Tips to Improve Ad Creative Performance

I spent $65,000,000+ on Meta ads so you don’t have to. Over the last few years, I’ve worked with dozens of DTC brands in beauty, skincare, fashion, CPG, and home goods. I’ve seen what works (and what burns money) across every stage of growth. Here are 10 hard-earned lessons from the trenches: 1. Creative strategy beats media buying. The algorithm already knows who to show your ad to. What separates great brands is creative: strong angles, messaging, and hooks that actually convert. 2. Creative is targeting. You don’t need 17 ad sets and crazy custom audiences. One powerful ad that speaks directly to your customer beats everything. 3. Static native ads are king. Especially for MOF and BOF. These don’t scream “I’m an ad”—and that’s why they work. 4. If you’re using video, your first 3 seconds matter most. People scroll fast. If the hook doesn’t hit immediately, they’re gone. Write for the scroll, not the script. 5. Start with a scalable hero product. High margin, low return rate, strong retention. If your product can’t scale profitably, your ad strategy won’t either. 6. Know your customer inside-out. Go beyond demographics. Read reviews, Reddit threads, TikTok comments. Mirror how they talk about their pain points. 7. Build an offer that’s worth sharing. Bundles, bonuses, exclusives—these boost AOV and give people a reason to act now. 8. Campaign structure matters. Use a consolidated setup: Advantage+ for broad reach, CBO for testing, and a clean retargeting stack for conversions. 9. Winning creatives come from a repeatable system. Map out angles, hooks, variations. Test weekly. Refresh often. The most creative brands have the most systems. 10. Track real performance metrics. CTR and ROAS are surface-level. Blended CAC, LTV, and contribution margin tell the real story. Scaling on Meta isn’t about hacks. It’s about strategy, consistency, and iteration.

Ever felt your creative pipeline is busy but not moving fast enough? One big reason: no one defines the buckets. Team mix new shoots, edits, influencer collabs, and UGC into one pile. Result: confusion on priorities, slower reviews, delayed launches. What works better is defining 𝗿𝗲𝗰𝘂𝗿𝗿𝗶𝗻𝗴 𝗯𝘂𝗰𝗸𝗲𝘁𝘀. Some buckets can be universal across D2C brands: ▪️ 𝗣𝗿𝗼𝗱𝘂𝗰𝘁-𝗳𝗼𝗰𝘂𝘀𝗲𝗱 𝘀𝘁𝗮𝘁𝗶𝗰 (clean, no talking) ▪️ 𝗘𝘅𝗽𝗹𝗮𝗶𝗻𝗲𝗿/𝘁𝗮𝗹𝗸𝗶𝗻𝗴 𝗮𝗱𝘀 (problem-solution, testimonials) ▪️ 𝗥𝗲-𝗲𝗱𝗶𝘁𝘀 𝗼𝗳 𝗽𝗮𝘀𝘁 𝘄𝗶𝗻𝗻𝗲𝗿𝘀 (shorten, add new hooks) ▪️ 𝗜𝗻𝗳𝗹𝘂𝗲𝗻𝗰𝗲𝗿/𝗨𝗚𝗖 𝗽𝘂𝗹𝗹𝘀 (organic turned paid-ready) And some should be brand-specific, depending on what’s working: For a yoga apparel brand: community-led content (tagged reels, authentic photos) For a beauty brand: before-after edits For a food brand: recipe/demo hooks The clarity comes when the team knows this week’s 4 ads are “X new shoots + Y re-edits + Z UGC.” No ambiguity. No bottlenecks. Buckets don’t just organize work. They create speed. 🏋️ Action Build recurring creative buckets based on the last 90 days performance. ▪️ Look back: Identify which ad types actually delivered profitable scale. ▪️ Create “working” vs “not working” lists. ▪️ Working: Turn into recurring buckets. Bake them into your weekly creative plan (e.g., statics + re-edits + testimonials if those are winning). ▪️ Not working: experimental buckets: Keep them small, treat them as test pilots, not core pipeline. #PerformanceMarketing #MetaAds #FbAds #CreativeStrategy

Stop trying to make people feel happy. Happiness is the most boring emotion in ads. It blends into the feed. It converts poorly for high-value offers. Want better performance? Use emotion like a scalpel, not a confetti cannon. Here’s a smarter playbook for emotions in creative: 1️⃣ Pick the micro-emotion, not the macro label ↳ Replace "happiness" with precise targets: curiosity, frustration, relief, pride, envy, nostalgia. Micro-emotions hit specific neural shortcuts that drive action. 2️⃣ Match emotion to funnel stage ↳ Top: surprise or curiosity to stop the scroll. Mid: frustration or aspiration to deepen intent. Bottom: relief, urgency, or security to close. Retention: pride and belonging to raise LTV. 3️⃣ Sequence emotions across your creative library - emotional cadence ↳ Don’t show 10 happy ads. Feed the algorithm a journey: intrigue → empathy → assurance. The system learns sequences better than isolated hits. 4️⃣ Use emotion to pre-qualify ↳ Angry, urgent ads attract deal chasers. Calm, authoritative ads attract high-LTV buyers. Your tone is the sieve that determines customer lifetime value. 5️⃣ Test with purpose ↳ Launch 4-6 emotional vectors, same format. Measure CTR and 30/90 day LTV, not just CPA. Iterate on what emotionally resonates, not what looks pretty. Controversial but true: emotional diversity beats brand sameness for performance. If your ads all feel the same, the algorithm and humans will ignore them. Found this useful? Like, follow, and repost ♻️ so others can too! ps. struggling with creative bottlenecks? We can help.

You don’t need 17 campaigns, 10 audiences, or complex structures to scale on Meta. Nick Theriot manages $1M+ per month in ad spend with one of the cleanest setups I’ve ever seen. On Ep 2 of Ad Spend, he walked me through his 7-part system for scaling brands and creating “purple cow” ad creative that actually stops the scroll. Nick isn’t a theory guy. He’s in the accounts, operating. His agency has driven $100M+ in revenue for clients. Here’s what we covered:  ✅ CBO vs ABO  ✅ Scaling creative testing  ✅ Campaign cleanup  ✅ Building Purple Cow ads Nick’s top insights 👇 1. Keep your account structure clean.  Nick runs 1 CBO campaign per product type, per country, per gender. No separate testing and scaling campaigns. No confusion. Just clarity. 2. Don’t overthink exclusions. He tested excluding site visitors and past purchasers, but found Meta spent the same way. That’s why he focuses on creating ads that speak to bigger audiences. Your creative is your targeting lever. 3. Stick to a creative cadence. Nick launches 3–6 new ad sets every week. Always on the same day. Each ad set = 1 concept based on a single hook. Consistency gives your pipeline clarity. 4. Never move “winners” to a main ad set.  If it’s working inside your test campaign, then leave it. Scale it in place. Meta doesn’t care about your naming conventions and account structure. Just let it cook. 5. Don’t just rely on creative quantity. Nick defines a quality ad test as “a specific angle, targeting a real pain point, with messaging no one else is using.” The better your customer research, the better quality ads you make, the fewer you have to churn out. 6. Purple cows win.  Want to really stand out? Try one of these:  → Beat your competition on ad production quality  → Go after a sub-identity they’re ignoring  → Make a bigger (but true) claim 7. Don’t just optimize your ad creative Too many brands stop at the ad. Nick pushes further: test a new landing page. The creative got you the click. Now improve conversion. My biggest takeaways from Nick:  → Clean structure scales better  → Creative diversity fuels Meta’s algorithm  → Quality ads come from knowing your customer deeply  → Winning ads are just the beginning.

I speak to 40+ DTC brand owners a month. 87% iterate on their creative the wrong way: Most of the brands I speak to waste time iterating on losing ads. They take a crappy ad, slap a new hook on it, and expect it to be transformed into their next unicorn winner. Here's why that doesn't work: There are two types of Facebook ad metrics. 1. Hard metrics → CPA, ROAS etc. 2. Soft metrics → Hook Rate, Hold Rate, CTR etc. Hard metrics are the ones that matter. They tell you if an ad is working or not, and hence if it is worth iterating on. Hard metrics help you differentiate the winners from the losers. If an ad has bad hard metrics, it's not worth iterating on. Soft metrics should only be used as a guide to help improve ads that already have strong hard metrics. As changing the hook on a crappy ad far below target CPA, will still leave you with a crappy ad far below target CPA. So if all of your ad creatives are bad, instead of creating iterations on bad ads do this instead: → Export all customer reviews from the last 6 months → Feed this data to an AI creative strategist → Analyze the data to uncover the 5 most common pain points → Use AI to ideate 3 new creative ideas for each pain point → Use AI to order these creative ideas in descending order of strength → Make 15 new ads starting at the top of the list and work your way down → Create 5 variations of each ad which will give you 75 ads in total → Create a testing campaign with an ad set for each pain point → Test with a budget suitable for your brand's AOV → Run for 5 - 7 days You will be left with some winning ads and some losing ads based on your target CPA/ROAS. 1. Iterate on the winning ads 2. Double down on the angles in your winning ads in your next creative round 3. Take learnings from the losing ads 4. Do more of what works, stop doing what doesn't work Simple, but not easy.

If your Facebook ads are too expensive… It’s probably not your budget. It’s not your niche. It’s not even your product. It’s your structure. Here’s how we took a client from $115 cost per purchase down to $55 while scaling spend above $2K/day: 1. Kill campaign clutter. The client had 25+ campaigns running. Lookalikes. Interests. Bid caps. Cost caps. Retargeting. It was chaos and none of it was working. We replaced all of it with one campaign. → One product → One country → One broad ad set → One clear objective Simplicity scaled. Complexity killed performance. 2. Use existing proven creatives. We launched 6 ads using existing post IDs from past top spenders. No new angles. No guessing. Just riding proven momentum already in the account. No one talks about this but it works extremely well when first taking over a messy ad account. 3. Rebuild your testing infrastructure. We rebuilt the creative system from scratch: • Defined 6–12 ad concepts • Each concept = 1 ad set • Each ad set = 3 variations (split by 1 visual element only) Video example: same script, same hook, different first 3 seconds. Image example: same copy, different background. We ran 3–4 concepts per week to maintain velocity. 4. Creative strategy starts with research. Most people write angles based on vibes. We write angles based on data: → Reddit threads → TikTok comments → Competitor reviews → Customer feedback → Site copy + heatmaps → Even YouTube comment sections We build angles around what people actually say, not what we think sounds good. 5. Once you win, scale with control. We found a $55 CPP winner. Then scaled 20% a day. All with structured budget increases with proven creatives. Now we’re pushing toward $1M/month while staying below the $70 CAC target the client hired us to hit. The biggest mistake most brands make? They throw more spend at broken foundations. We simplified the structure, fixed the strategy, and let the creative carry the weight. If your CAC is too high, don’t scale harder. Scale smarter. Follow me Nick Theriot for more content like this.

Struggling to make your Facebook Ads more effective? Here are 7 proven tactics from demand generation pros that could transform your ad campaigns:   1. Don’t fear “ugly” ads: → Ads native to the platform often perform better than polished, glossy ones. → Authenticity beats perfection. 2. Feature people, not just products: → Personal connections drive engagement. → Ads that feature real people, like team members or customers, tend to build more trust & boost conversion rates. 3. Leverage your data: → Focus on collecting & analyzing first-party data. → Use Facebook lead forms to gather insights. → Boost your ad performance by targeting the right audience effectively. 4. Dynamic ads for better results: → Use Facebook’s Dynamic Creative Optimization (DCO) to automate and test multiple versions of your ads. → Facebook’s AI will determine the most effective combinations to drive higher engagement & conversions. 5. Engagement-purchase-scale method: → Warm up your audience with engaging content, then move to targeted purchase ads. → Once you find what works, scale it by building lookalike audiences for exponential growth. 6. Trust the algorithms: → Meta’s algorithms are highly effective—let them do the heavy lifting. → Focus on testing creatives and messaging, not overcomplicating audience targeting. 7. Use the conversion setting for lead ads: → For higher-quality leads, direct users to fill out forms on your website rather than Facebook. → This method filters out spam leads and helps drive more serious inquiries. Focus on making your ads feel organic, trust the algorithms, and stay data-savvy to boost your ROI. Apply these tips to your next campaign and see the difference!

After overseeing $150M+ in ad spend for 8-figure finance, lead gen, app & omnichannel brands, these are the 5 pillars I came up with that go into every high-performing direct response ad. 1/ Alignment to your target audience Making sure the creative is aligned to your target audience is vital. This one obviously doesn’t need much explaining, but it’s important to note since a lot of brands seem to lose sight of this. 2/ Clear brand/product presence Your brand and product presence needs to be crystal clear in the ad. If it’s hard to tell what product/service you’re actually selling, chances are you won’t be getting the right people in the door. 3/ Value proposition is clear This answers the question of: “Why should I care about watching more of your ad/clicking through?” If the value you’re providing/problem you’re solving isn’t clear to the user, they have no reason to click or be interested. They could be the perfect, 10/10 ideal client who’s super problem aware… But if they aren’t even aware that you can help them with that problem, they won’t convert. 4/ Clear CTA This is a direct response ad. So that said, we want to make sure that the user knows you want them to click through and take some sort of next step. Whether it’s to book a call, buy now, or enter info for a free lead magnet… What their next step is, shouldn’t be up for debate. Make it nice and easy for those who are interested. 5/ Mixing of formats You can’t just be running videos. You can’t just be running images. You need to be mixing it up, running ALL formats. Reason why, is because Meta delivers ads to users with preference in mind. For example, if person A prefers to receive video ads, they’re going to end up getting more video ads. If person B prefers image ads, they’re going to get more of those. If you only run one type, you’re going to be neglecting an entire subset of your audience. Overall, those are the 5 most important principles that all high performance direct response ads have in common. Did I miss any? Let me know in the comments!

I have generated $400M with Meta ads. Most of these results came from a rock solid understanding of video creatives. To date, we’ve produced 1,000s of them that’ve driven millions for clients. Here’s 5 components that make up a winning video creative: COMPONENT 1: Write A+ hooks Poor quality hooks = throwing money down the drain. It’s that simple. Viewers aren’t going to stick around & watch the rest of your ad if your hook is inadequate. Your hooks should either:  - Lead with a benefit  - Agitate a pain point  - Call out the ICP & do one of the above Keep them simple. Being too clever usually backfires. COMPONENT 2: Show the product early Your product is the crown jewel of the video ad. If it’s a product that viewers genuinely want because it solves a pain point? Or using it is beneficial to the viewer? Sometimes that’s all it takes to hook & hold their attention. Burying the product too deep in the ad = it might be missed entirely. Rookie mistake. COMPONENT 3: Use the “heartbeat method” Heartbeat method = quick edits every 1-3 seconds so the viewer doesn’t get bored. The truth is that attention-spans are at an all-time low. So using retention techniques to keep your viewers engaged is essential. The change could be:  - Zoom-ins  - Pop ups  - New visuals New & novel creative elements that help to hold their attention. COMPONENT 4: Make the “problem” you’re calling relatable This one goes back to your market & customer research. Having a clearly defined list of problems makes it easy to agitate them in ads. The goal with the ad should be to amplify the pains that problem brings. Almost bringing those pains “to life” to capture your viewers’ attention. Agitate the problem using both the ad’s visuals & copywriting. Good research x relevant visuals x copywriting = a winning formula to convey pain. COMPONENT 5: Put your solution on a pedestal You’ve just gone *deep* into a big problem that your viewers face. You’ve put them in an emotional state - now it’s time to present a solution. You want to present the product like it’s the only solution on the planet… Aesthetic shots. Eye-grabbing visuals & imagery. Text that outlines key benefits & showing how it alleviates the pains you pointed out. This is how you balance agitating pains & presenting a solution to those pains. The biggest takeaway here is that a LOT of moving parts go into winning video creatives. There’s even more I could discuss. The goal being to make each component winners in their own right. So when they do come together? It leads to a *complete* video creative & outperforms other ads on the newsfeed. Video in paid ads is only becoming more integral as we approach 2025. My advice is to get a good grasp of the fundamentals sooner rather than later. You don’t want to be playing catch-up down the road!

Creative is the new targeting, but how do you execute on that? Every decision you make with regard to ad creative matters for who you reach: - Creator selection - style, age, interests, etc. - Music selection - genre, pace, etc. - Copy - subtle framing/phrasing, value props - Context - use cases, setting - Testimonials - from who, and what they say - And more... To focus efforts, we ALWAYS recommend starting with consumer research. Here is an example from a recent client project for a wellness brand: - Our 30-45 demo is 34% MORE likely to trust friends & family for advice, and 10% more likely to trust social media - Our 50+ demo is 32% more likely to ask a doctor for advice and 49% more likely to ask a pharmacist for advice So what does this mean for our creative strategy? To truly reach these people and drive conversion, we need to build trust through different avenues. We should (generally) not be showing a pharmacist to our 30-45 demographic and we should not be showing a non-credentialed/non-medical professional creator to our 50+ demo. The data proves this. - In ads mentioning a doctor's recommendation and leveraging an older creator, we see our younger demo underperform by 60%, with our older demo outperforming by 10%. - In ads featuring testimonials & social proof, our younger demo outperforms by nearly 50%. As is always the case ... remember to think audience-first! New Engen #DigitalMarketing #CreativeTesting *Ad screenshots are AI reproduced in general direction of actual assets to mask brand.