Meta’s Creative Diversity Mastermind: The 2026 Playbook for Ad Performance

0:00Forget everything you knew about creative testing and performance creative or meta because meta just flipped the script. The algorithm has evolved and creative diversity is now the name of the game. In today's episode of DTC Diaries, we break down exactly how to win and thrive in this new era. We go into the theory behind Meta's creative diversity rules, real examples of what's working for us since this update, and how we're adapting our process to drive incrementality and outperform the competition. This is how you future-proof your brand and how to dominate in the age of Andromeda. Welcome to Adid C Diaries and I'm excited to go through this mastermind with you today.

0:44Probably one of the most important training sessions that we have done and in this session we'll be going very very deep into firstly what does that mean so what is the difference between creative id, entity id and how to ensure that we're not being penalized for avoiding or for having too much creative similarity. We're also going to be going into loads of examples of iterations as well and what a good iteration looks like in the day and age of diversification of creative. And thirdly, what this actually means to your creative process and how it needs to evolve the way that you are working through not only iteration sprints, but also your net new creative.

1:23So there's lots in here that should give huge amounts of direction for brands, for agencies, for everyone that can really help impact and drive the best performance that you can going into 2026. Hey guys, very quick one. Our sponsors at Wavefly are offering our listeners 25k in upfront capital for a 1.8k on the run-ups to Black Friday, Cyber Monday to give you a boost when you need it most. Imagine this, your ads are finally dialed in, CPAs are down, AOV is up and you can feel the momentum building but then you run into unexpected costs which i'm sure we've all seen your budgets are nearly capped and your next move the one that could turn a good month into a fantastic month depends on cash that isn't landing for another two weeks this is how wayfly helped brands like wild true classics dock and bay to name a few achieve growth with funding when they actually need it.

2:19They're giving ambitious DTC brands$25 ,000 in upfront capital for$1 ,800 total cost in the run-up to Black Friday, Cyber Monday. That's a 1.2 % per month over six months. No equity, no hidden fees, just fuel to move when it matters most. So you can double down on what's working, scale your best performing ads while they're still pumping, stay stocked. As we know, stock can be an issue during this period so reorder your fastest movers before they sell out and move fast right connect shopify or amazon and get approved in hours not weeks timing is everything often the brands that win aren't the ones with the biggest balance sheets they're agile and ready to act when the window opens so go to the link in the description and get 25 000 in your account next week from the brand that helps you fund your favorite brands to kick off let's just firstly touch on why the algorithm has changed and how it has changed.

3:18And over the past few years, it's been, Meta as a platform has very much been shifting away from hands-on, in-account towards automation. It requires strategic volume-based approach over traditional concept isolation. As we know, Andromeda can support 5 ,000 ads a week versus 50. And the brands that understand this and cater to Andromeda's needs are the ones that are going to win on this platform. Ultimately, performance now depends on strategic volume, speed and iteration of creative. And you need to focus on high volume of content that appeals to distinctly different audience personas and preferences.

3:58However, as we know through this creative diversity, it's not just about volume. It's all of the things that I said above in terms of differentiating persona, but also differentiating visuals uh the content vehicles that we are utilizing and that in turn is what is going to drive incremental reach and net new eyeballs on the ads that we are delivering we've essentially gone from a deterministic to a probabilistic um feed so essentially social pre-tiktok let's say was based on social-based followers so in your algorithm what you would see would be the engagement based on your followers and friends.

4:36Now it's data backed based on your preferences and the level of intent the algorithm thinks you have based on targeting certain content to you. And if you extrapolate that forward, it's going to be hyper in-depth AI delivery that has such a deep understanding of all of us that we'll know exactly what creative or what ads to put in front of us because as we know um we all like to consume content in different ways and what resonates to me might not resonate to someone else even within the same persona range as myself so this kind of brings us to this idea of soulmate theory you should under you should essentially operate under this philosophy that there is a perfect creative for every single user instead of which audience fits the ad the question becomes which specific ad version will resonate most with this single user right now um and it's a shorthand for creator for andromeda's goal essentially which is every ad impression should feel as if the advertiser crafted that exact image line of copy you know persona based messaging specifically for the person viewing that and that is where we are moving towards and what we are going to be going through today a little bit more depth quick one we are hiring creative strategists at the uk so if you want to join the uk's number one independent meta agency named by meta themselves and be part of the most curious innovative and fast-growing creative strategy teams in the uk then send me an email down below we work with the uk's leading ddc brands like heights mindful chef space goods 47 skin and nutrition geeks we've recently been nominated for meta's creative diversity award because of the innovative ads that we are producing and also we have one of the best development programs for creative strategy in the UK so if you want to be part of that ecosystem and our team and want to find out more then make sure you send me your CV a link in bio lucas at ambimedia.co.uk used to be about more ads as I just mentioned but today it's very much about different ads and meta shared this our reps at meta shared this with us which essentially breaks down you know what that scale looks like essentially from less differentiated to more differentiated we want to be over here as much as we can because meta are going to reward us for that we are going to drive lower cpms we're going to drive more incrementality through the ads that we're producing and therefore get in front of new customers and you know you can see this one similar message same style so the messaging is different we might previously have run this as an iteration to just a slight differentiation of that messaging that's not going to cut it anymore and we need to be driving more differentiation through not only persona and messaging but also through the visuals that we are seeing this one in the middle very different message same style again we might run this as an iteration we can't do that anymore and thirdly this one which is where we need to be moving towards which is very differentiated message and a very different style because these people are going to want to consume this idea of soulmate theory these two ads that really are saying the same thing in different ways but it's going to resonate with different types of people so essentially they penalize us now based on creative similarities so if two ads look similar even if the copy is different it will be treated as the same entity id meaning you're we're going to be damaging damage from the point of incremental reach there's not going to be any fresh delivery and also lower performance as well.

8:11So let's actually break down what this looks like in terms of theory. So we have an entity ID and a creative ID. Essentially, entity ID is why two ads that look the same get treated as the same by Meta. So they will get grouped under the same entity ID. View entity ID almost like a fingerprint. We all have a unique fingerprint, whereas essentially Meta is saying now that if we use a fingerprint that is too similar to the existing one it's just going to be grouped into the same and it's going to piggyback off the exact same learnings um and it's going to do that because they they're too visually similar even if the messaging is different uh so essentially this means that performance data delivery and learnings are going to be shared so it's going to piggyback and absorb the learnings from the original which means we're not going to be driving any new incremental reach we're not going to be getting in front of new customers and our frequency is likely going to increase as well doesn't mean there's not a place for creative similarity but we need to be much more bullish on the differentiation um so essentially every ad that gets uploaded has a creative id uh as it gets uploaded into the ad account whereas it will likely group under the same entity id if they are too too similar so just to show a few examples that meta shared with us uh we have two here one is well they're both offer focus so but very bottom of funnel um they will have different creative ids right if they're launched into the ad account at different times however the back end is going to pick them up as having the same entity id because essentially the image is identical they are too similar and we're going to get penalized for that the funny thing is this will also get penalized because although the product is different here the message is the same the offer is the same they are very similar images right like the we've got that kind of top down tabletop shot of the sunglasses we're on the same marble backdrop so actually although they are different you know on paper in meta's eyes they are not different so these really you know you might have run a thousand ads but meta won't deem it as a thousand ads if they are too similar now this one it has the differentiation and these will be these won't be grouped under the same entity id because as you can see we've taken the same offer shop now we've taken the same sunglasses we've applied it to a very different setting so we've created that differentiation so therefore this could resonate more with a different type of persona potentially and this actually a great example is to share with one of our brands that we work with space goods and meta actually use this as an example they are grouping all of these individual buckets of creative as the same ad even though there are differences within the ad account so this one is an example same visual same hook but different video length so we're iterating on the video length meta's eyes don't care about that this one same visual angle and persona but different hot hook so that was an audio based hook was different so because there's no visual differentiation within the first three seconds which in meta's eyes is one of the most important things to ensure that we're not grouping under the same entity id it has it's essentially done exactly that and the third one which is the most shocking in my eyes different creators um but they look very similar and they've got a very similar setting it's grouped that under the same entity id which shows you know not only do we need to differentiate through the the visuals the scene the shots types that we are utilizing but also differentiate the personas that we're using so you know really making sure that we are tackling into different age demographics uh different ethnicities and you know ensuring that we are changing the setting and all of these areas because that has actually been grouped under the same entity id and even though they are two completely different creatives so why does this matter for creative diversity?

12:11Well, as I said, if creatives are assigned a new entity ID, it will start fresh in terms of audience and learning. So if we are trying to push incremental reach and getting in front of new customers, it's important that we are generating different entity IDs to allow us to do that, or otherwise we're going to plateau. True creative diversity compared to small creative tweaks is going to be key for that incremental reach and unlocking new audiences. and if a creative is given the same entity id it will share learnings and delivery which will essentially limit our ability to reach net new audiences so just to like give you an example of that how many ads do you see here if you were to count all of these well it's 20 ads so in our eyes we've uploaded 20 ads into the ad account whereas you know if meta attributes similarities under the same entity id then actually what we see uh is five ads right so it really doesn't make much of a difference so uh this is why we need to not just be driving volume without a real strategic approach to how we are differentiating the visuals the persona and the messaging and i'm going to show you a pyramid that we view now to ensure that we are really driving that differentiation um but i love this quote from Taylor Holiday, you know, if an ad is being struck with the same historical data as another ad, it's not going to give that ad any delivery opportunity relative to the desired outcome.

13:39So this just shows how important it is that we are leaning into that. And look, this is how we used to approach testing, right? We would test three different headlines that would approach the angle in a slightly different way. So this one, the first one, benefit first, The second one, fear mongering, and the third is more USP focused, whereas now they're going to get grouped onto the same entity ID. So how we need to approach it now, rather than say this, where they all look very, very similar. And, you know, there's just some small tweaks in terms of, you know, maybe what the benefits we're sharing within the static is the example here or the offer or the headline.

14:18Now, really how we should approach this is doing one version with a creator selfie. So maybe it's a creative selfie, a native style ad that is overlaying some of this, some of the messaging on top of that in a way that looks like it's come from the consumer. Maybe we do one that is like this example here with a bold overlay or headline. And then we could do one that's like a completely different version. Right. So it's taking the winning messaging or the messaging and applying that to a different vehicle. So one could be like, you know, the native Reddit format that we've seen. and these all would be targeting the same motivator whereas if we wanted to take this one step further we don't only just want to iterate on format or vehicle or the hook we also want to be thinking about persona and diversification of persona so the building blocks as i mentioned before in this pyramid that we're looking at it matters they're going to see the most differentiation and there's no like you know one is the best one is the worst uh but if we're looking through the lens of you know not grouping under the same entity id format is going to be viewed as the most different right so that's going from ugc to static static to high product etc vehicle or hook is the second most important so that is in terms of differentiating you know if you've got high production as a format a vehicle in high production might be a podcast ad it might be a street interview etc and and then the hook as well to go with that as i say the first three seconds is the most important part of differentiation to ensure that we aren't getting grouped under the same entity id the next is persona and angle which i think actually all the all three of these are the most important areas in terms of creating that um because if we're changing form with our vehicle we should also be diversifying and testing how we are more intricate about our persona based research to ensure that we are differentiating personas as much as possible so this is going much deeper into you know their core motivators as a consumer this should be going into you know what are below the surface level pain points if a surface level pain point is you know it's a product that helps with spots um if we really you know the the solution is it you know helps reduce spots whereas the and the pain point is acne as an example whereas if we go much deeper than that and tie in core human desires you know we're linking this to things like confidence and you know and other type another core human desires status potentially um so we need to be thinking about persona breakdowns in a much more intricate way to ensure we're creating differentiation there and cody uh i saw he he put something on twitter about this it's not just visuals it's the whole package visual copy offer a most importantly persona if meta doesn't think the changes fundamentally alter the experience or outcomes and thus the targeting it saves compute and buckets them together so as i say this is where that really you know the importance of differentiating that whole package comes into play and as i said you know the the idea of that soulmate theory this ties into that you know there's a perfect ad out there for everyone so it's important that we are making those iterative uh differentiation uh to ensure that we are driving that as much as possible so in terms of that like testing hierarchy if you're testing a specific persona or an angle and want to uncover winning messaging i would recommend to only do that through net new variations because if a if there's a creative that doesn't already exist within the ad account and you're running one new concept and you have four variations of that concept that is where you want to test the messaging because in meta's eyes when that goes into the ad account it is going to optimize for the ad that is performing best and therefore that is where uh we're going to see the entity id be taken versus if we are if we're testing messaging through an iterative approach it's going to group those under the same creative and therefore it's not as impactful or as important a creative test once you've found a winner and let's say you want to then iterate it on the ad account previously you know we would we would do lots of small iterations that would be isolating soft metrics like click-through rate, hold rate, thumb stop as an example.

18:53You then, when that's in the ad account, you need to make sure if it's a video that we are making visual, big visual changes within the first three seconds so it doesn't group it under the same entity ID. But also you want to iterate on things like format vehicle as well because I think that's really important. So you've got something that is working within the ad account and let's say we wanted to make a small tweak to that or adapt on that an iteration should be taking that winning messaging and applying it to a different vehicle so as i said i gave some examples of what that could be um in the nature of a static earlier but you know there's 50 100 plus different vehicles that will sit under each creative type if you want to iterate on a winner to unlock new audiences changes change the format vehicle or hook whilst also the persona or angle so if i've got a top spender i'm like okay i'd like to test this against a different persona let's say that in the first three seconds we've got a screenshot overlay of a woman in her 30s um and it's a green screen concept with a woman in her 30s let's say the it's a eyelash serum because we've got this example later.

20:07If we wanted to then test that against an older demographic, the visual change we might make to not only change the persona, but also the visual is put an older woman's eyelashes behind the screen versus the younger woman. And by doing that, we're changing the first three seconds and we are resonating with a different type of persona, which is an older woman, as an example, in this case um and if you've iterated on an existing ad with a big visual change you can test the same visual against multiple messaging tests um so yeah important to to think about these areas i think the last one is one that we are still experimenting with we've seen some success doing this some not but i think the key is we need to make those big visual changes uh to ensure that we essentially are aligning to soulmate theory as i said and we're getting in front of new eyes and as i said same messaging tests are still working but through the lens of net new concept so when you're running you know under maybe one ad set you've got four different ads or you know ad ids it is going to optimize for the messaging that is resonating most so it doesn't mean that we can't still do this and test four different ways to tackle that messaging.

21:26But the goal is to identify which message resonates with our target persona and how to communicate that message most effectively. So, you know, a winner will still emerge essentially, even if grouped under the same entity ID. And this just brings me on to like the importance now of differentiating the way that we are utilizing vehicles. So here's just a few examples of a couple of different vehicles that we might lean into founder ads podcast vsls and you know whiteboard education let's say that we had a podcast ad that performed really really well and it was more education first with an expert let's take that messaging let's put it in third person and apply it to a whiteboard concept so you can see by doing that you know we're going to be potentially appealing to a completely different consumer because meta knows and has that understanding that every there's a there's a perfect piece of creative for us out there.

22:20Um, and a great example of this is like what we did with 47 skin. We took a, an original founder VSL. We, uh, then did a before and after static winner taken into a different format. And then this one is a great example. Again, we've taken the original high prod founder story. We flipped it into third person narrative. So it's talking about Nick in the third person. Um, and then off the back of that, we have we've found another winner and we've been able to drive more incrementality as a result of that and we've used different visuals with a different hook as well so i'll just show you like a quick uh couple of moments of this so you can get an idea

23:26so yeah you get the idea again you don't need to reinvent the wheel all the time it's like taking that best performing messaging and think about through the lens of how can I how can I say this but in a different way through a different vehicle so let's go through some examples of iterating with visual differentiation because I think this is what is going to change the part of the process the most is how we're doing iterations now we do it at our agency agency through iterative sprints on a weekly basis um whereas um now we're gonna have to change that approach because you're never going to be able to draw the same volume and output as previously because a lot of those tests might have been subtle messaging tweaks as an example so how we can't iterate let's say we've launched this in the ad account it works well and then we want to test all these three as well this there's not as much point doing that anymore because you know although there's uh we can still piggyback off the learnings of the original creative um you know it's it's not going to to do what we set out of that iteration so an example of what you might actually do instead is this one which i love because it's not only changing the visual but it's changing the persona so we have a younger female uh individual here you know let's say we've created a really in-depth persona for women that are 25 to 34 this creative we ran it didn't work or maybe we were seeing it was uh it was driving more impressions from older demos and we were like okay well let's run this same creative but against a new persona by changing the visual so as you can see we've changed the visual of this persona and then we've done messaging tests off the back of that we wanted to test push this one step further maybe each of these headlines would be a slightly different angle so we could test joint pain with collagen we could test um you know reverse aging skin etc etc so as i say this is a perfect example because we're not only changing the visuals but we're also able to change the persona it's part of that wider layering system of iterations now

25:41so now i want to run through some hook iteration examples with you um you know as i say the first three seconds is what meta is going to look at so it's really really important we are visually changing the first three seconds it may ignore a new message if it's the same visual and another thing that our reps have told us is that a different thumbnail massively helps avoid entity id grouping as well so this is something we are currently in the process of testing so i want to show you an example now of how we're doing this uh one one means is through hook variations with vo3 because it's a great way of you know helping to really emphasize a message especially through visual metaphors.

26:23So here's a good example of, you know, an ad that wasn't performing that great in the account or the TSR was below a 30 % benchmark. We iterated on it by changing the visual and look at the outcome. It now has become a top spender in the account.

26:48so we've taken that and as you can see a great a great thing that we've done subtly here is again persona change right it's a younger persona here whereas now it's very obvious that we are targeting an older persona we are resonating with that audience we are our targeting is ensuring we're getting in front of the right consumer that is likely to buy from us.

27:15So we can see that rapid change of aging in the skin, which really emphasizes through a visual metaphor what the problem is. And I think that's a great example of a VO hook iteration that has worked well for us and that you can lean into as part of this creative diversification and differentiation another one is just evergreen iterations now this is just a very quick example of like it's a very cut you know quick cuts of um ugc low prod footage in the hook we've changed the visuals and now this the second one has become a top spender in the account and this was only a subtle change this isn't as much of you know layering layer stacking the vehicle the format the persona etc but yeah just to show you so this is another way that we can iterate and essentially what we've done is we've just gone into the asset library pulled different visuals that maybe drive a little bit more um what's the word just movement in the frame which again we we know can really help drive better tsr um but now this is really driven uh the account and become a top spender

28:35cool the next one again i love this example because it is visually changing the persona and it's a great example of that so this first one um this is the last brand i mentioned to you younger persona in the frame so you can see younger eyes to this one which now just by changing the visual behind the screen is going to be appealing to an older demographic i would say this is still playing you know if we remember that uh that chart at the start and you know going from less differentiated to more differentiated i would still say this is still playing the edge the line a little bit too much because there's not huge visual differentiation We might want to do more there to be able to differentiate that as much as possible.

29:28But in this case, it worked well. I think another one you can do is a one-shot UGC green screen. So this is another way you can really differentiate the visuals. You can take a standard UGC concept from a creator that maybe is a top spender. If you've got her speaking face to camera, as we know, there are apps like CapCut, Premiere, that now can just remove the background entirely. and you can put it in a completely different setting. So taking the same creator and layering them over a different background, this is an example we found.

30:05So again, that might have just been a straight face to camera concept, but we can differentiate it through an iterative approach by changing the visuals and what we see on screen. so now to go through some static iteration examples um again you know this is a good example of showing how we are changing the frame right and this is where things like nano banana are going to become even more important to your team so you're creating loads of differentiated ai image based you know product selections whole hand uh holding the product all of these different things so we can differentiate those visuals although this is great and this um this can work i think we can improve this further by changing angle and persona within the headlines so not only are we going from this to this but what are we going to do with the headline or the messaging to be able to uh differentiate the persona selection as i say we can go really in depth with that persona selection by going much deeper than what we see on the surface layer with a persona what are their biggest motivators um you know what are their failed solutions what are some of those un uncovered pain points that we're not currently addressing within the creative this is another iteration that worked really well for us um after these uh creative diversity updates from meta this is essentially just a quite a quick win taking um a reddit style uh static and what we wanted to do was make it very, very clear who this person that we wanted to target was and who we wanted to resonate with.

31:47So we just added an image of the persona on screen as part of the comment. And again, that has helped drive that differentiation. Like I said before, we might want to push that even further through even getting in front of a potentially new persona through that messaging.

32:12cool another good example of just a very easy native based one now again a good example from our previous slide let's say that reddit one did well taking that applying it to a text message again creating that differentiation getting in front of a different type of person so we could be from this one here and then we iterated it on and put a split screen and put the product front and center by doing that it actually you know slightly brought it down the funnel um to people that are aware of the brand but putting now in boots really helped because it gave it almost that pr stamp of approval and helped build trust and drive ctr as a result of that another one which again is playing it very on edge in terms of differentiation but this one actually not the second one you see here on the right became a top spender so um what we've done here is essentially just adding much more context around what some of the features are of that product and we've changed the composition of the frame as well as a result of this but this again is playing on edge in terms of you know probably sitting in the middle of the park of differentiation and that's a this is another one that we did that was more seasonal based this is a great example because we're ticking a lot of boxes with this one we're going we're changing the format which is the biggest swing change we can make to create that differentiation we've gone from a static we've taken the text on the static and we've applied it to a mashup here which again has worked really well

33:57very simple very quick change you could do this could even be a static in itself right you know that frame there that thumbnail um but you know again that that has really really helped create that differentiation by taking that messaging from a static applying it to a hybrid or a mashup style creative another one that again is technically changing format is going from a static to a motion graphic there are loads of tools that can help you do this now um we did this actually i think through adobe uh after effects but there's loads of ai tools that can help bring statics to life uh so we've gone from this static here which is the top spender to this one.

34:43Again, just through subtle changes in visual emotion, we might be able to appeal to a slightly different consumer. And this is my favorite that we've already gone through, but completely changing persona with the image, which is one of my favorite ways that we can iterate at scale. And again, this is why I think it becomes so, so important that you're thinking about, you know, how this impacts every stage of the process. Right. Because this now means that we need to be even more bullish when we're working with UGC creators to get static shots of them, evergreen shots of them holding the product.

35:20Because we know that the more differentiation we have in the personas down to age, race, you know, all of these different factors that is going to help us potentially get in front of new customers. And we've seen this before with other accounts, even from a UGC perspective. You know, we had very southern based accents in the ad account and we were we weren't driving incremental reach. And then as soon as we brought in more diversity with northern accents, as an example, that's subtle change. But a big one from a perspective of persona that really helped drive incrementality. And that's how we need to be thinking about, you know, creative diversification.

36:01communication it's not just through the lens of how we iterate how we make it different the key and the motivator behind everything and why we change anything there needs to be a hypothesis behind why we're doing something and we need to be thinking about it in the research phase not when we're coming to iterating on ads this is another one so actually plugging together two different format types so static and high production split screen um and this is a good example I saw from Dirty where it's a static on the bottom we got a high prod on the top and we've merged both of those together into one again quite a lazy iteration but again can work so now and we're very early in this process right of what this actually means to us and and we're doing lots of tests right now I think I see on Twitter and X there's huge amounts of controversy and debate around what this means and how we should be using it so we're early days but some of the big things that we are changing right now is less but more intentional iterations i think the way you need to view it is an iteration now it has to be viewed in a bigger swing nature you're never going to achieve the same volume of iterations you are right now to ensure that you are not being grouped under the same entity id so less but more intentional iteration so therefore as a result we're probably you know if we've got say we need to do 100 iterations in the month we're probably now doing 50 to 60 that are more intentional because therefore they are going to drive more impact and therefore performance and putting the rest of the 40 percent towards net new variations of concepts because that is where we can be testing persona headline tests messaging adjustments a lot more of those more subtle based tweaks that previously we might have tested in iterations now we need to test them as net news um we are moving to more of a bi-weekly iteration sprint because we are being much more intentional about the iterations that we're putting forward we want to be spending more time on not only the how we are hypothesizing those changes and briefing in those changes but also to allow post-production teams and design teams to be able to actually you know make those bigger changes to ensure they're they are visually different um more intention when briefing hook visuals so whether that's for UGC creators whether that's for um you know high production whatever it is be more intentional about what the visual direction is of those hooks because that example I shared at the start of space goods where two different personas same setting we could have got around that by differentiating one of their briefs with being very clear about how to differentiate the visuals or be in a slightly different setting than the kitchen.

38:59Messaging tweaks through net new variations, we've already touched on that, but that is where you want to be doing the large chunk of your messaging tests. More intentional persona research, as I said, really diversifying the personas and getting much deeper to their core motivators is going to be really, really important because that those subtle messaging tweaks headline changes hook adjustments is how we are going to be getting in not only the core bucketed persona but also the uh the the different verticals of persona that sit beneath that um brief creators to send through native statics as i mentioned this is this a moment ago that is now extremely extremely important um because obviously uh you know the the more statics we have through you might as well kill two birds with one stone right if you've got a ugc video from a creator it takes them two seconds to take a photo of them with the product that there's as as we've shown through the examples within this deck there's huge amounts of things we can be testing there to differentiate personas um through those visual changes within the iterations uh and then finally you know and this just goes from what we've said previously but i feel like the new creative diversity changes really lend themselves to more whitelisting and more emphasis on whitelisting because as we know like if we are whitelisting with a creator uh every time we do that it's going to to have its own entity id there's not going to be that differentiation there so it lends itself to creative diversity and differentiation um yeah that's just a very quick rundown of how we're using and or thinking about creative diversification it's very early days uh i'll probably do a follow-up here in the new year once we have a little bit more data to go by but every example that i shared or at least the 80 of them that came from us specifically um every example i shared that came from us is a top spender and and it has become a top spender from the point of when we shifted to creative diversification and meta's new updates there so uh if you've got any questions please let me know in the comments and we will make sure to get back to you on that um and yeah thank you for watching please subscribe give us a like uh on and subscribe on both spotify and youtube and we will see you on the next episode