Have you ever wondered why your ads keep running without delivering the performance you expected? You have tried everything from changing images and rewriting content to increasing budgets. Yet, the results remain stagnant.
The harsh reality is this: you might be making a mistake that Meta's algorithm is quietly "punishing" you for without you even knowing it. That mistake is creating ads that look different on the surface but are actually identical in the eyes of the machine.
Meta is changing the game. And if you do not adapt, you will be left behind.
The Problem: Meta Is No Longer "Lenient" with Duplicate Creatives
Meta recently updated its best practices guide for advertisers, and the core message is very clear: diversify your creatives, or the system will not rank your ads highly.
Specifically, according to industry experts, Meta has clarified that their ad ranking machines will evaluate two ads with the same "look" or the same starting "hook" as essentially identical. And when they are identical, they compete against each other instead of winning together.
Let me illustrate with an extremely simple example you might have fallen for yourself. You have a video of yourself talking about a product. You create a second video, also of you talking about that product, but this time you overlay the text "Buy today" instead of "Limited-time offer." You think you have created two different ads. But in the eyes of the Meta algorithm, that is still the exact same creative. Same face, same layout, same opening. Nothing new for the system to learn.
Olivia Cripps, Global Paid Media Director at agency Billion Dollar Boy, confirmed this: "If you have a video of me, and then another video of me, and this one has 'Text A' and this one has 'Text B', realistically they are treated as the exact same creative."
The Consequence: Competing Against Yourself
What happens when you launch multiple ads that are fundamentally identical? The Meta algorithm treats them as variants of the same thing, and instead of allocating budget to optimize, it gets "confused."
As a result:
- Your ads compete with each other instead of rising together, driving costs up.
- The algorithm lacks sufficient diverse signals to learn and optimize distribution.
- You waste your budget on variants that do not provide added value.
- Overall performance drops because the system cannot determine which creative actually works.
This is not just a technical issue. This is a strategic issue. If you do not understand how Meta operates, you will forever be trapped in a vicious cycle of useless testing.
The Solution: What Does Meta Want From You?
Meta is not just complaining. They are providing specific guidance. And the overarching message is: deeply diversify your creatives, do not just change the surface level.
Here is what Meta recommends:
1. Mix Video and Static Images
Do not focus on just a single format. Meta encourages advertisers to combine both video and static assets within the same campaign. This not only gives the algorithm more data to learn from, but also helps you reach different user groups with different format preferences.
2. Change the Opening Hook
The hook is the first 3 seconds that determine whether a user stops scrolling or not. Meta emphasizes that two ads with the same opening hook will be treated as the same creative, no matter how different the rest is. Experiment with various opening methods: questions, shocking numbers, humorous situations, or action shots.
3. Change Messaging and Tone
Do not just change words. Change how you deliver the message. One ad can be serious and professional, another can be fun and casual. One focuses on features, another focuses on emotions. Diversity in tone helps the algorithm better understand which audience responds to which style.
4. Diversify the "Look"
The look is not just colors or layout. It is the entire visual experience. One ad can use real people, another can use motion graphics. One shot vertically, one horizontally. One with subtitles, one without. The more visually distinct they are, the more the algorithm has to compare and optimize.
Why Did Meta Change Its Strategy?
Behind these new guidelines lies a revolution in Meta's advertising system. In the Q2 financial report, CFO Susan Li revealed that Meta changed user behavior understanding models to analyze organic activity and connect users with the right ads more effectively.
The results are staggering: ad clicks increased by 8.3% thanks to these changes, combined with the use of the Gem recommendation model. This figure is by no means small when you are running ads at a multi-billion dollar scale.
Meta is also testing creative insights metrics in the Insights API, providing more detailed data on the performance of each creative. Currently, only a select few agencies and partners are testing this feature, but it shows Meta is heavily investing in helping advertisers better understand how their creatives are evaluated.
Meta also emphasizes that their guidelines are results-based, not model-specific. A Meta spokesperson shared: "We do not go into the details of how our ranking models represent each creative. Our guidance to advertisers is results-based (not model-specific), like 'diversify creatives, test, and let the system optimize'."
How Are Advertisers Adapting?
Ankit Jadav, VP of Paid Social at Rain the Growth Agency, said these detailed guidelines have helped his team restructure their creative testing strategy for clients.
Instead of creating dozens of similar variants, they focus on creating creatives that are fundamentally different – from storytelling and format to target audience. This not only helps the Meta algorithm work more efficiently, but also helps advertisers better understand what truly attracts customers.
Strategy for Vietnamese Advertisers: Where to Start?
If you are running ads on Facebook or Instagram in Vietnam, here is what you need to do right now:
1. Audit All Existing Creatives
Review the ads you are currently running. How many of them are truly different in terms of hook, look, and message? If you find them similar, it is time to change.
2. Build a Creative Testing Matrix
Instead of making minor tweaks, create a matrix of variants: 3 hooks x 3 looks x 3 messages = 27 completely distinct creatives. This sounds expensive, but it is much more effective than running 27 variants of the same ad.
3. Do Not Fear "Quirky" Experiments
Sometimes unfamiliar-looking creatives perform the best. A "raw" phone-shot video can outperform a heavily produced one. A minimalist image can attract more than a colorful one. Let the data decide.
4. Track and Learn from the Insights API
If you have access to the new creative insights metrics, take advantage of them. Understanding which creatives are working well and why will help you make wiser decisions for future campaigns.
5. Be Patient with the Algorithm's Learning Process
Meta needs time to learn and optimize. Do not change creatives too frequently. Let the system gather enough data to understand what works before you intervene.
Creative Is Not Just Art, It Is Science
I have witnessed the digital advertising industry change over decades. From the days when just a pretty picture and a catchy headline were enough, to the era where algorithms decide everything. But one thing has never changed: understanding.
Meta may ask you to diversify creatives, but they cannot ask you to understand your customers. That is your job. The algorithm is just a tool. It can help you reach the right people, but it cannot create the right message. Only you can do that.
And in a world where everything can be automated, the ability to do deep creative work – producing truly unique creatives that genuinely touch emotions – will be your biggest competitive advantage. Do not just chase the algorithm. Make the algorithm chase you.
Have you ever experienced your ads "competing" against each other? Do you think Meta's requirement for creative diversification is a step forward to help advertisers work more efficiently, or just a way for Meta to force you to spend more money on content production? If you are a small advertiser with a limited budget, which of the three elements would you prioritize diversifying: hook, look, or message – and why? Share your views below, because in this digital advertising game, someone who understands the algorithm is no match for someone who understands the customer, but someone who understands both will be the winner!
0 Comment
Add your comment to this article