Facebook & Instagram Ads for Small Businesses: A Practical Guide by Matan Guy

By Matan Guy · November 11, 2026 · 7 min read

I'm Matan Guy, and most of what I know about paid social I learned the slow way: by spending money on ads that didn't work and then figuring out why. Facebook and Instagram ads can be one of the most useful tools a small business has, but only if you treat them as a system you test and measure, not a button you press when sales are slow. This guide is the version I wish someone had handed me at the start.

One note before we begin: Meta changes Ads Manager often. Menu names, campaign options and default settings move around. Use this as a way of thinking, and always check the current options inside your own account.

How Facebook ads campaigns are structured

Every campaign in Ads Manager has three levels. Understanding them saves a lot of confusion.

Campaign — where you choose the objective, meaning what you want Meta to optimize for.
Ad set — where you choose the audience, placements, budget and schedule (in some setups the budget sits at the campaign level instead).
Ad — the creative itself: the image or video, the text, the headline and the link.

A simple rule I follow: one campaign per goal, a few ad sets, and several ads inside each. Too many campaigns and ad sets spread a modest budget so thin that nothing gets enough data to learn from.

Choosing the right objective

The objective tells the system who to show your ad to. Choose traffic and Meta looks for people likely to click; choose sales and it looks for people likely to buy. Those are not always the same people.

Awareness — reaching as many relevant people as possible. Harder to measure directly.
Traffic — getting clicks to a page. Cheap clicks are easy to get, but they don't always turn into customers.
Engagement — likes, comments, messages, video views. Good for building social proof or starting conversations.
Leads — collecting contact details through a form or your website.
Sales — optimizing for purchases or other conversion events on your site.

Choose the objective that matches the action you actually want. If you sell online and your tracking works, start with sales. If you book calls, leads usually makes more sense.

Broad vs defined audiences in Facebook ads

A few years ago, success depended on detailed targeting: interests, behaviors, lookalikes stacked on top of each other. Today the delivery system is much better at finding buyers on its own, especially when it has good conversion data.

Approach	What it means	When it fits
Broad	Basic filters only (location, age, maybe gender)	You have working tracking and some conversion history
Defined	Interests, behaviors or custom audiences	New accounts, niche products, or local services
Retargeting	People who visited your site or engaged with you	Always worth having once you have traffic

In practice, I often test a broad ad set against one defined audience and let the results decide. What I don't do anymore is narrow the audience so tightly that the ads can barely deliver. Your creative does a lot of the targeting now: the message in the ad tells the right people "this is for you." I go deeper into that in my framework for ad creatives.

Pixel and Conversions API, in plain words

The system can only optimize for what it can see. Tracking is how it sees.

The Meta Pixel is a small piece of code on your website. When someone views a product, adds to cart or buys, the pixel reports that event back to Meta from the visitor's browser.
The Conversions API (CAPI) sends the same kind of events from your server or store platform instead of the browser. Browser tracking can be blocked or lost; server-side tracking fills in many of those gaps.

Most store platforms offer an official integration that sets up both. Once it's installed, use Events Manager to confirm that events arrive and aren't counted twice. If tracking is broken, every later decision rests on bad information.

Testing with a small daily budget

You don't need a large budget to learn. You need a budget that is consistent and a test that is clear.

Here's the process I use:

Start one campaign with a small daily budget you're comfortable losing while you learn.
Use one or two ad sets, each with three to five different ads.
Let it run for several days without touching it. Daily swings are normal; constant editing resets what the system is learning.
Compare ads on the metric that matters for your goal, not on likes or comments.
Turn off the clear losers, keep the winners, and add new creatives to test against them.

Think of the early budget as paying for information.

Key metrics, defined simply

Ads Manager shows dozens of columns. These are the ones I check first.

CPM (cost per 1,000 impressions) — how expensive it is to show your ad. It depends on audience, competition and season.
CTR (click-through rate) — the share of people who saw the ad and clicked. A low CTR usually points to a creative or message problem.
CPC (cost per click) — what you pay for each click. Useful, but cheap clicks alone don't mean much.
Cost per result — what you pay for the outcome you optimized for, like a lead or a purchase. This is usually the most important number.
ROAS (return on ad spend) — revenue attributed to ads divided by ad spend. It's helpful, but it ignores product costs, shipping and fees, so you need to know your margins to read it properly.

Read them together. A healthy CTR with a poor cost per result often means the landing page or offer is the problem. A low CTR means people aren't stopping to look.

When to scale and when to stop

Signs it may be time to scale:

Cost per result has been stable for a while, not just one good day.
The numbers make sense once you account for your real margins.
You have more than one ad performing, so you're not dependent on a single creative.

When you scale, increase the budget gradually rather than doubling it overnight, and keep feeding the campaign new creatives.

Signs it's time to stop or rethink:

You've given a test enough time and spend and it still has no conversions.
Costs keep rising and new creatives don't bring them down.
People click but don't buy — which usually means the problem is the product page, price or offer.

Stopping isn't failure; a test that quickly says "this doesn't work" has done its job. More on that in what losing my first stores taught me.

Common Facebook ads mistakes small businesses make

Boosting posts and calling it advertising. Ads Manager gives far more control over objectives, audiences and testing.
Skipping tracking. Without a working pixel and Conversions API, the system optimizes in the dark.
Editing too often. Constant changes keep campaigns from stabilizing.
Judging too early. One day of data is noise.
Running one ad for months. Every creative eventually tires out its audience.
Ignoring the landing page. An ad can only bring people to the door. The page has to do the rest.

Where Facebook ads fit in a bigger marketing plan

Paid ads work best on top of a clear offer with some trust behind it. If people search your name after seeing an ad, what they find matters, so invest in building trust through a personal brand alongside your ads. More about my background is on the about page.

FAQ

How much should a small business spend on Facebook ads?

There's no single right number. Start with a small daily budget you can treat as a learning cost, keep it consistent for several days, and increase it only once results are stable and make sense for your margins.

Should I run ads on Facebook, Instagram or both?

For most small businesses, I start by letting Meta place ads across both platforms automatically, then look at the placement breakdown in Ads Manager. If one placement clearly performs worse, you can adjust from there.

Do I need the Conversions API if I already have the pixel?

It's strongly recommended. The pixel alone can miss events because of browser settings and privacy tools, and the Conversions API helps fill those gaps. Use your platform's official integration and confirm events in Events Manager.