What Is Facebook Dynamic Creative (DCO)?

What Dynamic Creative actually is

Dynamic Creative is a setting at the ad level in Meta Ads Manager. Instead of building one finished ad, you hand Meta a pool of parts: several images or videos, a few headlines, a handful of primary-text options, some descriptions, and a set of call-to-action buttons. Meta then assembles those parts into many ad variations and tests them in delivery, steering budget toward the combinations that earn the best response. You supply the ingredients. The system does the mixing.

The cleanest way to picture it is a structured test of your own inputs. You are not asking Meta to invent anything. You are saying, here are four hooks and four visuals I believe in, find the pairings that work. That framing matters, because it is exactly what separates this feature from the newer generative tools Meta now defaults on. Dynamic Creative recombines. It does not create.

Meta introduced Dynamic Creative back in 2017, and for years it was the standard answer to a common problem: you have a few promising creative ideas and you do not want to build, name, and manage a dozen separate ads by hand. Turn the setting on, load the assets, and the system does the combinatorics for you. Per Meta's own documentation, a single Dynamic Creative ad can hold up to 10 images or videos, 5 headlines, 5 primary-text options, 5 descriptions, and 5 CTA buttons, all pointing at one website URL. Even a modest fraction of those slots produces dozens of possible combinations, which is the whole appeal: broad coverage from one ad, not twelve.

There is a structural rule worth knowing before you set one up. A Dynamic Creative ad set holds a single dynamic ad. You cannot run that dynamic ad next to a handful of separate static ads in the same ad set. The dynamic ad is the ad set's one occupant, and every asset you load into it still has to compete for delivery against the others. That competition is healthy when you give it a tight, varied set. It works against you when you stuff every field to the limit, because the budget gets sliced thinner across more variations and each one gathers less data before the system decides what to favour.

So the mental model is simple. Dynamic Creative is a combination engine fed by your assets. It is fast, it is native, and it is genuinely useful for discovery. The nuance, and the reason this page exists, is what it reports back to you afterward.

The reporting blind spot: elements, not combinations

Here is the catch that almost every guide skips, and it is the single most important thing to understand about Dynamic Creative. When the test finishes, you cannot see which full combination won.

Meta gives you a breakdown for this, and at first glance it looks like the answer. In Ads Manager, open the Breakdown menu and choose By Dynamic Creative Element. You get performance split by creative (which image or video), by text (which primary-text option), by headline, by description, and by CTA button. It tells you, on average, how each individual element performed across every combination it appeared in. That sounds like a verdict. It is not.

The problem is the word average. The element breakdown flattens each piece across all the ad variations it ran inside. Suppose Headline 3 happened to get paired mostly with your weakest image. Its average looks mediocre, even though Headline 3 with your best image might have been the standout ad of the whole test. The reverse hides winners too: a great headline can prop up the average of a poor image, so the image reads better than it deserves. You see element averages. You do not see interactions, and the interaction is where the real winner lives.

What the breakdown does not give you is just as telling as what it does. There is no per-combination report, so you never learn that Image 2 plus Headline 3 plus the Shop Now button was the ad doing the work. There is no view of how elements performed together, only apart. And there is no per-demographic element split, so you cannot see that one image carried younger viewers while a different one carried older ones. The result is a tool that points you in a direction without ever naming the destination.

This is why Dynamic Creative belongs in the discovery column, not the proof column. It is excellent at telling you that a certain image and a certain angle both seem to resonate. It is incapable of telling you that a specific assembled ad is your winner, because it never measures ads, only their parts. Treat the element breakdown as a set of leads to follow, not a result to bank.

That table is the heart of the matter. If your question is "which assets seem promising," Dynamic Creative answers it. If your question is "which exact ad should I scale, and how sure am I," it cannot, and pretending otherwise leads to scaling decisions built on averages that mask the truth.

Dynamic Creative versus Advantage+ Creative

This is the comparison the feature is most often confused with, so it deserves a clean line. The two sound similar and live near each other in Ads Manager, but they do fundamentally different jobs.

Dynamic Creative recombines assets you supply. You give it discrete, finished parts, and it assembles them into combinations and tests which ones land. Nothing it shows a viewer is content you did not upload. The image is your image, the headline is your headline, the CTA is one of the buttons you chose. It is a structured test of your inputs.

Advantage+ Creative modifies and generates creative itself. At delivery, and per viewer, it can rephrase your primary text, generate or swap a background, animate a static image, add music, stitch stills into a video, and crop your asset to extra aspect ratios. It is not picking among versions you made. It is making new versions on the fly. As of early 2026 these enhancements default on for new Sales, Leads, and App Promotion campaigns, so the creative spending your budget may already be a Meta-altered take on what you uploaded.

The distinction in one sentence: Dynamic Creative reshuffles your assets, Advantage+ Creative alters and invents them. Here is the side by side.

The practical consequence shows up most when exact wording matters. With Dynamic Creative, every headline a viewer sees is one you typed, so a price or a regulated claim stays exactly as written. The system only decides which of your approved lines to show. With Advantage+ Creative, text generation can rephrase your copy, which is fine for a casual lifestyle product and dangerous for anything where a number or a claim has to read precisely. If you want the toggle-by-toggle detail on what each Advantage+ enhancement does and which to switch off, the Advantage+ Creative enhancements guide covers exactly that. This page stays on the recombination side of the line.

The 2026 change: Dynamic Creative, Flexible, and Advantage+

If you went looking for a Dynamic Creative toggle recently and could not find it where you remembered, you are not imagining things. The feature has been moving, and a reader searching for what Dynamic Creative is in 2026 needs the current map, not the 2017 one.

The lineage runs like this. Dynamic Creative arrived in 2017 and mixed individual components: it chose one of your images, one of your headlines, one of your CTAs. Then Meta introduced the Flexible format, which did everything Dynamic Creative did and added one thing: it also auto-selected the format. Load up to 10 images or videos and Flexible decides whether to render the ad as a single image, a video, or a carousel based on what it predicts will perform for each audience and placement. Flexible was the natural successor, and along the way Meta discontinued Dynamic Creative for the Sales and App Promotion objectives, steering those campaigns toward Flexible instead.

Now the next step. Per Meta's own Help Center, the Flexible format will no longer be available as a standalone option in ad setup starting March 2026. The automation is not being deleted. It is being absorbed into the wider Advantage+ layer: Advantage+ creative and Advantage+ campaigns, plus the cluster of controls Meta groups under format display options and flexible media that let an ad appear in additional formats and adjust to placements automatically. In other words, the recombination logic you knew as Dynamic Creative is becoming a built-in behaviour of the Advantage+ system rather than an explicit toggle you flip.

This connects to a parallel cleanup on the API side, which is a good signal of the direction. In Marketing API v22.0, released on January 21, 2025, Meta deprecated the single Standard Enhancements bundle field, the one switch that turned on a group of Advantage+ Creative features at once. Developers now have to opt in to individual enhancements instead, with at least a 90-day grace period from that date to migrate before old calls could fail or be upgraded. The theme is consistent across both the UI and the API: Meta is dissolving the old all-in-one toggles and distributing their behaviour across the Advantage+ stack.

Treat the exact menu labels here as directional, because Meta keeps reshaping this part of Ads Manager. The durable takeaway is not where the button sits this quarter. It is the concept: you can still hand Meta a pool of assets and let it assemble and test combinations, but that capability now lives inside Advantage+ rather than behind a clearly named Dynamic Creative switch.

Dynamic Creative versus DCO, the term confusion

Searchers routinely conflate two things that share initials, so it is worth separating them cleanly.

Dynamic Creative is Meta's native feature, the in-platform setting this page has been describing. DCO, short for Dynamic Creative Optimization, is the broader industry category: the general practice of assembling and personalizing creative in real time against data signals. DCO as a discipline often runs through dedicated third-party platforms that swap creative elements based on a viewer's location, the weather, a product they browsed, or a live feed, frequently across many channels at once.

The clean way to hold it: Meta's Dynamic Creative is a lightweight, native expression of the wider DCO idea. It does real-time assembly, but within Meta and within the asset pool you supplied, rather than pulling from external data feeds or personalizing across an open web. When you read DCO in a vendor's pitch, it usually means the broad, multi-signal version. When you read Dynamic Creative in Ads Manager, it means Meta's specific, contained feature. Same family, different scope, and worth clarifying before a conversation runs in two directions at once.

When to use Dynamic Creative, and when to reach for something else

Definitional intent still leaves you with a decision to make, so here is the practical guidance. Dynamic Creative earns its place in one clear situation: you have several genuinely different images and a few real copy angles, and you want Meta to find the promising combinations fast without you building and managing a dozen ads by hand. It is a discovery instrument. Used that way, it shortens the path from a folder of assets to a sense of what resonates.

It is the wrong tool in three situations. The first is when you need a clean causal verdict on one variable. If the question is "does the testimonial hook beat the demo hook, and how confident can I be," Dynamic Creative cannot answer it, because it never isolates one variable or reports confidence. That is a controlled A/B test, which splits the audience with no overlap and gives you a confidence score to act on. The second is when exact wording must be protected: a price, a percentage, a regulated health or finance claim. Dynamic Creative will reshuffle which of your text options shows, which is fine if every option is approved, but you lose the certainty of one fixed message. The third is when you simply want tight, deliberate control over how much each specific creative spends, which is what a manual ad set budget setup gives you.

The healthiest way to use it is as the front of a two-step loop. Run Dynamic Creative to surface which assets and angles seem to work, then take the leading idea into a controlled A/B test to prove it before you put real budget behind it. Discovery first, proof second. The two tools are complements, not rivals: one casts a wide net quickly, the other confirms a catch.

If you want the rigorous version of step two, the way to run a controlled experiment with a non-overlapping split and a real confidence read is covered in the guide to A/B testing Facebook ads. And if you want the repeatable weekly system that turns each read into the next test, the Facebook ad creative testing method lays out the one-variable-at-a-time loop. Dynamic Creative feeds both of those. It is the wide first pass, not the final word.

How delivery uses the combinations you feed it

It helps to understand why a pile of well-chosen assets matters beyond the test itself. Meta's delivery system has become a creative-reading machine. Its retrieval and ranking models, the ones that decide who sees which ad, increasingly evaluate the creative itself, the visuals, the hooks, the language, to predict relevance for each person. Meta credits its generative ads recommendation model, GEM, announced in November 2025, with a 5% lift in ad conversions on Instagram and a 3% lift on Facebook Feed in its own Q2 testing, figures that are directional and come from Meta. The broader point holds regardless of the exact number: the system rewards strong, distinct creative because it has more signal to read.

Dynamic Creative plays directly into that. When you give it several genuinely different assets, you are handing the delivery models a richer set of variations to score and a better chance of finding the combination that clicks for a given slice of the audience. When you give it five near-identical images and three almost-the-same headlines, you are handing it noise, and the combinations all blur together. The quality and distinctiveness of what you load decides how much the recombination is worth. This is also why a strong, on-brand base library pays off: a tool such as AdPlay.ai can help generate distinct on-brand assets to feed a combination test, and the more genuinely different those inputs are, the more the system has to work with.

That is the quiet lesson under the whole feature. Dynamic Creative does not manufacture a winner out of thin material. It finds the best arrangement of what you supply. The leverage is upstream, in the assets, which is exactly where the modern Meta auction puts the weight too.

Getting the most out of Dynamic Creative

A few habits make the difference between a useful discovery run and a muddy one.

Load variety, not volume. Three genuinely different images beat ten variations of the same shot, because the test can only teach you something when the options actually differ. The same goes for copy: distinct angles, not synonyms. Maxing every field to the limit feels thorough but thins the data each variation gets, and the structural fact that a Dynamic Creative ad set holds only that one dynamic ad means the budget has nowhere else to go.

Read the element breakdown as a compass, not a map. By Dynamic Creative Element will point you at the image and the angle that seem to carry, and that is genuinely useful for deciding what to test next. Just never mistake a strong element average for a proven ad, because the breakdown cannot see the combination, and the combination is the thing that actually ran.

Keep wording you cannot change out of the rotation. If a price, a claim, or a compliance line has to read exactly, do not scatter five variants of it and hope the right one shows. Use the text you must, control it, and let Dynamic Creative vary the parts that are safe to vary.

Then close the loop. Use the run to generate hypotheses, pull the leading image and angle out, and put them through a controlled A/B test to confirm a winner before you scale. The combination engine is the fast first read. The clean verdict comes from the test that isolates one thing and tells you how sure to be. Use each for the job it is actually good at, and Dynamic Creative becomes the front door to better creative rather than a black box you hope produced a winner.