campaign on impressions. Match the metric to the objective.
• Editing too soon. Panicking on day one and resetting the learning phase over and over. Give it room.
• Boosting weak posts. Paid distribution amplifies quality; it doesn’t create it. Put budget behind your genuinely good content, not your leftovers.
• Skipping the pixel. Flying blind on conversions because tracking was never set up. Install it before you spend.
• Set-and-forget. Launching and never looking again. The whole value is in reading results and reallocating.

None of these are hard to avoid once you know them, and avoiding them puts you ahead of most advertisers on the platform.

Bringing it together

Here’s the honest truth this whole X (Twitter) ads guide has been building toward: advertising on X isn’t about secret hacks or the perfect budget number. It’s about matching your objective to your goal, reaching the right people without over-narrowing, putting real creative behind it, and then, this is the part most people skip, measuring honestly and doing more of what works.

Start small, keep good notes, let campaigns learn before you judge them, and treat every campaign as a lesson that makes the next one sharper. The advertisers who win on X aren’t the ones with the biggest budgets. They’re the ones running that learn-and-reallocate loop with the most discipline. You can start running it this week. And if you want your paid efforts sitting on a strong organic foundation, keeping a consistent, well-planned posting rhythm is where it starts, our advice on managing your social presence without burning out pairs perfectly with everything here. Now close this tab, open Ads Manager, and go build that first campaign, one objective, a few audiences, a couple of great creatives. You’ve got this.