How to Run Facebook Ads (Meta Ads) in 2026: The Complete Beginner's Guide

To run Facebook ads, which Meta now calls Meta ads, you install tracking, choose an objective and an optimization event, and build one simple campaign in Ads Manager with varied creative. Then you launch, leave it alone through the learning phase, read your results over days and weeks, and test new creative before you scale what works.

Running Facebook ads yourself is simpler than most people think. If you spend under $5,000 a month, it's often more effective than handing everything to an agency. That's exactly the budget range where the small business strategy article maps the $20 a day framework before you touch Ads Manager. Meta ads isn't rocket science. I've managed $35M+ in Meta ad spend across 50+ brands over about ten years, and winning accounts usually do a few simple things well. This guide walks you through each of them, from an empty ad account to scaling your first winner.

Last updated: September 2026. By Victoria Alenich, Meta Ads Consultant | $35M+ managed across 50+ brands including foodspring, Asana Rebel, and Kitchn.io.

What Are Meta Ads and How Do They Work?

Meta ads are the paid ads you run across Meta's apps from one tool, Meta Ads Manager. A single campaign can appear on Facebook, Instagram, Messenger, the Audience Network, Threads and WhatsApp Status. So how to run Facebook ads and how to run Meta ads are really the same question. Facebook was simply the first placement, and the old name stuck.

Each time someone opens one of these apps, Meta runs an auction for the ad slot. According to Meta's page About ad auctions, the winner is the ad with the highest total value, which combines your bid, the estimated action rate and ad quality. The estimated action rate is how likely this person is to take the action you asked for. That's why an ad people actually want to see can beat a larger bid.

I call the algorithm behind all this your silent super employee. Imagine hiring the best media buyer in the world, someone who works 24/7 and knows exactly who to show your ads to. Now imagine getting them for free. The catch is that this employee can't talk. You communicate through the events your tracking sends back, which tell it what success looks like, and through your creative, which tells it who the ad is for. Most of this guide is about getting those two signals right.

Meta also works differently from Google. On Google you wait for someone to search for what you sell, which is pull marketing. On Meta you reach people as they scroll, often before they know they need your product. My Facebook ads vs Google ads comparison explains when each one makes sense.

What You Need Before You Start

You need a Facebook Page. Even if you only want to advertise on Instagram, Meta asks for a Facebook Page as your business identity. It can be completely empty, and facebook.com/pages/create takes about five minutes.

You need a Business Manager account. This is your headquarters on Meta, where ad accounts, Pixel, audiences, billing and team access live. Meta now calls it a business portfolio, and you create it at business.facebook.com. Connect your Page and your Instagram account there, and give team members their own access so nobody needs your personal login.

You need a product people already buy. Ads amplify what's working, and if nobody buys organically, ads will only show you that faster. A handful of organic sales or a few clients from word of mouth is enough proof to start.

You need a landing page that converts. Most people will see your ad on their phone, so the page has to load fast on mobile and make the next step obvious. Every extra second of load time can raise your bounce rate by around 20%. Keep it to one page, one offer and one call to action.

Read Meta's advertising standards too. Repeated violations can restrict you personally from advertising, and ads for credit, employment, housing, social issues or politics need a special ad category.

---

Step 1: Set Up Tracking (The Foundation Everything Else Depends On)

This is the most important step and the one beginners skip most often. Without tracking, your super employee works blind, because it can't see who buys or who is worth showing your ads to. I've seen businesses spend thousands before realizing their Pixel wasn't firing.

Install the Meta Pixel

The Meta Pixel is a small piece of code on your website that records what visitors do after they click your ad. It sees which pages they view, what they add to their cart and whether they buy or submit a form.

To create it, open Events Manager from business.facebook.com, click Connect data sources, choose Web and follow the prompts. On Shopify, install Meta's Facebook & Instagram sales channel app and choose the Maximum data sharing setting. That sets up the Pixel, standard events and the Conversions API in a few minutes. On WordPress or WooCommerce, use the official Meta Pixel plugin or Facebook for WooCommerce. Wix, Squarespace and most other platforms have a direct integration in their settings, and custom sites can use the setup wizard in Events Manager.

Set Up Your Conversion Events

Events are the actions that matter to your business, and choosing them is how you tell the algorithm what success looks like.

For e-commerce, track the whole shopping path. That means ViewContent on product pages, AddToCart, InitiateCheckout and Purchase. Pass the product ID, price and purchase value with them. The more detailed the receipt you give Meta, the better it understands what drives your sales.

For lead generation, track Lead and CompleteRegistration. Lead fires when someone submits a form or books a call, and CompleteRegistration when they finish a signup.

For apps, use the Meta SDK, the app version of the Pixel, to track installs, first opens and purchases or subscriptions inside the app.

Verify everything with the Meta Pixel Helper Chrome extension by visiting your product, cart, checkout and confirmation pages, or use the Test events tab in Events Manager. Some older guides tell you to rank your top eight events for Aggregated Event Measurement. You can skip that, because Meta removed the eight event limit in 2023, as Adviso explains.

Set Up Conversions API (Don't Skip This)

Since Apple's iOS 14.5 privacy changes, people can opt out of tracking and browsers block more signals. The Pixel alone now misses part of your conversions. The Conversions API, or CAPI, sends conversion data from your server or shop platform straight to Meta. The Pixel watches from the browser, where it can be blocked, and CAPI reports from the server, where it can't. Together they give Meta the most complete picture. Turn on Advanced Matching as well, which sends hashed details like email addresses so Meta can match more actions to your ads.

Keep Meta's default attribution of 7 day click and 1 day view to begin with. Expect Ads Manager to differ slightly from Shopify or your CRM, too. One client was ready to pause everything because Shopify showed 50 purchases and Meta showed 43. Neither number was wrong, because each tool counts in its own way.

---

Step 2: Understand How Meta Ads Are Structured

Meta ads work on three levels, like a waterfall.

The campaign level is your goal. Here you choose what you want, whether that's sales, leads or installs, and you set your budget if you use a campaign budget.

The ad set level holds your rules. Here you decide who can see your ads, where they appear and, most importantly, which optimization event the algorithm should chase.

The ad level is your message. This is the image or video, the text, the headline and the button that people actually see.

The mistake I see most often is building too much. I've audited accounts with 15 campaigns, 40 ad sets and 200 ads competing against each other. The budget was split so thinly that nothing could learn. Simple structures beat complex ones for the vast majority of advertisers, so start with one campaign.

---

Step 3: Create Your First Campaign

Open Ads Manager at business.facebook.com/adsmanager and click Create.

Choose Your Campaign Objective

Meta has six objectives, Awareness, Traffic, Engagement, Leads, App promotion and Sales, as listed in Meta's campaign objectives help center. For a growing business, three of them matter.

I've honestly never seen Traffic used profitably for a growing business. Every week I audit accounts where someone chose it because they just wanted visitors, then got thousands of clicks and no sales. The algorithm finds exactly what you tell it to find.

What Advantage+ Means in 2026

Advantage+ is Meta's name for its automation, and since February 2025 it has been part of the normal campaign setup. Social Media Today reported that an Advantage+ on indicator now appears when a campaign uses Advantage+ audience, Advantage+ placements and Advantage+ campaign budget together. Advantage+ Sales, which grew out of the old Advantage+ shopping campaigns, is a campaign type inside the Sales objective. For a beginner, I'd leave these automations on. They hand your super employee the jobs it's best at and leave you in charge of your event, your offer and your creative.

Set Your Budget

Use a campaign budget, which Meta now labels Advantage+ campaign budget and most people still call CBO. It moves money between your ad sets in real time toward whichever one performs best. With a daily budget, Meta can spend up to 25% more on a good day while keeping your weekly average on target.

How much should you start with? Honestly, start with what you can invest today. Meta's minimums are roughly $1 a day for ad sets charged on impressions and $5 a day for ad sets charged on clicks. Check Ads Manager for the amounts in your currency. I've seen businesses start at €3 a day and scale to €300 a day with the principles in this guide.

A small budget mostly changes how fast you learn. My rule of thumb is a daily budget of at least three times your target cost per result. If you want purchases at $30, that's $90 a day or more. If that's out of reach, keep Highest Volume bidding and optimize for a more frequent event, as Step 4 explains. The minimum budget section of my cost guide shows what $5, $10 or $20 a day can realistically do.

Choose Your Bid Strategy

Start with Highest Volume, which gets you as many results as possible within your budget while you learn what results cost. After a few weeks of data, you can try a Cost per result goal. Start above your average and lower it by about 10% at most every three days. Save ROAS goal and Bid cap for when you have around 50 conversions a week and clear profit numbers.

---

Step 4: Set Up Your Ad Set

Choose Your Optimization Event (Most Important Setting)

Your optimization event is possibly the most powerful targeting tool you have. It tells the algorithm what kind of person to look for, and it studies everyone who completed that event to find more people like them.

Pick the event closest to your business goal that still happens often enough to learn from. According to Meta's help center, an ad set usually needs around 50 optimization events within seven days to exit the learning phase.

If you sell products, optimize for Purchase. If you only get a few purchases a week, step up to InitiateCheckout, or to AddToCart if checkouts are rare. Move back to Purchase once volume grows. I wouldn't go further up than AddToCart, because ViewContent and PageView find people who like to browse.

If you generate leads, optimize for Lead. With volume, you can optimize for a custom conversion, like leads who say in the form that their company has more than 100 employees. That turns the algorithm into your lead qualifier.

If you promote an app, optimize for installs first, then move to a purchase or subscription inside the app once you have volume.

Avoid Link clicks and Landing page views when you want sales or leads. The algorithm will find serial clickers who buy nothing, so if you want purchases, tell your super employee you want purchases.

Set Your Targeting

Broad targeting usually beats detailed interest targeting in 2026, which surprises most beginners. With a good optimization event, your creative does much of the targeting work. The algorithm reads your ad, tests it on a slice of a broad audience and finds more people like those who convert.

Meta has moved in the same direction. Advantage+ audience is on by default for new ad sets, as Meta for Developers announced in June 2025, and detailed targeting exclusions were removed in 2024, as Social Media Today reported.

For your first campaign, set your location, keep the age range wide and leave Advantage+ audience on. You can add a few broad interests or a lookalike as audience suggestions, and Meta will start there before it expands. For a deeper look at these signals, read Facebook ads targeting in 2026. Custom audiences and lookalikes come later, once you have visitors and customers, and a lookalike works best from a source of at least 1,000 people.

Choose Your Placements

Leave Advantage+ placements on, so your ads can run wherever they perform best across Feed, Stories, Reels, Marketplace, Messenger, the Audience Network and newer placements. TechCrunch reported that Threads ads rolled out worldwide in January 2026, and Meta announced ads in WhatsApp Status in June 2025. Advantage+ placements can include both. The Instagram version of this guide covers Promote vs Ads Manager and creative for each placement.

Build your creative in 9:16 vertical format. It fills the screen in Stories and Reels, usually the strongest placements, and Meta can crop it for Feed. My Instagram Story ads guide covers the specs and the safe zone in detail.

---

Step 5: Create Your Ad Creative (The Make or Break Factor)

In my experience, better targeting might improve results by 10 to 15% and smarter bidding by 10 to 20%. The right creative can make an ad perform three, four or five times better. What makes a Facebook ad creative work matters more than targeting, bidding and structure combined. Creative is also how you talk to your super employee, because the algorithm reads your ad to decide who should see it.

One skincare client had a messy account, with competing campaigns and dozens of audiences. Still, one simple, authentic before and after video drove 80% of their sales at a third of the cost of their other ads. A B2B software client had a perfectly structured account running corporate stock photos, with a cost per lead of €280. We replaced the creative with raw videos of their CEO explaining the product, and within days the cost per lead dropped to €45.

Video vs Images

Video tends to outperform static images for most businesses, and the bar for good enough video is lower than people think. A founder talking to camera, a quick demo on the kitchen counter or a customer recording a short testimonial on their phone is exactly what works. Products with strong visual appeal, like jewelry or fashion, can still do well with striking images. One easy trick is turning your best image into a video by adding music. Test both and let the data decide.

Hook, Body and Close

The hook has to stop the scroll in the first second or two. If people keep scrolling, nothing else in your ad matters. Pattern interruption works best, like a surprising statistic, a provocative question, a bold statement or a moment from behind the scenes. An image gets about half a second, so give it one clear focal point, like a before and after split.

The body delivers the value and backs it up. In the next five to seven seconds, say what the product means for the viewer. Skip the talk about your advanced formula and show that they'll see results in seven days. Until around 20 seconds, prove it with the product in use, results, social proof and an answer to the main objection. Cut to something new every two to three seconds so attention doesn't drop.

The close tells people what to do. Many viewers never reach the last second, so weave softer prompts in early and finish with a clear ask that matches your button. Use urgency only when it's real.

Format Specifications and the Safe Zone

Always start with 9:16. Cutting a vertical asset down for Feed is much easier than stretching a square one. In my tests, square videos in Stories and Reels often perform 30 to 40% worse because they use about half the screen.

Design for the safe zone. In Stories and Reels, your brand name and the Sponsored label sit at the top, engagement buttons on the right, and the button and caption at the bottom. Leave roughly 20% clear at the top and bottom, and keep your product, key text and main action in the middle. That part also survives when Meta crops the ad into a square. I've audited hundreds of accounts where the headline or product was hidden behind the interface.

Authenticity Beats Production Value

Ads that look like organic content, slightly imperfect, shot on a phone and featuring real people, usually beat polished commercial ads. People have learned to scroll past anything that looks like an ad. I've seen iPhone videos outperform €50,000 agency productions many times. A recent phone, a ring light for about €50, a tripod and a free editor like CapCut are enough. One hour of filming several hooks and angles back to back can give you material for 15 to 20 ad variations.

Write Your Ad Copy

The primary text has to earn the See more click. Only the first few lines show, so open with a question about a pain point, a surprising fact or a specific promise. Use line breaks so people can scan. Nobody cares about a welcome to a brand they don't know yet.

The headline should state the value in 25 to 30 characters. "Save 40% On Running Shoes" beats "Great Deals on Shoes" because specific numbers beat vague promises. The description doesn't show in every placement, so a line of social proof is plenty there.

The button should match the decision you're asking for. A furniture brand I worked with sold more of its $13,000 sofas with Learn More than with Shop Now, because nobody buys a sofa like that on impulse. For a B2B software client, changing Sign Up to Get Started lifted conversions by 15%, with the same page and the same offer.

---

Step 6: Launch and Wait (This Is Where Discipline Matters)

Click Publish, then leave the campaign alone for five to seven days. Your first campaign needs to be running, and perfection can come later.

A new ad set starts in the learning phase, while the algorithm explores who responds. As Meta explains, it typically needs around 50 optimization events within seven days of launch, or of its last significant edit, to exit learning. Results swing until then. Changes to targeting, creative, the optimization event or the bid strategy, and large budget changes, count as significant edits that can send an ad set back into learning.

The most expensive beginner mistake is panicking during the learning phase. People see one bad day, change the budget, swap the creative and adjust the targeting, and the campaign never stabilizes. One client changed something every morning. When we convinced them to leave a promising campaign untouched for a week, their cost per conversion dropped by 30%. You wouldn't fire a new employee after one day without a sale, and your silent super employee deserves the same week to learn the job.

During the learning phase, watch and take notes. After five to seven days, look at the average for the whole period. If your cost per result is within 1.5 to 2 times your target, you have something promising. If you want €10 leads and you're paying €40 or €50, keep testing new creative before you build anything bigger. My checklist in Facebook ads not working? 9 things to check helps with a full diagnosis. If an ad set shows Learning limited, it isn't getting enough events. Combine ad sets or move your event one step up the funnel.

---

Step 7: Read Your Results (The Metrics That Actually Matter)

Ads Manager shows dozens of metrics, and most of them are noise. Read the ones below together, and look at both where you are today and where you're heading.

Is Your Ad Getting Attention?

CPM, the cost per 1,000 impressions, tells you how expensive your attention is. There's no good or bad CPM on its own. I've seen campaigns succeed at $50 and fail at $2, because what matters is whether your click and conversion rates can carry the price. The cost per 1,000 views section of my cost guide has current benchmarks.

Link CTR counts only clicks to your site, so watch it more closely than overall CTR, which includes likes and comments. I once had an ad with a 5% overall CTR and a 0.5% link CTR, entertaining people without selling anything.

Hook rate, the share of impressions where someone watched at least three seconds, tells you whether your opening works. Frequency shows how often the average person saw your ad, and for cold audiences I keep it under about 4 a week.

Is Your Landing Page Working?

Your click to page view ratio compares landing page views with link clicks, and anything below 80% means people leave before your page loads. Test it at pagespeed.web.dev. Conversion rate is results divided by link clicks. If CTR is healthy while conversion rate is poor, look at your page and offer before you touch the ad.

Are You Making Money?

Cost per result (CPA or CPL) decides profitability. If a product sells for $60 and costs $30 to deliver after fees and overhead, $30 is your breakeven CPA. ROAS is revenue divided by ad spend, and your breakeven ROAS is one divided by your margin, so a 50% margin needs 2x to break even. How much Facebook ads cost walks through breakeven ROAS by business model. Watch average order value too. One brand I worked with cut its cost per purchase from $40 to $25 while its order value fell from $70 to $55.

How to Judge Cost per Result Against Your Target

Judge trends over days and weeks. Three or more periods moving the same way make a real trend, and one or two bad days are usually normal variation.

Which Metrics Matter Most by Business Type

---

Step 8: Set Up Retargeting (Your Second Campaign)

Once your first campaign has run for a week or two and built up visitors, add retargeting. These people already know you, and I regularly see retargeting deliver 3 to 5x ROAS while prospecting delivers 1.5 to 2.5x.

How to Set It Up

Retargeting shows ads to people who visited your site, viewed a product, added to cart, watched a video or engaged with your posts without converting. Create a campaign with the same objective as your prospecting campaign and build custom audiences from these actions. Good starting points are website visitors from the last 7 to 30 days and people who added to cart in the last 7 to 14 days. Video viewers who watched 50% or 75% of your ad work well too. So do people who engaged with your Facebook or Instagram in the last 30 to 90 days. For e-commerce, a catalog sales campaign can automatically show each person the products they viewed.

Advantage+ audience is on by default for new ad sets. With it on, a custom audience works as a suggestion Meta can go beyond. Switch to the original audience options when you want a retargeting ad set to stay with your warm audience.

Budget for Retargeting

Keep most of your budget on prospecting. For e-commerce I split it roughly 75 to 95% for prospecting, 5 to 20% for retargeting and 2 to 5% for a loyalty layer aimed at past customers. At $30 a day, $25 for prospecting and $5 for retargeting works well. The loyalty layer can wait until you have a real customer list. As prospecting grows, you'll have more people to retarget, and that share can grow with it.

Retargeting Creative Strategy

Don't show retargeting audiences the ad they already saw. Something stopped them, and the same ad won't remove that barrier. Answer the likely reason instead. Use reviews for trust, a small incentive like free shipping for price and a simple reminder for the distracted. People who need more information want answers to the common objections. One skincare brand I worked with used a video opening with the problem for cold audiences. For retargeting it used a before and after version, because people who already knew the brand wanted proof. Online shops that want help with catalog, creative and retargeting can work with me through Meta ads consulting for e-commerce.

---

Step 9: Test, Learn, Scale

Your first campaign probably won't be a home run, and that's expected. The businesses that succeed treat creative testing as their main ongoing job. About 60 to 70% of my own time on an account goes into creative.

Creative Testing Is Your #1 Ongoing Priority

Meta's A/B tests suit big strategic questions. For regular creative testing they're slow and expensive, because they split your traffic and need many events before they answer. I prefer to let the algorithm compare concepts inside a CBO campaign, and your first campaign can already work this way.

1. Set up one CBO campaign with Highest Volume bidding and broad targeting.
2. Create three or four ad sets with a genuinely different concept in each, like a testimonial, a demo, a before and after and a founder story.
3. Keep the optimization event, targeting, placements and bid strategy identical across them.
4. If your budget allows, give each test ad set a minimum daily spend of 1.5 to 2 times your target CPA.
5. Leave it for five to seven days, and aim for around 50 results across the test before you call a winner.

The campaign budget flows toward concepts that work, and weak ones get little spend. Keep any concept that matches or beats your current average, then test variations of it, like new hooks or first frames. For one skincare brand, four versions of the same serum video with different openings each attracted a different kind of buyer. Log what you changed in each variation, and after a few months you'll see clear patterns.

Creative fatigue is real, and it arrives faster than it used to. In 2019 you could run an ad for months. Today I see ads fatigue within a week in competitive industries, so build a routine for continuous creative production.

Scaling What Works

When a creative stays profitable for five or more days in a row, scale it gradually. My guide on how to scale Facebook ads without wrecking your CPA covers the full protocol. Raise the budget by 15 to 20%, then wait about 48 hours and check your metrics before the next step. That way $30 a day becomes $35, then $42, then $50. If costs rise, wait until they settle, and keep any budget cuts to about 20% a day as well.

Never double your budget overnight. A client once found great results at $50 a day and jumped straight to $200. Their cost per result doubled overnight, and it took three weeks to recover.

What If Nothing Works?

If you've tested 10 or more concepts over two or three weeks and nothing comes within 2x of your target, look outside your campaign settings. Your tracking may be broken, especially after a website update, so check Events Manager again. Your landing page may not convert, so test it on your phone and make sure it delivers exactly what the ad promised. One client's lead form asked for 12 fields, and cutting it to three doubled their conversion rate. Your offer may not be strong enough, and no advertising skill sells a product nobody wants at a price nobody will pay.

Sometimes costs stay at 2x your target or more, you get fewer than 15 to 20 events a week and learning drags past seven days. Then it's time to start fresh with fewer, broader ad sets and a more frequent event. One client with an online store went from ten small ad sets to three, and their cost per purchase dropped by 30% in two weeks. For the full walkthrough, see Facebook ads not working? 9 things to check.

---

What About Meta Andromeda?

Andromeda is Meta's ad retrieval system, the first stage of delivery that picks which ads are even considered for a person before ranking and the auction. Meta deployed it in the second half of 2024, and Meta Engineering described it in December 2024. Since then, people have blamed it for almost every change in their results.

My take is simple. Andromeda is an updated part of the same delivery system your super employee has always used, and for a beginner there's nothing new to learn. Because retrieval is personalized, it rewards accounts that give the algorithm genuinely different creative to match to different people. It also works best with broad targeting, a good optimization event and a simple structure that keeps your data together. Those are the principles this guide is built on, so if you follow the steps above, you're already set up for it. For the full story of what changed and what didn't, read my Meta Andromeda explainer.

---

How Much Do Meta Ads Cost, and Are They Free?

Meta ads are paid. Creating a Page, a business portfolio and an ad account costs nothing, and organic posting is free. Ads run through the auction, and you pay for the delivery you win, usually per 1,000 impressions. There's no way to run Facebook ads for free, although you control exactly how much you spend with your daily or lifetime budget.

When people ask what Facebook ads cost per 1,000 views, they usually mean CPM. The cost per 1,000 views section of my cost guide compares current ranges across several data sources. If you see a charge you don't expect, it's usually Meta's billing threshold, since new accounts are billed each time spend reaches a small amount. Why did Meta charge me $2? covers this and the other common reasons.

---

Common Beginner Mistakes (And How to Avoid Them)

Don't rely on the Boost button. Boosting a post gives you limited targeting, weak conversion optimization and poor tracking. Use Ads Manager instead, and see Facebook ads vs boosted posts for the full comparison.

Don't launch without tracking. Without firing events the algorithm can't learn, so set up tracking before you spend your first dollar.

Don't change things too quickly. One bad day is no reason to edit a campaign, and panic editing kills more campaigns than weak creative does.

Don't target too narrowly or optimize for clicks. Stacking 15 interest layers leaves the algorithm no room, and click optimization finds people who click, while purchase optimization finds people who buy.

Don't ignore mobile or run too many campaigns. Try to buy from your own site on your phone before you launch, and start with one campaign, adding retargeting once you have visitors.

---