Solved: Optimizing Meta Ads Campaigns by Cost Per Result

Hi there,

Thanks for reaching out! I saw your question about Meta ads and since you've got a long history with Google Ads, it's completely understandable why you're asking about those specific optimisation levers. The two platforms are quite different beasts. I'm happy to give you some of my initial thoughts and guidance based on my experience running these campaigns day in, day out. It might be a bit of a mind shift from what your used to with Google.

There's a fair bit to cover here, so I've tried to break it down into a few areas that should hopefully shed some light on how to approach Meta ads for better results. Tbh, it's less about fiddling with micro-adjustments and more about getting the foundations right from the start.

The first thing we'll need to look at is your campaign objective...

Okay, so the most immediate and glaring issue from your post is the campaign objective. You mentioned you're running an 'Engagement' campaign with the goal of getting 'button clicks on your website'. This, right here, is almost certainly the root cause of your problems and why you're not getting the performance you want. It's a very common mistake for people new to Meta, especially coming from other platforms.

Here’s the thing about Meta's algorithm: it is incredibly literal. It does *exactly* what you tell it to do. When you select 'Engagement' as your objective, you are telling Facebook's AI, "Please go and find me people within my target audience who are most likely to like, comment on, and share my ad." The system then looks at user history and finds the serial 'likers' and 'commenters'. These people are brilliant for creating social proof and making an ad look popular, but they are often terrible at actually leaving the platform to visit a website, let alone clicking a specific button once they get there. You're basically optimising for vanity metrics that have little to no correlation with your actual business goal.

For your stated goal – 'button clicks on my website' – you absolutly need to be using a 'Conversions' objective (or 'Sales' as it's often called in the newer Ads Manager interface). By selecting 'Conversions', you tell the algorithm, "Please go and find me people within my target audience who are most likely to visit your website AND complete a specific action." This is a completely different instruction and sends the AI off to find a completely different type of person. It will look for users who have a history of clicking on ads, visiting external sites, and converting.

To do this properly, you need the Meta Pixel installed on your website, and you need to have a specific 'Standard Event' or 'Custom Conversion' set up to fire when someone clicks that target button. This is non-negotiable. Without this tracking in place, Meta has no way of knowing who is converting, and therefore it can't learn or optimise. The pixel is the brain of your entire operation. Giving it the wrong goal is like giving a satnav the wrong destination address; you'll definitly end up somewhere, but it won't be where you wanted to go.

I can't stress this enough: changing this one setting from 'Engagement' to 'Conversions' and ensuring your pixel is tracking the button click correctly will have a more profound impact on your 'cost per result' than any other tweak you could possibly make. You'll likely see fewer likes and comments, but the number of actual button clicks and the quality of the traffic should improve dramatically. It's the difference between getting a crowd of window shoppers and a line of actual buyers.

I'd say you need to shift your mindset from Google to Meta...

Your question about optimising by device, demographic, location, and day-parting is a classic Google Ads way of thinking. In Google Search, you're managing an auction based on intent, and manually adjusting bids for these different segments is a core part of optimisation. You might increase bids for mobile devices if they convert better, or for certain postcodes, or during business hours. It's very hands-on and granular.

Meta works very differently. While you *can* see performance breakdowns for all those segments in the reporting dashboard, you don't typically manage them with manual bid adjustments. Meta's approach is far more automated and relies on the power of its algorithm to make those decisions for you in real-time. The platform's philosophy is that its AI can process billions of data points far more effectively than a human can to determine who to show an ad to at any given moment to achieve your objective at the lowest cost.

So, instead of you telling Meta "I want to bid 20% more on iPhones in London between 9am and 5pm," you simply tell Meta "My audience is people in London, and my goal is a website conversion." The algorithm then does the heavy lifting. It will automatically learn that people on iPhones convert better and will start showing your ads to more of them. It will learn that conversions happen more frequently in the evening and will shift budget there. This is all done dynamically within the ad set's delivery.

The real optimisation levers on Meta are not these granular adjustments. The levers you need to be pulling are much higher up:
-> Audience Targeting: Who you are showing your ads to. This is the biggest factor.
-> Ad Creative: What you are showing them (the image/video and the copy).
-> The Offer: The value proposition on your landing page.

Trying to manually control device or placement performance is often counter-productive. For instance, many advertisers used to manually exclude the Audience Network placement because it had a reputation for low-quality traffic. However, with a 'Conversions' objective, Meta is smart enough to only use that placement if it can find cheap conversions there that you wouldn't have gotten otherwise. By trusting the system (and giving it the right objective), you often achieve a better overall 'cost per result'. Your job is to feed the machine with good audiences and good creative, not to micromanage its every move. This is a big adjustment, but once you embrace it, managing Meta campaigns becomes a lot more effective.

You probably should structure your account for conversions...

So, if manual bid adjustments are out, what does a good structure look like? As an agency, when we audit new client accounts, the most common issue after the wrong objective is a messy or non-existent structure. People will have dozens of ad sets with overlapping audiences, all competing against each other, with no clear strategy.

A proven method is to structure your account based on a marketing funnel: Top of Funnel (ToFu), Middle of Funnel (MoFu), and Bottom of Funnel (BoFu). This organises your efforts logically, from reaching cold audiences to converting hot prospects. This is for an eCommerce account, but the logic applies to almost any buisness.

META ADS AUDIENCE PRIORITISATION & FUNNEL STRUCTURE

ToFu: Prospecting Cold Audiences

This is where you find new people who have never heard of you before. The goal is to introduce them to your brand and drive them to your website for the first time. Your main audiences here are:

-> Detailed Targeting: These are the interest, demographic, and behaviour-based audiences Meta provides. The key here is to be specific. As you're coming from Google, think of this like picking broad match keywords vs. exact match. For example, if you're selling high-end photography courses, targeting the broad interest 'Photography' is probably a bad idea. It's too vast and full of amateurs. Instead, you'd target interests like 'Leica Camera', 'Hasselblad', pages of famous professional photographers, or magazines like 'Aperture'. You need to think, "what interests are people in my target audience far more likely to have than the general population?" That's the secret to good interest targeting. Avoid the temptation to just pick the biggest, most obvious interests.

-> Lookalike Audiences (LALs): This is one of Meta's most powerful tools. You give Meta a source audience (e.g., a list of your past customers, or people who have converted on your website), and it builds a new, much larger audience of people who share similar characteristics. The quality of your source audience is everything. A Lookalike of your best customers will always outperform a Lookalike of all your website visitors. As soon as you have enough conversion data (you need at least 100 events, but tbh you want more like 1,000 for a really good LAL), you should be building Lookalikes based on your most valuable conversion events. Start with a 1% Lookalike in your target countries, as this will be the most similar to your source, and then test broader percentages (1-3%, 3-5%) as you scale.

MoFu: Engaging Warm Audiences

These are people who have shown some interest but haven't taken that final step. They know who you are, but they need more convincing. You are not trying to sell to them directly yet, but rather nurture them, build trust, and keep your brand top of mind. Audiences here include:

-> Video Viewers: People who have watched a significant portion of your video ads (e.g., 50% or more).
-> Page/Profile Engagers: People who have liked, commented, shared, or saved a post, or visited your Facebook/Instagram page.
-> Website Visitors (Excluding Converters): A broad audience of everyone who has visited your site in, say, the last 30-90 days, but hasn't completed your desired action.

The messaging for this group should be different. You might show them testimonials, case studies, behind-the-scenes content, or educational material that addresses their potential pain points. You're building a relationship.

BoFu: Closing Hot Prospects

This is your retargeting audience. These people are on the verge of converting. They've shown strong intent, and your job is to give them that final nudge. These are your most valuable audiences and should deliver the highest Return on Ad Spend (ROAS).

-> Website Visitors (e.g., last 7-14 days): A more recent, and therefore hotter, segment of your site visitors.
-> 'Add to Cart' or 'Initiate Checkout' (but not Purchased): For eCommerce, this is gold dust. For a service or lead-gen model, the equivalent might be "visited the contact page but didn't submit the form" or "clicked to view pricing but didn't sign up". You can create custom audiences for these specific URL visits.

Your ads for this group should be direct. You might offer a small discount, remind them of the benefits, create a sense of urgency ("limited spots available"), or show social proof like "Join 5,000 others who...".

Structuring your campaigns this way (e.g., one campaign for ToFu, one for MoFu/BoFu) with proper audience exclusions (e.g., exclude all website visitors from your ToFu campaign, exclude converters from your BoFu campaign) prevents audience overlap, allows you to tailor your messaging correctly, and gives you clear visibility on what's working at each stage of the journey.

You'll need to focus on testing your targeting and creative...

Once you have the right objective and a logical structure, optimisation becomes a systematic process of testing. This is where you'll spend most of your time. Your job is to constantly find new winning audiences and new winning creatives to beat your current controls.

On the targeting side, you should be methodically testing different audience hypotheses. In your ToFu campaign, you could have different ad sets for:
-> Ad Set 1: Lookalike (1%) of Purchasers
-> Ad Set 2: Interest Group A (e.g., Competitor Brands)
-> Ad Set 3: Interest Group B (e.g., Related Software/Tools)
-> Ad Set 4: Interest Group C (e.g., Industry Publications/Influencers)

You let them run for a few days (how long depends on your budget, but a general rule is to let an ad set spend at least 1-2x your target CPA before making a decision) and then you analyse the results. If Interest Group B is getting you conversions at half the cost of the others, you allocate more budget to it. If Interest Group C isn't delivering, you turn it off. Then you introduce a new audience to test against your winner. It's a perpetual cycle of 'test, learn, scale, repeat'.

On the creative side, this is arguably even more important. A powerful creative can make a mediocre audience work, while a poor creative can make even the best audience fail. You need to be testing constantly. We've seen massive sucess with some clients just by changing the creative. For one B2B software client, we reduced their CPA from a painful £100 all the way down to just £7, and a huge part of that was moving to a better creative strategy and campaign structure.

You should be testing:
-> Formats: Single Image vs. Carousel vs. Video. Video usually performs best but requires more effort. Don't discount simple images, they can be very effective for getting a message across quickly.
-> Hooks: The first 3 seconds of your video or the headline of your ad. You need to grab attention immediately. Test different opening lines and questions.
-> Angles: The core message of your ad. Are you focusing on a pain point? A benefit? A specific feature? Social proof? We've worked with SaaS clients who have seen brilliant results from simple User-Generated Content (UGC) style videos – just a quick screen recording of a happy customer talking about the product. This can feel more authentic and trustworthy than a polished corporate ad.
-> Call to Action (CTA): The text on your button. 'Learn More' vs. 'Sign Up' vs. 'Get Offer' can make a difference.

Your creative will eventually fatigue, meaning your audience gets tired of seeing it and performance drops. Having a constant pipeline of new creative to test is essential for long-term success and scaling your spend.

We'll need to look at what's a realistic cost per result...

This is the million-dollar question, isn't it? The honest answer is: it depends. Your cost per result (in your case, a button click, but let's call it a 'lead' or 'signup' for benchmarking purposes) is influenced by so many factors: your industry, the competitiveness of your audience, the country you're targeting, the quality of your creative, and the conversion rate of your landing page.

However, I can give you some general ballpark figures based on what we see across many accounts. We can break it down by developed vs. developing countries, as this has a huge impact on click costs.

In developed countries (UK, US, Canada, Australia, Western Europe, etc.), you can expect a Cost Per Click (CPC) to be in the £0.50 - £1.50 range. A decent landing page might convert visitors at a rate of 10-30%. Doing the maths on that gives you a potential Cost Per Acquisition (CPA) of anywhere from £1.60 to £15.00 per lead/signup.

In developing countries, the costs are much lower. A CPC might only be £0.10 - £0.50. With a similar conversion rate, your CPA could be between £0.33 and £5.00. However, and this is a big however, the quality of leads from these regions can often be much lower, so cheaper isn't always better. You have to balance cost with lead quality.

Here's a table to visualise it for a lead/signup objective:

These are just averages, of course. As an example, I remember one campaign where we achieved 4,622 registrations for a B2B software at a cost of just $2.38 each. But we've also seen other niches where a £20 CPA is considered a great result. The key is to establish your own baseline and then work methodically to improve it through the testing I mentioned above.

If your costs are higher than these ranges, it's a strong signal that something is wrong – likely your objective, your targeting, your creative, or your landing page. You need to diagnose which part of the funnel is broken and fix it.

I know this is a tonne of information to take in, especially when you're used to a completely different system. To make it a bit more digestible, here's a summary of my main recommendations for you to implement.

Summary of Main Recommendations

It's not just about setting up an ad and hoping for the best. As you can see, it's about understanding your audience, having a robust testing methodology, creating compelling ads, and fine-tuning your entire funnel. It can be a full-time job in itself, and it's a very different skillset from managing Google Ads.

That's where a professional consultancy like us can make a huge difference. With years of experience and a deep understanding of the Meta advertising landscape, we can help you bypass the painful and expensive learning curve. We can provide insights that you might not have thought of and take over the implementation of the entire optimisation process for you, ensuring that every pound you spend is working as hard as possible to grow your business.

We often find that getting an expert second opinion is incredibly valuable. We offer a free initial consultation where we can take a proper look at your account together on a call, review your strategy, and give you some more specific, tailored advice. It's a great way to get a taste of the expertise you'd be getting if you decided to work with us.

Hope this detailed breakdown helps you get things on the right track!

Regards,

Team @ Lukas Holschuh