Hi there,
Thanks for reaching out!
That's a great question, and you've hit on something that confuses a lot of advertisers. Meta isn't always great with their naming conventions, and it often feels like they roll out features without explaining properly how they fit into the bigger picture. I'm happy to give you some of my thoughts on this, based on what I've seen running campaigns.
The short answer is that these two features, despite sounding similar, are designed for two completely different jobs. One is for broad creative *testing* to find a winning message, and the other is for *optimising* a specific, proven ad for all the different places it might show up. Getting this distinction right can make a huge difference to your results.
TLDR;
- Dynamic Creative (Ad Set Level) is for TESTING: You give it a bunch of separate assets (images, headlines, text), and it mixes and matches them to find the single best-performing *combination*. Its job is discovery.
- Flexible Ad Format (Ad Level) is for OPTIMISATION: You have one core ad idea, and you provide different versions of its assets (e.g., a square image for Feed, a vertical video for Reels) to make sure it looks great everywhere. Its job is tailoring the ad to the placement.
- They are mutually exclusive: You can't use both on the same ad because they serve conflicting purposes. One is for finding a winner, the other is for scaling a winner.
- The most important piece of advice is to stop worrying about the feature names and instead build a proper testing process: first use Dynamic Creative to find your best message, then build a new 'champion' ad with that message and use Flexible Formats to optimise it for every placement.
- This letter includes a simple flowchart to help you decide which tool to use and an interactive calculator to show the potential impact of proper placement optimisation on your ROAS.
Let's clear up the confusion: Dynamic Creative first...
Before we get to the new 'flexible' option, let's make sure we're on the same page about the older feature: Dynamic Creative Optimisation, or DCO. You turn this on at the Ad Set level. When you do, you're essentially handing Meta a box of creative LEGO bricks.
You can upload:
- Up to 10 images or videos
- Up to 5 primary texts
- Up to 5 headlines
- Up to 5 descriptions
- Up to 5 Call-to-Action buttons
Meta's algorithm then takes all these individual pieces and starts building different ads. It'll combine Headline 1 with Image 3 and Primary Text 2, then it'll try Headline 4 with Image 1 and Primary Text 2, and so on. It creates hundreds of potential combinations and rapidly tests them to see which ones get the best results for your chosen objective (which should almost always be conversions, by the way).
The whole point of DCO is discovery. It’s a testing tool. You use it when you're not entirely sure which image will work best, or what headline will grab the most attention. It’s perfect for the early stages of a campaign when you have a few different ideas and want to let the data tell you which one is the winner without having to build 50 different ads manually. We often use this for clients when launching a new product or testing a completely new marketing angle. For instance, for a software client, we tested 5 different feature benefits as headlines against 3 different user-generated videos to quickly find the combination that drove the most trials at the lowest cost.
The downside is that you lose some control. The reporting can be a bit of a pain to dig through, and you can't force it to spend on a specific combination you like. You're trusting the algorithm to do the work. It's a brilliant tool for testing, but it's not the right tool for every job.
So, what's this new 'Flexible Ad Format' all about?
The feature you're seeing at the Ad level—sometimes called 'Flexible Ad Format' or more accurately 'Asset Customisation for Placements'—does a completely different job. Its purpose isn't to test dozens of combinations to find a winner. Its purpose is to take a single, specific ad and make sure it looks its absolute best in every possible placement.
Think about all the different places your ad can appear: Facebook Feed (which is often square), Instagram Stories & Reels (which are full vertical 9:16), the Audience Network (which can be landscape), and so on. If you just upload one square image, it's going to look terrible in Reels. Meta will just slap it in the middle of the screen with ugly coloured bars above and below it. Everyone can see it's an ad that doesn't belong, and they'll scroll right past it.
This is the problem Flexible Ad Format solves. Within a single ad, you can say:
- For Feeds (1:1), use this square image and this headline.
- For Stories & Reels (9:16), use this full-screen vertical video and maybe a slightly shorter headline.
- For landscape placements (1.91:1), use this other version of the image.
You're not testing different messages here. The core idea of the ad is the same. You're just providing different *formats* of the creative assets to ensure the ad looks native and professional wherever it's shown. This isn't about discovery; it's about optimisation and performance. When an ad fits the placement perfectly, it feels less jarring to the user. This almost always leads to a higher Click-Through Rate (CTR), better engagement, and ultimately a lower Cost Per Acquisition (CPA). I remember one campaign we worked on for an eCommerce client selling cleaning products, where we managed to get a 633% return. A significant part of that success came from creating custom vertical video ads for their top-performing products specifically for Reels, instead of just letting their square feed images run there.
To make it clearer, here is a simple way to decide which one you should be using.
(at the Ad Set Level)
(at the Ad Level)
You probably should use them separately, and that's the point...
As you've probably gathered, these two features are mutually exclusive. If you turn on Dynamic Creative at the ad set level, you won't see the option for Flexible Ad Format at the ad level, and vice versa. This is by design.
Meta is forcing you to make a strategic choice. Are you in a testing phase or a scaling/optimisation phase?
Trying to do both at once would be chaos. Imagine giving the algorithm 10 videos, 5 headlines, and 5 texts (DCO), and *then* also asking it to use different crops and versions of each of those for 5 different placement groups (Flexible Format). The number of possible combinations would be astronomical. The system would never get enough data on any single variation to make an intelligent decision. Your budget would be spread so thin that you'd learn absolutely nothing. It's what we call 'diluting the data'.
The real secret to success on Meta isn't about finding a magic button or a new feature. It's about having a methodical, disciplined process. This is something we drill into all our clients. You have to separate your testing from your scaling.
I'd say you need a structured testing framework, not just new features...
So, how do you actually use this in practice? Don't just turn one of them on randomly. Build a process around them. Here’s a simple but incredibly effective framework:
Phase 1: Discovery & Testing (with Dynamic Creative)
- Create a new campaign with a Conversion objective. Please, don't use "Reach" or "Brand Awareness"—you are just paying Meta to find people who will never buy anything. We've seen so many accounts wasting money on this.
- In your ad set, turn ON Dynamic Creative.
- Feed it your best creative hypotheses. For example:
- -> Images/Videos: 3-5 distinct creative assets. Maybe one is a user-generated style video, another is a clean graphic, and a third is a lifestyle photo. Test different concepts.
- -> Primary Text: 3 distinct angles. One could be focused on a pain point, another on the main benefit, and a third could have a strong social proof element.
- -> Headlines: 3 distinct, punchy headlines. Maybe one is a question, one is a clear statement of the offer, and one is a call-to-action.
- Let this run until you have statistically significant data. Don't touch it for at least 3-4 days. You need to let it spend enough money to make a real decision—I usually recommend at least 2-3 times your target CPA per ad set.
- Analyse the results. Go into the 'Breakdown' menu and select 'By Dynamic Creative Asset'. Meta will show you which specific headlines, images, and texts are performing the best. You're looking for the clear winners.
Phase 2: Scaling & Optimisation (with Flexible Ad Format)
- Now that you've identified your 'champion' creative elements, create a new ad set (or even a new campaign, sometimes called a CBO or ABO campaign).
- In this ad set, keep Dynamic Creative turned OFF.
- Create a single, new ad. Use ONLY the winning elements you found in Phase 1. So you'll use your best-performing image, your best headline, and your best primary text. You are now creating your 'master' ad.
- Here's where Flexible Ad Format comes in. Click to edit the ad by placement (you'll see options for 'Advantage+ Placements', then you can select specific groups like 'Stories & Reels').
- Now, for each placement group, upload a properly formatted version of your winning creative. Create a 9:16 vertical video cut of your winning video. Create a properly cropped 1:1 version of your winning image. Make sure your headline isn't getting cut off on mobile feeds.
- You now have one killer ad concept that is perfectly tailored to every single place it can appear. This ad should now outperform the messy, un-optimised ads from your dynamic test. This is the ad you put your serious budget behind.
This two-phase process ensures you're not guessing. You're using DCO for what it's good at—broad testing. Then you're using Flexible Format for what it's good at—squeezing every last drop of performance out of your proven winner. The impact of this simple optimisation can be quite significant, especially if a large portion of your audience is on mobile-first, vertical placements like Reels.
Interactive Calculator: Potential ROAS Uplift
You'll need a final recommendation to put this into practice...
To summarise everything, stop thinking of these as competing features and start thinking of them as two different steps in a single, logical workflow. One finds the message, the other perfects the delivery of that message. Too many advertisers either skip the testing phase entirely and just guess, or they find a winning ad and then get lazy, failing to optimise it for different placements and leaving money on the table.
By adopting a structured approach, you take the guesswork out of creative development and rely on data to make decisions. It's a more disciplined way to manage your account, but it's also far more effective in the long run.
I've detailed my main recommendations for you in a table below to give you a clear, actionable plan to implement.
| Step | Action | Tool to Use | Primary Goal |
|---|---|---|---|
| 1. Creative Discovery | Test multiple distinct creative concepts, headlines, and texts together in one ad set. | Dynamic Creative (Ad Set Level) | Find the highest-performing combination of ad elements without manual setup. |
| 2. Analyse & Identify Winners | After sufficient spend, use the asset breakdown report to find the single best image/video, headline, and primary text. | Ads Manager Reporting | Isolate the creative components that resonate most with your audience based on real data. |
| 3. Build 'Champion' Ad | Create a new, standard ad in a separate ad set using ONLY the winning elements identified in Step 2. | Standard Ad Creation | Consolidate your best-performing creative into a single, high-potential ad for scaling. |
| 4. Placement Optimisation | Within this new champion ad, use the placement customisation options to upload different asset formats (e.g., 1:1, 9:16, 4:5). | Flexible Ad Format (Ad Level) | Maximise relevance and performance by tailoring the ad's appearance to each specific placement. |
| 5. Scale & Iterate | Allocate the majority of your budget to the fully optimised champion ad. Begin the process again with a new batch of creative hypotheses. | Budget Management | Drive results efficiently and create a continuous loop of improvement for your campaigns. |
This whole process of methodical testing, analysis, and optimisation can feel like a full-time job in itself, and getting it wrong can be an expensive way to learn. It requires patience and a good understanding of how to interpret the data without jumping to conclusions too early.
If you're finding it a bit overwhelming or just want a second pair of expert eyes on your setup to make sure you're on the right track, that’s where getting some professional help can make a real difference. We offer a completely free, no-obligation initial consultation where we can look through your account together and identify the biggest opportunities for improvement.
Hope this helps!
Regards,
Team @ Lukas Holschuh