Solved: Facebook Ad Shown Outside Targeted Age & Gender

Hi there,

Thanks for reaching out regarding your Facebook ads and your question about age targeting. It's a really common point of confusion, and frankly, it gets right to the heart of how these platforms actually work versus how people *think* they work.

The short answer is yes, what you're seeing is completely normal. The slightly longer answer is that you're looking at the problem from the wrong angle. The solution isn't about forcing Facebook to be more 'accurate' with demographics; it's about understanding the algorithm's real goal and using it to your advantage. I'll walk you through how I'd approach this.

We'll need to look at how the algorithm really thinks...

First off, you need to ditch the idea that any ad platform's targeting is perfectly precise. It's not. Think of the age and gender you input as a strong suggestion, a starting point for the algorithm, not a rigid set of walls it can't go beyond. Unfortunatelly, people lie about their age on their profiles, or they create a profile when they are 20 and never update it for 15 years. The data is messy.

But the main reason you're seeing this 'spillover' is because of your campaign objective, assuming you've set it correctly to something like 'Conversions' or 'Sales'. When you give the algorithm that goal, its #1 priority is not "show this ad to males aged 22-32." Its priority is "find me the people, whoever and wherever they are, who are most likely to convert for the lowest possible cost."

The algorithm has an insane amount of data on user behaviour. It knows who clicks, who buys, who fills out forms. If its data models suggest that a 45-year-old man has a similar behavioural profile to the 28-year-old men who are converting, it's going to show him the ad. Why? Because it's following your primary instruction: get me conversions. It's making a bet that this 45-year-old is a cheap and easy conversion. Sometimes it's right, sometimes it's wrong, but its always learning.

In many campaigns, especially with Advantage+ settings, 'targeting expansion' is automatically enabled. This is Facebook explicitly telling you it reserves the right to go outside your defined audience if it thinks it will get you better results. You're actually paying the world's most powerful advertising machine to find you customers, not to just talk to a demographic you've guessed at. The fact that only 30% of your results are from your core target group isn't necessarily a failure; it might be valuable market feedback telling you that your initial assumption about your customer was wrong or too narrow.

The real question isn't 'Why is my ad showing to older people?'. The real question is 'Are those older people converting, and at what cost?' If they're converting cheaper than your target audience, then great! You've just discovered a new, profitable customer segment for free. If they're not converting at all, then we have a different problem to solve.

I'd say you need to shift your mindset from demographics to behaviour...

This brings me to a much bigger point. The most common mistake I see people make is defining their customer by sterile demographics. "Males, 22-32" tells you almost nothing of value. It leads to generic ads that speak to no one.

You need to stop thinking about your Ideal Customer Profile (ICP) as a demographic and start thinking of it as a *nightmare*. What is the specific, urgent, expensive problem that your product or service solves? That's your customer. The demographic is just a container.

Let's imagine you're selling a B2B SaaS tool that helps developers manage their workflow. Your ICP isn't "a male developer aged 25-35". Your ICP is a Head of Engineering who is terrified her best developers are about to quit because they're frustrated with a broken system. That fear, that 'nightmare', is what you target. It's an emotional state, a problem state. And guess what? A 45-year-old female Head of Engineering could be living that exact same nightmare. By focusing only on the young male demographic, you'd miss her completely.

Once you define the nightmare, you can get much smarter with your targeting. Where do people with this problem hang out online?
-> What niche podcasts do they listen to?
-> What industry newsletters do they actually read?
-> What software tools do they already pay for?
-> What specific Facebook groups are they members of?

This is the intelligence that builds a real targeting strategy. For instance, instead of just "male, 22-32," you could target people interested in 'HubSpot' or 'Salesforce', who are also 'small business owners'. Or for a consumer brand, you might target people who follow specific competitor pages, or magazines related to your niche. This is infinitely more powerful because it's based on actual behaviour and intent, not a guess about someone's age.

Do this work first. Map out the problem you solve in detail. Before you spend another pound on ads, you need to become an expert in your customer's pain.

You probably should focus on what you're actually measuring...

Let's go back to your observation: "Only 30% of my results are coming from the 25-34 range". What do you mean by "results"? Is that clicks? Impressions? Reach? Because if it's not conversions (leads, sales, signups), then the metric is mostly useless.

You need to go into your Ads Manager reporting and customise the columns. You want to see the breakdown by Age and Gender, but you must look at the metrics that matter: CPA (Cost Per Acquisition) and ROAS (Return On Ad Spend). You might discover that while the 25-34 group gets you some conversions, the 45-54 group, while smaller in number, actually has a much better ROAS. That's a goldmine of information.

This is where understanding your business numbers becomes absolutly essential. The real question isn't "How low can I get my cost per click?" but "How high a Cost Per Lead can I afford to acquire a great customer?" The answer to that is your Customer Lifetime Value (LTV).

Let's do a quick, dirty calculation. You need three numbers:

1. Average Revenue Per Account (ARPA): What's a customer worth to you each month? Let's say it's £150.
2. Gross Margin %: What's your profit on that? Let's say it's 70%.
3. Monthly Churn Rate: What percentage of customers do you lose each month? Let's say it's 5%.

The calculation is simple: LTV = (ARPA * Gross Margin %) / Monthly Churn Rate

So, in our example: LTV = (£150 * 0.70) / 0.05 = £105 / 0.05 = £2,100.

Each customer you acquire is worth £2,100 in gross margin over their lifetime. A healthy LTV to CAC (Customer Acquisition Cost) ratio is about 3:1. This means you can afford to spend up to £700 to acquire a single customer and still run a very healthy business. Suddenly, a £50 CPA doesn't seem so bad, does it?

When you know this number, you can analyse your ad performance properly. You can look at the results from the 45-54 age group and say, "Okay, the CPA here is £65, but my LTV is £2,100. This is incredibly profitable. I should create a new campaign specifically targeting this audience." Without the LTV calculation, you're just guessing. You're flying blind, turning off potentially profitable audiences because they don't fit your initial assumption.

You'll need a proper testing structure...

So, how do you put all this together into a campaign that actually works? You need a systematic way to test audiences and find winners. Relying on one ad set with a simple demographic target is a recipe for failure. Here's how I structure campaigns, and you can apply a similar logic.

I think of the funnel in three parts: Top of Funnel (ToFu), Middle of Funnel (MoFu), and Bottom of Funnel (BoFu).

ToFu: Prospecting for New People
This is where you find new customers. Your goal here is to test different audiences to see what works. Instead of just one demographic ad set, you should have multiple ad sets running against each other.
-> Ad Set 1: Interest Stack. Group together 5-10 highly relevant interests based on the 'nightmare' you defined earlier. Target people interested in your competitors, related software, industry leaders, magazines, etc.
-> Ad Set 2: Lookalike Audience (LAL). Once you have enough data (at least 100 purchases, but more is better), you can create a Lookalike Audience. You tell Facebook, "Here is a list of my best customers. Go find me more people who look just like them." This is the most powerful targeting tool on the platform because it uses thousands of data points, not just age and gender. You can test LALs of purchasers, LALs of people who initiated checkout, LALs of your highest-value customers.
-> Ad Set 3: Broad Targeting. This is for when your pixel has a lot of data. You might literally just target a country with no interest or demographic targeting at all, and just let the algorithm do its thing. It sounds crazy, but for mature accounts, it can work spectacularly well.

MoFu/BoFu: Retargeting
This is where you bring back people who have shown interest but haven't bought yet. It's criminal how many advertisers neglect this. These audiences are warm and much more likely to convert. You should have separate ad sets for them.
-> Ad Set 4: Website Visitors. Anyone who visited your site in the last 30 days but didn't buy. Show them a different ad, maybe with a testimonial or overcoming a common objection.
-> Ad Set 5: Add to Cart / Initiated Checkout. Anyone who got *really* close in the last 7-14 days. These are your hottest leads. Show them an ad with a reminder, maybe a small discount or a free shipping offer to get them over the line.

By structuring your account this way, you are no longer just guessing. You are systematically testing, gathering data, and making informed decisions. You can clearly see which ToFu audience is bringing in the cheapest customers, and you can re-invest your budget there. We've run campaigns for eCommerce clients where this structure took them to a 691% return, and for a B2B SaaS client, a medical job matching platform, where it dropped their cost per user acquisition by huge margins. I remember that medical job matching platform, as I just talked about, where we reduced their £100 CPA down to just £7 by applying this kind of rigorous structure.

This is the main advice I have for you:

I've covered a lot of ground here, from high-level strategy to in-the-weeds tactics. It can be a lot to take in, so I've put the main, actionable recommendations into a table for you to use as a checklist.

As you can see, running paid ads effectively in todays world is much more than just picking a few demographics and writing a line of text. It's a full-time job that sits at the intersection of psychology, data analysis, and financial modelling. Getting it right can transform a business. Getting it wrong is just an efficient way to burn cash.

This is where expert help comes in. An experienced eye can spot the opportunities you're missing, build the structures I've outlined, and navigate the complexities to drive down your acquisition costs and scale your results. We've done this for dozens of clients, from generating $115k in course sales in a month and a half to achieving over 1000% ROAS for subscription box companies.

If you're serious about making your ads work and would like an expert to go through your account with you, we offer a free, no-obligation initial consultation. We can take a look at your current setup together and map out a clear, actionable strategy to get you on the right track.

Hope this helps!

Regards,

Team @ Lukas Holschuh