Happy Sunday, Everyone!
I hope your Q2 is off to a fantastic start. This week’s issue is one that I’ve been working on (intermittently) for the last few weeks in response to a number of frustrations I've heard from CMOs, investors & operators about Meta.
The gist of each of those conversations has gone something like this: the ads are running, the creative seems like it should resonate, the account "should" be performing, but in place of sales, all we’re seeing is more money flying out the door. We’re at the point where we’re beginning to wonder, “Why doesn’t Meta work anymore?”
My response each time I hear it is this: you’re asking the wrong question.
Meta Ads do still work; they just don’t work the way they used to. The days where marketers could fire up a few ads, micro-target a few hyper-niche audiences and print money are long gone. Over the past 5 years, Meta has matured. It has methodically reduced the surplus value generated by advertisers while doubling ad spend (from $84B in 2020 to $164B in 2024). That’s a fancy way of saying: if you want to succeed on Meta today, you can’t run the same old tired playbook and expect sensational results.
Why Your Meta Ads Are Underperforming
Reason #1: Your Offer Isn’t Ready For Prime Time
The rising cost of traffic has made weak offers unsustainable. What worked at $9 CPMs fails at $19 CPMs. This is the inevitable result of supply and demand: in the US, Meta users increased by ~6.4% from 2020 to 2025 while ad revenue doubled. All the while, Meta has made massive investments in their ad products to capitalize on the avalanche of both advertisers and data flooding the platform.
The outcome? Meta has gone from a supplemental discount platform (something added to a media plan to make it “digital”) to a premier performance engine.
Unfortunately, many advertisers have not evolved their approach to Meta to reflect this new reality. I still see far too many accounts running bland, generic offers (save 10% or $10 off) or worse, dominated offers (an offer that is objectively inferior to at least one other available option).
Here’s an anonymized example from an audit we conducted last year:
Let’s not mince words: this is complete obliteration.
Put yourself in the shoes of a member of your target audience who is being served both of these ads on Meta. There’s no contest: the competitor offers are objectively better in each reviewed month. When the economics are this lopsided, it doesn’t matter how glowing your testimonials, how clever your copy or how beautiful your ad creative - you’re not even in the running for the business.
While this example uses a highly commoditized industry (home improvements), the reality is that this occurs far more often than we want to admit - it just takes different forms: (a) feature domination (one option offers staggeringly more features/functionality at a comparable price point); (b) experiential domination (one option offers exponentially better service/support than the others, at the same price point with the same features) and/or (c) perceptual/brand domination (for the same price + features, you can work with a proven brand vs. a risky upstart).
In study after study, the results are the same: the most impactful driver of Meta ads performance is the offer itself (30% to 50% of the total); the ad creative is 5-10%. The remaining 40% to 65% is distributed across the moment, the experience, the lander and the brand perception.
To put a finer point on it: the best Meta ad account in the world will not rescue a bad product or a poor offer from failure. If you want to win, you must have offers that are competitive in the marketplace (and if you’re not sure - then you might want to start with some competitive research, like I discuss here).
Reason #2: You're Conflating Awareness & Performance Creative
The biggest mistake performance marketers make in 2025? Confusing awareness creative for conversion creative. Meta isn’t TV or YouTube. The median IG user scrolls about 303 feet per day - that’s equivalent to a football field. Anyone moving that fast is not going to wait around for 30 seconds while you build intrigue or create tension. If you can’t stop your audience’s scroll in under 2 seconds, it doesn’t matter how fabulous the rest of your ad is, because they’re already onto the next.
Far too many brands I review are still running TV-style ads in their accounts, with branded visuals, moody music, dramatic b-roll and/or overly long founder monologues. There’s nothing inherently wrong with any of that - it’s wonderful for storytelling and TV - but using it for performance on Meta is akin to bringing a bicycle to a Formula 1 race: we all know how it’s going to end.
The ads that do perform on Meta tend to exhibit these five characteristics:
- Quick, Immediate Hook - stops the scroll + creates the opportunity for the rest of the ad.
- Real & Authentic - not overly polished or salary. Great ads don’t feel like ads.
- Breaks The Pattern - most ads tend to fall into the same mold; the ones that are truly effective don’t adhere to marketing “best practices”
- Makes Your Target Audience The Hero - as a brand, your #1 priority should be to help your audience solve their problem
- Constantly Closing - great ads continually nudge your audience toward a specific, defined outcome, not some vague nonsense like “learn more”
You’ll notice that none of these five points say anything about the creative format (video, static, carousel) or type (UGC, comparison, testimonial, unboxing, virtual sales letter, etc.) – that’s because every one of those can be effective if it is executed properly.
Most brands fail here by trying to do too much or too little. Either the ad tries to be clever and ends up confusing, or it’s so generic it could apply to any competitor. Even worse are the brands that try to use TV cuts on Meta (seriously, why is this still a thing?)
If you want your creative to perform on Meta, design it for Meta.
Reason #3: Your Creative Velocity Sucks
Creative velocity is the primary driver of creative value, and creative value is the single-most-impactful lever you have to move the needle on Meta.
This sounds simple enough, but what most brands miss is that it’s not just about how much creative you produce, it’s about how fast you can test it, how well you can measure performance, and how quickly you can scale what works.
Most brands are flying blind. They still rely on gut instinct, isolated performance anecdotes (the sales team said this ad resulted in a whale deal or a customer reached out to an exec raving about that ad), or recycled “top ads” from last quarter. Worse, they obsess over vanity metrics like thumb stop rate & ROAS without understanding how those metrics connect to the bigger picture question: is the account (and the brand) performing?
To succeed on Meta today, you need a system to assess your creative performance like an investor would assess his/her portfolio - dispassionate, grounded in data and focused on the thing that matters (bottom-line results), not statistical significance.
When I review ad accounts, here’s what I’m looking for:
- Cost Per Creative (CPCr): How much does it cost to create a single piece of creative? Every dollar that goes to creative development is a non-working dollar that isn’t available for anything else, but must be recovered via ad performance. If an ad costs $10,000 to make, and your brand has $20 in contribution margin on every sale, the first 500 sales go to paying back the cost of the creative before you have any profit available.
- Hit Rate: The percentage of creatives that meet or exceed your performance threshold (for us, it’s 25x CPA in spend with a CPA >= your target in a 30-day period). Hits are the straw that stirs the drink – if you don’t have a way to manufacture them with regularity, the risk profile of your account goes through the roof.
- Creative Velocity: The number of net-new creatives launched per unit time. Brands that introduce more creative will compound performance exponentially faster than those that do not. This is simple math.
- Payback Period: How many sales/leads are required for your creative to “pay back” its cost of creation? If the CPCr is $1,000, and your contribution margin from a sale is $100, then the Payback Period = $1,000/$100 = 10 sales. This metric is one of my favorites because it forces you to think about both the cost per creative AND the contribution margin generated by the advertised bundle/product/service, while giving you a way to normalize investment decisions across different SKUs/offers.
- Time to Fatigue: How long a “hit” creative performs before declining / hitting diminishing returns. If most of your hits have a high time to fatigue, you’ll be able to get more mileage out of them vs. a flash-in-pan ad (i.e. something that newsjacks or is trendy).
This approach reframes creative testing as a capital allocation exercise. You’re not just creating for the sake of content; you’re investing resources to uncover assets with a positive expected performance.
The implications of that statement are incredibly difficult for most marketers to swallow, because doing so entails decoupling the cost (time, money, energy, blood, sweat, tears) of a specific creative from the expected future performance of that asset.
The rubber meets the road on that when someone (usually an executive) screams about how much money was spent on the brand video and demands to know why that asset can’t be used going forward. The answer should be: that asset has a negative expected future value. The answer usually is: well we can try again to make it work. Stop it. Don’t double down on losers; cut your losses and invest your resources where they have the highest probability of generating returns in excess of your target.
From a logistics/operational standpoint, executing against this requires a structured cadence. Set creative testing goals each week. Measure performance using the metrics above. Kill underperformers quickly. Scale & iterate on winners using AI tools (ChatGPT is your friend, not your enemy). But, most importantly: treat your creative strategy like an evolving portfolio, not a static campaign.
As you implement this, remember that your internal team is a goldmine. Founders, product managers, designers, and support reps can (and should) become content creators. In a world flooded with polished, impersonal ads, a shaky iPhone video from the founder explaining “why we built this” might be the best performer in your account. The era of founders and executives outsourcing/delegating communication is over.
Reason #4: Poor Account Structure & Day Trader Management
Meta’s ad algorithm thrives on data, volume, and signal; it falters when accounts are muddled, over-segmented and constantly changing.
Many underperforming accounts suffer from antiquated structures: too many campaigns, too many overlapping ad sets, too little budget and too much micromanagement. Every time you pause, restart, duplicate or otherwise change a campaign, you disrupt the learning process. Far too many accounts I review have both way too much going on AND slews of daily, non-essential changes being made, rather than once/twice weekly bulk updates. The end result? An account that is never permitted to reach its full potential, simply due to confused management.
Over the course of evaluating 1,000s of Meta Campaigns, spanning 100s of accounts in just about every imaginable industry, there are 8 common campaign structure issues, pitfalls and blunders that tend to surface in poor-performing accounts:
- Too Many Campaigns. You don’t need 17 “micro-campaigns” to look sophisticated. Unless you’re running distinct offers to distinct audiences, with custom creative for each, more campaigns just means diluted data and slower optimization. If you’re going to build a campaign, build it with purpose, and fund it accordingly. Otherwise, take your money outside and light it on fire – at least that’ll keep you warm.
- No Testing Strategy. This isn’t just about A/B testing. It’s about having a real, repeatable test-learn-evolve system that applies to everything: hooks, formats, angles, CTAs, landing pages, audiences, offers. Testing isn’t something you do once—it’s how you operate.
- Optimizing for Bullshit. Engagements, link clicks, video views... none of that pays the bills. Meta will optimize toward exactly what you ask for. If you tell it to get you clicks, it’ll find click-happy browsers. Want buyers? Optimize for purchases. It’s not complicated. But it does require clarity.
- Trying to Outsmart Meta. If you’re still trying to “hack” the algorithm with some ancient TOF/MOF/BOF nesting doll or psychographic witchcraft, stop. You’re not going to gimmick Meta into submission, or find some hidden alpha that the millions of other advertisers have missed (and even if you do, I guarantee someone else will find it too - then what?). The real cheat code? Get brilliant at the basics. Offer. Targeting. Creative. Experience. Spend your time nailing those things, not chasing ghosts.
- Not Whitelisting. If you’re using influencers or creators and not running ads through their handles, this is your invitation to start. Whitelisting unlocks new audiences, allows you to leverage the trust of familiar faces (i.e., the creators they’re already following), and immediately boosts your relevance. It works.
- Lack of Proper Exclusions. Your prospecting campaigns shouldn’t be targeting people who already bought, people who live in places where you don’t (or can’t) operate, or people you’d never want as customers. While the tools you have to exclude all of these people are imperfect, they’re better than nothing – so use them.
- Manage The Comments Section. Social proof matters. Leaving negative comments visible, letting trolls dominate your ad thread, or ignoring actual prospects asking questions is lazy and expensive. Manage your comments. Reuse Post IDs. Treat every ad like a public storefront.
- Day Trader Management: I know most marketers fancy themselves as “traders” in the arena, here’s the reality: all that does is waste money. Meta is a cargo ship. You cannot operate it like a speedboat. Bucket your changes, allow the algorithm to adjust and stabilize, then push the next set. As long as you’re headed in the right direction, you’re good to go.
- Forgetting the Fundamentals. While there are all manner of sins in this bucket, the big ones tend to be: (1) not having the right data in the account; (2) over-segmentation; (3) an ad hoc account structure with no rhyme or reason behind it; (4) running DPAs without (at least) price bucketing – preferably bucketing based on contribution margin and (5) not aligning creative – offer – audience in each ad set / campaign.
Fundamentally, your account structure is a value statement. It is how you communicate to Meta what is important to your brand. A clean, stable structure allows Meta to understand what you’re trying to achieve (sales, leads, demos, whatever), and lays the foundation on which your targeting and creative will operate. It reduces volatility and increases Meta’s ability to identify high-probability converters. Stability is a competitive advantage in an ecosystem where most brands are in constant reset.
Reason #5: Ignoring The Post-Click Experience
Most marketers are obsessed with what happens before the click - structure, bid strategy, audience and creative - to the detriment of what happens after it. That may sound harsh, but it’s true: most brands/marketers are not ruthlessly optimizing what happens after the click. That’s a problem, because no matter how good your ad is, if the experience it leads to is broken, you're just paying to disappoint (at best) or alienate (at worst) your audience.
Here’s the truth: after the offer, the post-click experience is often the single most powerful driver of poor performance. If your ads aren’t performing, the best place to start identifying the issue is the landing page, followed by the form, checkout process, thank-you page, data passback (i.e. form-to-CRM connectivity or checkout-to-CDP) and email/SMS experience. Odds are, something is broken on at least one of those.
Some immediate red flags:
- Greater than 2.5s load time
- Poor page experience
- No above-the-fold CTA with a clear directive + value statement
- Mis-alignment between ad, offer + LP
- Overly long/complex form
- Convoluted checkout (this really shouldn’t happen with Shopify, but it does)
- Out-of-stock product or unavailable service
Fixing this means treating your landing page experience like a product, not an afterthought. Ad-to-landing-page continuity isn’t optional. It’s performance-critical. Same hook. Same message. Same urgency. It should feel like one seamless narrative from scroll to checkout. If you’ve got a killer ad driving to a mediocre page, you’re not underperforming because Meta “isn’t working.” You’re underperforming because the experience you’re delivering is terrible.
Improving Your Meta Ads Performance
Upgrade #1: Improve Your Offer
Your offer is the single-most-impactful thing in your ad account. Full stop. If the offer isn’t compelling, nothing else matters. Strong offers aren’t just economically compelling; they’re psychological motivators that create urgency, increase trust, and provide an impetus to act.
If you’re struggling to see performance from your Meta Ads, resist the urge to pause/pull your budgets. Instead, ask yourself this one question: “After seeing this ad/offer, why should a member of my audience interact (buy, submit a lead form, schedule a demo, subscribe) with my brand right now?”
The honest answer to that question is usually a painful one to accept: there’s no reason. That’s a problem. But the first step to solving a problem is recognizing that one exists.
Once you have that, it’s time to re-think your entire offer. Start with discounts. I genuinely don’t know who did the math on giving away 10 points of margin for an email, but I’m immensely confident that it doesn’t pencil as well today as it did way back when. Discounting is a race to the bottom, which is not a game most of us ever want to play (and one that no one ever wins).
Instead, consider bundling products or services to raise AOV. Create perceived exclusivity. Introduce non-traditional (and harder-to-quantify) incentives, like free gift with purchase, charitable donations, or exclusive access. Position the offer in the context of what matters most to your customers, whether it be saving time, reducing effort, getting a better outcome. Once you’ve dialed in the offer, communicate it clearly in the ad, the landing page and the follow-on experience. avoid hidden exceptions, fine print, and bait-and-switch.
Start thinking like a direct response copywriter, not a brand marketer. Translate value into action. Price is what you charge. Value is what they perceive. The offer is the bridge between the two.
Upgrade #2: Increase Creative Velocity (Again)
This is where most brands fall behind. They know creative matters, but they treat it as a campaign instead of a pipeline. If you're not generating 10–15 net-new creative concepts each week, you're going to fall behind.
Velocity isn’t just about quantity. It’s about consistency and structure. Set weekly creative testing targets. Create a backlog of ad ideas across different angles: problem-solution, feature-benefit, testimonial, demo, urgency, and lifestyle.
This is an area where AI can be a game-changer: feed it the relevant details on your brand, upload your top-performing creatives and allow it to generate ideas for you into a sheet. Review the sheet, indicate which ones are good/bad, and feed it back to the AI to refine even further. Once you’re satisfied with the concepts, task ChatGPT or Gemini to produce them. In a fraction of the time it would take you to do all of this, you’ll have a library of assets ready for testing.
As you introduce these new creatives into the account, critically evaluate performance:
- Hook rate is a strong indicator of hook strength.
- Total watch time or watch time to 25%, 50%, 75% and 90% (for videos) is a decent proxy for the hold strength.
- Unique click rate tells you if interest is translating into action.
- Landing page drop rate tells you if you’re attracting window shoppers or serious buyers.
- Conversion rate is a strong indicator of how effectively your landing page is monetizing the traffic generated from your campaigns.
Learn from each of those metrics – an ad might perform terribly from a conversion rate perspective, but have a high hook rate. Instead of discarding the whole creative, try integrating the same hook (or close variants of it) into other creatives with better conversion performance.
Upgrade #3: Use LALs + Interest Stacks
For much of the last 3 years, broad targeting was the single-most-common setup on Meta - and for good reason: broad delivered some eye-popping results early on, which were amplified and leveraged by Meta to compel more advertisers to use it.
Now, the broad-only party is over. The performance advantage of 'just let Meta figure it out' has been eroded by saturation, skyrocketing competition, and diminishing marginal returns. The algorithm has picked all the low-hanging fruit (and most of the mid-hanging fruit, too). Broad is now the default for the vast majority of accounts, which means the surplus value generated by using it is approaching zero.
That’s why high-performing brands are moving back to signal-rich targeting: not because Meta can’t optimize on its own, but because it optimizes faster, more efficiently, and more profitably when it’s fed smarter inputs. Smart marketers + smart machines = magical results.
The best analogy I can give for this is condensing Meta’s search space. You’re not restricting Meta; you’re reducing the complexity of the problem you’re asking it to solve by introducing data that has incremental value (i.e. your existing customer lists, your in-depth knowledge of your target audience, etc.). That gives your account a real, sustainable performance edge.
Start with leveraging your existing customer data. Take your full customer list, segment it based on performance, and then upload those segmented lists for use as seed audiences:
- Build Lookalikes from your highest LTV customers.
- Separate one-time purchasers from loyalists.
- Identify lists of single-line or single-SKU buyers
- Create lists of demo bookers who have never purchased
Each of these will behave—and perform—differently. The overlap between a 1% Lookalike from “all leads” vs. one from “engaged demo attenders who never bought” is smaller than most people think. Segmenting your customer database is what unlocks scale without waste.
Once you have a few lookalikes rolling, create a few ad sets with interest stack audiences. Contrary to popular belief, interest stacks can (and do) still work – they just have to be created from real, well-done customer insights work (you know, use Sparktoro). If you can identify the right combinations of interest stack elements (check out this article for more), the outcome will be giving Meta’s algorithm fewer degrees of freedom, which makes it easier for it to converge on the right audience.
Think of using these tools as akin to fishing with sonar vs. tossing dynamite and hoping something floats up. Given enough time (and enough dynamite), the latter approach will work. The former approach just gets you a delicious dinner much faster.
At its core, this approach isn’t about rejecting automation. It’s about collaborating with it. You give Meta your best inputs - segmented LALs, well-researched interest stacks, relevant cost caps - and let the algorithm do what it does best: identify patterns, allocate budget, and scale winners.
Signal-rich inputs produce profitable outputs.
Upgrade #4: Improve Your Landing Pages
The job of the creative is to earn the click; the job of the landing page is to earn the sale.
Landing pages are the single biggest bottleneck in most Meta ad accounts. Most are either too slow, too confusing, too generic or too disconnected from the ad that preceded them. A great landing page is like an infinitely scalable salesperson: it welcomes the users generated by your ads to your brand, converts curiosity into genuine interest, guides them through the remainder of the buying journey, overcomes their objections and closes the deal. That’s a tall task - and most pages aren’t up for it.
The fastest way to improve performance isn’t always changing your ads; it’s often fixing the experience that occurs after the click. Here’s a few ways you can do that:
- Align Offer, Creative & Lander - this is the single-easiest win for most brands: make sure your lander is aligned to the preceding touchpoints in the experience. There’s nothing that undermines trust and kills conversion quite like clicking on an ad, only to be deposited on a page that bears little resemblance to the thing I just saw/watched/clicked.
- Experiment with Different Landers: not all landers need to be product pages or shop pages or services pages; in fact, most probably shouldn’t be any of those things. Instead, experiment with radically different page types - listicles, quizzes, faux homepages, advertorials, long-form pages, comparisons, testimonial pitches, influencer pages, etc.
- Build Trust Early & Often: the first priority of any lander should be to establish trust with your visitor. The first step in that process is identifying which sources your target audience is predisposed to believe (after all, I likely trust different news sources and organizations that you do), then feature those.
- Weed Out The Pretenders: there’s a common misconception in marketing that landing pages should have low bounce rates. I disagree. One of the primary roles of a great landing page is to quickly weed out the pretenders/window shoppers. Your goal should be to convert as many of the right people as possible, which involves intentionally creating friction points for those that don’t fit that definition.
- Emphasize Benefits, Not Features: Great landers focus on showing how your product/service/widget makes your target audience’s life better. There are two components to this: highlighting the benefit and showing, not telling. Most brands fall into trap of bragging about their features (which no-one cares about) and telling you how great they do something (vs. just illustrating it). Avoid those traps. My favorite example of this is the (unintentional) Stanley ad, where the person pulls a Stanley out of a burning car and dumps the contents, to reveal there’s still ice inside. Sure, Stanley could say that we keep your drink cold on the hottest summer days – but then again, so could anyone else. What sets the Stanley example apart is that they showed it (and an extreme example at that).
If you’re curious how to do that, check out this article on Landing Pages (and why they’re probably holding back your ad accounts).
Upgrade #5: Measure What Matters
Direct, person-to-sale attribution was always a fugazi. It was an illusion. It worked fantastically well for selling digital advertising, but in the process, it trained an entire generation of CMOs (and executives) to think that something as complex and dynamic as marketing could be distilled into a linear, mechanical process.
My advice: Stop chasing granular precision. Start tracking progress. The best marketers + operators I know don’t ask, “What specific ad drove this sale?” They ask, “Is our marketing driving incremental profit?”
The reality is that most users will interact with your brand multiple times before purchasing – some of them you’ll know about, some you won’t. Some will be public (like on Meta), while others will be private (WhatsApp groups, Slack, etc.). If you try to track and make sense of all that, you’ll end up driving yourself insane (and you’ll be no wiser for it).
Instead, focus on your big picture metrics: is your cost/acquisition acceptable? Are you reducing your cost/creative while increasing hit rate? What is the trend on your landing page conversion rate? Get the direction right, then optimize as you go.
At the end of the day, your bank account is the single-best source of truth you have in marketing - so trust that above all else.
Meta Isn’t Dead. Most Marketers Just Aren’t Disciplined Enough To Make It Work.
Meta is still one of the most efficient, scalable, and profitable ad platforms in the world; it just doesn’t reward mediocrity anymore.
Winning in 2025 means:
- Relentless creative iteration
- Strategic targeting driven by genuine audience understanding
- Offer-led campaign design
- Clean, high-converting landers
- Manage everything based on what actually matters (people + profit)
It’s not a difficult playbook - but it’s one that relatively few marketers have the courage to implement.
Until next week,
Sam
Loving The Digital Download?
Share this Newsletter with a friend by visiting my public feed.
Follow Me on my Socials