Ad relevance: why good ads cost less?

Two businesses sell the same product. Both spend $5,000 on Facebook ads this month.

Business A pays $20 to show their ad to 1,000 people. Their $5,000 reaches 250,000 people.

Business B pays $10 to show their ad to 1,000 people. Their $5,000 reaches 500,000 people.

Business B reaches twice as many people for the same money. Several factors affect the price per ad view — audience, competition, time of year — but one of the biggest is how people react to the ad.

The platform charges less to show ads that people actually interact with — click, like, comment, or share.

This price difference is not random. It comes from how the ad auction works.

An auction is a system where multiple buyers compete for the same thing — in this case, the chance to show their ad to a specific person.

Every time someone opens Instagram or Facebook, the system decides which ad to show them. Hundreds of advertisers want that slot.

The system picks the winner using three things:

1. How much you are willing to pay (your bid)
2. How likely the person is to take action (estimated action rate — will they click, buy, sign up?)
3. How good the ad is (ad quality — do people react positively or negatively?)

The system combines all three.

An advertiser with a lower bid but higher predicted action rates and better quality can beat an advertiser who bids more but has a boring or annoying ad.

Why does the system care about engagement?

Facebook makes money from advertising. The more time people spend on the platform, the more ads Facebook can show them.

• If people see ads they hate, they use the app less — and Facebook shows fewer ads
• If they see ads they find interesting, they stay longer — and Facebook makes more money

So the system is designed to show ads people are most likely to interact with.

Facebook shows you these scores in Ads Manager. Here is what they look like.

Facebook gives every running ad three grades. Think of them like report cards — each one measures a different part of the experience.

Quality ranking

This measures if people think your ad is low quality. The system looks at negative signals:

• People clicking hide this ad
• People reporting the ad
• Known patterns like too much text, blurry images, or clickbait

If many people react negatively, your quality ranking drops.

Engagement ranking

This measures if people interact with your ad compared to other ads competing for the same audience. The system counts:

• Clicks
• Comments
• Shares
• Saves
• Video watches

If people scroll past without doing anything, your engagement ranking drops.

Conversion ranking

This measures what happens AFTER someone clicks.

Do they actually do what you want — buy, sign up, download?

This ranking is relative: it compares your conversion rate to other advertisers targeting the same audience.

If your rate is lower, this ranking drops.

Each ranking shows one of three results: above average, average, or below average.

Below average on any ranking = you pay more. Above average = you pay less.

Why? In the auction, the system picks ads most likely to get a good reaction.

If your ad has low grades, it loses to better ads — and you need to bid higher to compete for the same slot.

These grades tell you WHAT is broken. The next step is understanding what each one actually means for your business.

Each low grade points to a different problem. Fixing the wrong one wastes time.

Low quality ranking → people dislike the ad itself

The ad is triggering negative reactions. This could be anything from looking like spam to feeling misleading.

If the ad looks low-effort, people assume the product behind it is low-quality too. Common causes:

• Image or video looks low-effort or confusing — people associate visual quality with product quality
• Text makes promises that feel exaggerated — if it sounds too good, people distrust it
• The ad feels intrusive or irrelevant — people hide ads that feel like they do not belong in their feed

Fix: change the creative — that means the ad itself (the image, video, text, or combination you are running). Test a different visual style, a different message, or a different format entirely.

Low engagement ranking → people ignore the ad

The ad is not bad enough to report, but not interesting enough to stop for.

People just scroll past. Common causes:

• The opening does not grab attention (covered in depth in Lesson 2)
• The message is too generic — it could be about any product
• The ad does not give people a reason to react

Fix:

• Change the hook — the first thing people see or hear that is supposed to grab their attention
• Try a different angle — a different reason why someone should care about your product

The goal is to make people stop scrolling, not just see the ad.

Low conversion ranking → people click but do not buy

People click the ad but then leave without taking action.

This can be a problem with the website, with the audience, or with the gap between what the ad promises and what the page delivers. Common causes:

• Slow website — the longer the page takes to load, the more people leave before it finishes
• Confusing checkout process
• Price or offer does not match what the ad promised
• Targeting people who are curious but not ready to buy

Fix: start with the landing page — the webpage people see after they click your ad.

• Check its speed
• Simplify checkout
• Make sure the page matches what the ad promised

If the page is fine, check whether your targeting is reaching people who are actually in a position to buy.

Each grade points to a specific problem. Here is what they cost you in practice.

Here is what happens to your budget when grades are low.

Say your ad reaches 1,000 people and 20 of them click. That is a 2% click rate — typical for Facebook.

With above average rankings, you might pay $10 per 1,000 impressions.

An impression is one person seeing your ad one time. The industry calls this cost CPM (cost per mille — mille means 1,000 in Latin).

Now imagine your rankings drop to below average. Your CPM might double to $20.

You now need to spend twice as much to get the same 1,000 impressions — and the same 20 clicks.

There is a compounding effect. If your conversion ranking is also low, those 20 clicks produce fewer sales.

You are paying more per click AND getting fewer sales per click.

This is why fixing ad quality often produces better results than raising the budget. A better ad gets cheaper distribution and higher conversion from the same spend.

One caveat: better reach does not guarantee proportionally better sales.

Reaching 500,000 people instead of 250,000 does not automatically mean twice the sales. The additional people you reach may be less interested.

But more reach at the same budget means more chances.

The next question is how to find out what actually works for your specific product and audience.

There is no single ad style that always wins. What works depends on your product, your audience, and your platform.

Some brands do well with polished, professional ads. Others do well with raw, phone-shot content. Some audiences respond to humor, others to data.

You can study competitors and industry benchmarks, but the most reliable way to know what works for YOUR business is to test with your own audience.

How to test:

1. Create 2-3 versions of the same ad with different approaches — different visuals, different opening lines, different formats (video, single image, or carousel — a set of images people swipe through)
2. Run them at the same budget ($50-100 each) for 5-7 days — you need enough impressions for the rankings to stabilize
3. Check the three rankings for each version
4. The version with the best rankings costs the least. Gradually increase its budget and monitor whether rankings hold.

Results at small budgets do not always scale perfectly, but they give you a starting point.

What to look at:

• Which version has the highest engagement ranking? That tells you which approach YOUR audience responds to.
• Which version has the best quality ranking? That tells you which format people do not find annoying.
• Which version converts best? That confirms the whole chain works — from ad to sale.

Common pattern (not a rule):

For many consumer products under $200, ads that feel like regular posts — someone showing the product in a natural way, rather than a polished commercial — tend to get more interaction.

But this is a trend, not a law. Premium brands, technical products, and B2B often need a different approach.

Running both types and comparing the rankings shows which works for a specific audience.

Testing works for most businesses. But there are specific situations where ad quality matters less — or works differently.

Luxury and premium brands

If you sell $5,000 handbags or $50,000 watches, your audience expects polished ads that make people dream about a lifestyle.

A phone-shot video of your product on a kitchen table would damage trust — because luxury buyers judge product quality by how it is presented.

If the ad looks cheap, they assume the product is cheap or fake.

For these brands, quality means high production value — not raw authenticity. The principle still applies (better ads = lower costs), but what counts as better is completely different.

Example: Louis Vuitton and Rolex use cinematic-quality ads. Their audience associates production quality with product quality.

B2B with long sales cycles

If your product costs $50,000+ and takes 6-12 months to sell, a single Facebook ad will not close the deal. The buyer needs meetings, demos, proposals, and internal approvals.

Ad rankings still affect how much you pay to get each potential customer (called a lead — someone who shows interest, like filling out a form or booking a call).

Optimizing ad quality helps you get cheaper leads, but the conversion ranking will always be low because nobody buys immediately.

Example: Enterprise software companies use ads to get people to book a demo call. The sale happens months later after multiple meetings.

Tiny niche audiences (under 10,000 people)

If only 5,000 people in the world need your product, the ad system cannot learn what works.

The algorithm finds patterns by analyzing thousands of reactions — who clicked, who bought, who ignored. With a tiny audience, there are not enough reactions to find patterns.

Your costs will stay high regardless of ad quality because the system is guessing instead of optimizing.

Example: Specialized industrial equipment with 200 potential buyers globally. The auction mechanics barely apply.

Heavily regulated industries

Healthcare, finance, alcohol, and gambling have legal rules about what you can say and show.

Engaging content may get rejected by compliance. Even a great ad might get blocked before it runs.

In these industries, the legal constraints override creative optimization.

Example: Pharmaceutical ads must include side effects, disclaimers, and approved language. There is limited room for creative testing.