A client once ran three Facebook ad sets targeting "women 25-45 interested in yoga," "women 25-45 interested in fitness," and "women 25-45 interested in wellness" at the same time. All three landed in the same auction, bidding against each other, for the same handful of people. Nobody noticed for two months. That's not a rare story. Most wasted ad spend doesn't come from bad creative or a weak offer. It comes from audience setup mistakes that look harmless on the surface but quietly push costs up while results stay flat.
Here are six of the most common ones, and how to fix each without touching your budget slider.
Mistake 1: Stacking Overlapping Audiences in One Campaign
When you build multiple ad sets that target similar interests, locations, or behaviors, you're not expanding your reach. You're splitting the same pool of people into separate buckets and making them compete against each other in the auction. The platform ends up charging you more to show ads to the same person twice.
This happens most often when advertisers try to "test" audiences by stacking interest categories that already overlap heavily, like "yoga," "pilates," and "meditation" for a wellness brand.
How to check for it: Most ad platforms have a built-in overlap or audience overlap tool. Run it before launching, not after. If two ad sets show high overlap, either merge them or split them by a variable that actually separates the audiences, like age range or purchase stage, not just interest labels.
Mistake 2: Going Too Broad, Too Fast
Broad targeting has a place, especially once you have enough purchase data for the algorithm to find patterns on its own. But broad targeting without that data usually means the platform is guessing, and guessing costs money.
A common pattern: an account with no prior conversion history launches a campaign targeting "everyone in the US, ages 18-65." The algorithm has nothing to learn from, so it spends the first week or two just sampling the population, often at a higher cost per result than a narrower audience would have produced.
In our experience, broad targeting works best after you already have a working narrow campaign generating conversions. Use the narrow campaign to build a dataset, then widen once you see stable results.
A simple sequence that avoids this
- Start with a defined audience based on real signals (past customers, email list, specific interests tied to your product).
- Let it run until you have a meaningful number of conversions, typically a few dozen at minimum.
- Only then test a broader or lookalike audience, and compare cost per result directly against the narrow one.
Mistake 3: Skipping Audience Exclusions
This is the one that costs money silently for months. If you're not excluding people who already converted, already applied, or already purchased, you're paying to show ads to people who can't buy again (or who apply, sign up as a subscriber, and now see the same lead-gen ad every day).
Common exclusions worth setting up:
- Existing customers, if the campaign goal is new customer acquisition.
- People who submitted a lead form in the last 30 to 90 days, depending on your sales cycle.
- Current employees or internal teams (surprisingly common leak, especially in retargeting campaigns).
- Anyone who already completed the on-page goal in the last touch.
Set these up once at the account or campaign level so you don't have to remember them every time you launch something new.
Mistake 4: Letting Interest Targeting Go Stale
Interest-based targeting isn't static. Platforms update how they categorize interests, audience sizes shift, and the people inside an interest category change over time as the platform's own tagging evolves. An interest group that performed well eight months ago may now be mostly cold traffic, or may have quietly ballooned in size and lost precision.
The mistake isn't picking interests. It's picking them once and never revisiting them.
A practical habit: review interest-based ad sets on a set schedule, monthly for high-spend accounts, quarterly for smaller ones. Look at two things: has the audience size changed significantly, and has cost per result drifted upward over the last few weeks without a clear creative cause. If both are true, it's usually time to refresh or replace the interest.
Mistake 5: Using the Same Audience Setup Across Every Platform
An audience built for Facebook doesn't translate directly to Google or LinkedIn, even if the underlying customer is the same person. Each platform categorizes and targets differently. Copying an audience definition word-for-word across platforms usually means you're either too narrow on one platform or too broad on another.
For example, an interest like "small business owners" might be a precise, well-populated category on LinkedIn but a vague, oversized bucket on Facebook. Using the identical setup on both platforms means one of them is almost certainly underperforming, and you may not notice because the other platform is carrying the results.
Instead, define the audience by outcome first (who buys, why, what problem they have) and then translate that into each platform's own targeting options separately. It takes more setup time upfront, but it prevents one platform from quietly eating budget for weak results.
Mistake 6: Never Auditing Automatic or "Advantage+" Style Targeting
Most major platforms now push advertisers toward automated targeting options, broad matching, automatic placements, expansion features that let the algorithm find audiences beyond what you specified. These tools can work well, especially with a strong pixel and conversion history. But "automatic" doesn't mean "unmonitored."
The mistake is turning on automated targeting and treating it as a permanent, hands-off setting. Without regular checks, these features can drift toward audiences that convert on a shallow metric (a click, an add-to-cart) but don't turn into actual revenue.
What to check periodically:
- Compare cost per purchase (or your real bottom-of-funnel goal) for automated targeting versus your manually defined audiences, not just cost per click or cost per lead.
- Look at the breakdown by placement or age/gender if available. Automated targeting sometimes concentrates spend in one segment that looks efficient on the surface but has poor downstream conversion.
- If a manual audience is consistently outperforming the automated one on real revenue metrics, don't be afraid to scale down the automated version, even if the platform recommends otherwise.
A Quick Audit Checklist
Run through this list once a month, or before scaling any campaign's budget:
- Check audience overlap between active ad sets.
- Confirm exclusions are set for converters, existing customers, and recent leads.
- Compare current interest audience size and cost trend against last month.
- Review whether each platform's targeting is genuinely tailored, not copy-pasted.
- Pull cost-per-real-result numbers for any automated or expansion targeting features.
- Kill or pause any ad set that's been running flat or declining for three-plus weeks without a creative refresh.
None of these take more than a few minutes per campaign, and together they usually surface the leaks that are quietly inflating spend without anyone noticing.
Summary: What to Do This Week
- Run an audience overlap report and merge or separate any ad sets with high overlap.
- Add exclusion lists for existing customers and recent converters if you haven't already.
- Set a recurring calendar reminder to review interest targeting size and cost trends monthly.
- Rebuild your audience definitions separately for each platform instead of copying one setup everywhere.
- Pull revenue-based cost comparisons for any automated targeting feature before letting it scale further.
- Document today's baseline numbers so next month's audit has something to compare against.
Fixing targeting mistakes rarely requires a bigger budget. In most cases, it just means spending the budget you already have on fewer, better-defined people.