쉐도잉 연습: Meta Ads Insider Exposes How The Algorithm ACTUALLY Works - 영상으로 영어 말하기 배우기

Andromeda, Gem, and Lattice were the biggest Facebook algorithm changes that I have ever seen.

Today, I sat down for the last 25 minutes with a Meta executive who was responsible for these massive algorithm changes.

So in this video and interview, I was able to ask my most personal questions on how the algorithm has changed, how we should be thinking about it from Meta's perspective,

not just what's in their white papers, not just what they tell us publicly, but a true behind the scenes understanding of the internal people who have actually built the algorithm themselves.

Matt Steiner, welcome in.

How are you?

I'm doing well.

Thank you.

Great to be here.

I appreciate you having me on the show.

There's a lot of questions that I want to ask you today.

You have a pretty major role on the meta side of things

and specifically in the realm of known things like Andromeda and the algorithm in general,

which has a million questions I could have asked so much to you today.

But what I really wanted to focus on was a group of questions

that I think are going to be most impactful to selfishly some of the brands

that I work with and also ideally a lot of people that are watching this right now.

There's a lot of confusion on the algorithm.

There's a lot of confusion on what goes on.

Can you just tell the world what you've done on your side on the meta end, what you do, what you're responsible for?

Because I think people being able to hear the answers from the source is very rare

and in your case going to be pretty unique.

Great.

Yeah.

Thanks.

Happy to.

So my name is Matt Steiner and I support monetization, infrastructure, ranking, and foundational AI at Meta.

This is effectively the set of systems that are retrieving

which ads can be ranked for a person that shows up and then ranking those ads,

trying to pick the best ad that will both maximize a value for advertising partners, as well as deliver the best experience for the people who are seeing the ad,

really trying to match people to their interests and products

and services that they will love which is a good segue to say, this is what people have known to be.

This is Andromeda, right?

In a way.

So I want to actually ask you that question.

What is Andromeda?

Why was it built?

And how is it impacting advertisers so far from what you've seen?

Yeah, it's a great question.

So when you think about ranking for advertising, the first step is to select the ads

that are candidates to be shown for the person who appeared on a meta property, and there's an ad slot to fill.

That is where Andromeda comes in.

Andromeda is our retrieval system that's been co-designed,

the hardware, the software, and the machine learning model to optimize performance for retrieval and specifically personalized retrieval.

With the rise of generative AI tooling, advertisers have created a lot more creatives,

whether that's creative backgrounds or a creative copy and mixing them all together in a large number of variations.

That kind of Cambrian explosion of creatives made the retrieval problem much more difficult for our advertising system.

There were just so many more ads to consider when determining which ones to rank.

So we built Andromeda, which is a much more capable hardware system powered by NVIDIA GPUs on a custom hardware SKU,

plus a machine learning model that is tailored to work really well on that hardware system.

And the hardware system was selected to work really well with the model.

So we can maximize the performance in personalizing which ads to show for you.

Because if you think about ad retrieval, a bunch of advertisers are going to try to target me with an ad,

but not every advertiser that tries to target me with an ad has a product that I'm interested in.

My interests are a subset of the advertisers who've targeted me with their products.

And so Andromeda is really the sophisticated hardware and machine learning model

that selects of the advertisers who have targeted products at me, which ones am I most likely to be interested in?

And then those progress further to the ranking stage, which is where you hear about our investments in the generative ads ranking model, GEM, and the adaptive ranking model,

which we just started discussing a couple weeks ago.

Yeah, cool.

I want to hear a bit more about Gem as well, but I have a question on the Andromeda side, or maybe even like a summation of things because keeping it as simple as possible, right?

Like people started making more ads, a lot more ads, right?

People went down the route of essentially saying, hey, we have something that works.

Let's go make 20 versions of it.

Not old school.

I remember the days you'd have one ad and you just smash it.

You'd spend as much money as possible against it.

And I think that that shifted naturally.

And there's reasons for that, right?

We see less fatigue on ads when we do that.

And we see more scale when we actually have more variations of ads.

I think there's a misconception.

I know you're not totally tuned in on the creative side, but I do think there is a misconception that now in this Andromeda era,

that building iterations and building versions of good ads doesn't work.

And what I hear from what you've just said, which is perfectly described of what I've read and understood, is that it's not about you can't build variations.

It's just that they're going to literally be retrieved and perceived by the meta algorithm in a different way.

That's a good description.

I think the way the Andromeda system would look at this is if you create 25 versions of an ad, maybe only three of those would be appealing to me potentially.

And a bunch of them may not be appealing to me because of the types of content that I look at, because of the backgrounds that it appears on,

whether the color matches, the specific context that the ad appears in.

So a particular subset of those is going to be relevant to me.

Not all of the creatives that you've created variations for may be relevant to me.

And Andromeda is going to retrieve the ones that are most likely to perform for me.

And some of the ones that will perform for me may not perform for you.

And some of the ones that perform for you may not perform for me.

So as a result, Andromeda is selecting the most likely candidates.

It's good to have a reasonable candidate set of ads creatives for Andromeda to consider

across your various audiences at the same time, creating an extremely large number of them probably isn't going to help you more than having a reasonable diversity of ads.

That is the part that I'm so happy you just mentioned, right?

Like creative diversity doesn't mean that you have like all these like crazy different ads all over the place.

Some that are, you know, built for one avatar and some built for a totally different avatar that has nothing to do with your brand.

And I think people get that so confused these days.

And what you've just described is like, hey, you could build more you should probably build more.

I won't put words in your mouth on that.

But from my perspective, you should probably continue to build more diverse ads or clusters of diverse ads.

And even if you build 20 and three get run, three get delivery, that's almost to be expected here.

Because a lot of the work

that we would have been done by spending is now potentially saved

and done by the loads of computer rooms that Meta probably has somewhere that's doing a lot of that processing upfront, right?

100

Yeah, that's exactly right.

101

The machine learning models, they're very good at figuring out what works

102

and what doesn't work and optimizing for what is working, right?

103

So they'll attempt delivery on all the variations of the creative, then they'll figure out which ones are working best, and they'll start moving the advertising dollars to the ones that perform best for you.

104

And that doesn't necessarily mean that the other creatives were bad, it's just that they were lower performant.

105

And as a result, you probably don't want your dollars spent on those lower performing creatives.

106

The machine is doing the optimization and it's doing the optimization based on real world performance data.

107

And that's what really the machine learning models are best at.

108

And that's why we have all of these large and powerful computers that were fairly expensive sitting in that server room.

109

I'm sure.

110

How does Gem and Lattice, which are two other products or whatever we call them, services products that Meta has developed over really 2025 and into 26,

111

How are they playing in this Andromeda era?

112

That's a great question.

113

So Lattice is effectively our model consolidation technology.

114

Previously, you would have a variation of a machine learning model that was extremely specific.

115

This model is predicting clicks on Facebook.

116

This model is predicting clicks on Instagram.

117

This model is predicting conversions on Facebook.

118

This model is predicting conversions on Instagram.

119

It turns out that if you take all of those machine learning models

120

and you collapse them into one model and provide all of the training data

121

that used to go into those four models into the one

122

model the performance of the one model is better than the performance of each of the individual models because

123

that larger model is able to learn from more data more

124

of your interests more of your interactions with businesses product services etc and

125

that just improves the performance for the advertiser as well as

126

the experience for the person who is seeing an ad you're going to see a more relevant ad

127

because we consolidated these models and the training data used on these models to draw better outcomes for advertisers.

128

So that's Lattice.

129

GEM is quite a bit further expression of

130

that where we're creating this large generative recommendation model

131

that is really our foundation model

132

that all of our other models kind of learn from in that kind of second stage ranking environment.

133

And then we try to take that super large foundation model, the generative ads recommendation model and we build a LLM scale

134

and complexity inference model out of that foundation model

135

and that's what we started calling the adaptive ranking model where some number of people interact with brands

136

and content a lot have a lot of interest they interact

137

a lot on the platform they purchase more than other folks

138

they have longer sequences of interactions we can learn more from about them

139

and make better recommendations for them from their more more interactions on the platform.

140

Other folks, they don't interact as much as blip brands.

141

They don't purchase as much.

142

They have shorter interaction sequences, interaction histories.

143

And so we don't need to use as much compute in recommending an ad for them

144

because there's just not as much input data.

145

And so adaptive ranking model allows us to dramatically scale up on the kind of power user segment

146

and produce substantial improvements in performance for those users who are engaging a lot with brands and content,

147

etc. Whoa, okay, super good explanation.

148

I've not heard that explained.

149

So I need things sometimes in the simple version.

150

And Matt, that was the simple version of something that's extremely complex.

151

And I really do appreciate it because that to me is like, simply put, that is tying together the mass adjustments at Andromeda as this, this era that we're in,

152

it's kind of like, I don't want to say like the sidekicks, but like, they don't, they all work together.

153

They have to work together, which is really cool.

154

I want to ask you a question around what my audience has observed, what we've observed on a lot of the brands that we work with for context, right?

155

We work with around a hundred brands right now.

156

We're spending millions of dollars every month across the meta suite.

157

And then we also have a lot of folks

158

that watch in different modes that essentially like absorb a lot of what's going on

159

and they see they're feeling it and not necessarily spending hundreds of thousands of dollars a month, maybe spending a couple thousand dollars, five thousand dollars, whatever.

160

So something I've heard more is that brands are reporting increased delivery volatility.

161

So just overall volatility of ads, big shifts in where the spend goes, sometimes where they feel like campaigns are destabilizing,

162

especially without obvious triggers, like they're not seeing like, hey, I changed something on site that day.

163

They can't figure out what changed in that period of time.

164

So my question, is this known?

165

Is this the trade off

166

that we just have to take for faster optimization cycles where volatility is actually the feature potentially and not the bug?

167

Yeah, this is a really great question.

168

And We've heard the same question from a number of our advertising partners as well.

169

And it's a really important one that goes to how the auction actually works.

170

So because we are running a very large number of auctions in real time every second, and you have a very large number of bidders who are bidding every second,

171

the more advertisers on the platform and the more frequently they change their bid prices or bid volume,

172

the more changes that everyone sees in the auction dynamics that play out in the marketplace place.

173

It's like the stock market.

174

There are a bunch of different tickers, a bunch of people, retail investors, institutional investors, they're bidding up the prices of the stocks

175

or down the prices of the stocks in real time as they think those stocks are more

176

or less valuable to those investors.

177

The same effect is playing out in the advertising auction.

178

There are a bunch of people and maybe some information comes in

179

that this customer is much more likely to be interested in the product or service you're offering.

180

Maybe you increase your bid because they would purchase from you if they saw your ad.

181

Or maybe some new information comes in and they just bought a refrigerator.

182

They're unlikely to buy a new refrigerator soon and

183

that customer is no longer as valuable to you and you change your bid.

184

So as better information propagates throughout the ecosystem, more advertisers are changing their bids more frequently based on what they learn about customers and their desires.

185

So those bid changes at volume create the volatility

186

that people experience on their campaigns because it's a dynamic marketplace that is optimized for delivering the best price for each advertiser.

187

And so as you change your bid, different people and conversions will appear to you, not just because of the changes that you've made, but because of the changes that everybody else is making in response.

188

So volatility and associated price optimization are the opposite end of the spectrum of low volatility and higher prices.

189

So you kind of have to decide which matters more to you, lowest prices or low volatility, because they're on opposite ends of the spectrum.

190

Which is like essentially the more you're spending and the more conversions you're driving, the less overall volatility you will have because in a very simple term,

191

the algorithm will be more confident in the decisions that it's putting in front of you.

192

And then one or two or 100 people or 1,000 people

193

that may have been considered to purchase that drop out of your funnel because there's a signal like they bought that refrigerator, that's not impacting your total funnel.

194

When you're spending less or when your average order values or cost per acquisitions are significantly higher, you then therefore have to just accept that you have to deal with volatility

195

because the algorithm isn't just going to endlessly keep spending on

196

people who are no longer actually a part of your true purchase funnel anymore.

197

Yep.

198

Yep.

199

Yeah.

200

Um, as we are now again in this like Andromeda era that everybody talks about,

201

and we see people get more comfortable with how advertising is supposed to work and basically how to treat the algorithm.

202

Now, I would love to understand a bit more on Meta's long-term vision of the advertiser's role.

203

Is there, you know, an end state where we just, Hey, here's my 10 creatives.

204

Here's my hundred creators go rip it meta.

205

Like, what do we do?

206

I'd love to understand just like POV on that.

207

And I know there's no perfect vision, but an idea would be nice.

208

Yeah.

209

I mean, it's a great question.

210

I think if you look at the trends of the industry, the machine learning models and the computers, they're just getting better at what computers do.

211

They compute things.

212

They compute things quickly.

213

They can make dynamic decisions based on really good data really quickly.

214

And I think we're entering an era where humans are never going to outperform the machines at that type of work again.

215

Humans are never going to out-optimize the machine.

216

They're never going to move the budget as quickly, as responsibly to price changes up or down,

217

kind of based on the conditions of the auction market on Meta or really anywhere else, right?

218

But what are humans really great at?

219

Humans are great at coming up with novel ideas to test new messages for your brand, new messages for a particular audience, and then looking at the performance data

220

and reasoning about why did this message work with this audience and why did this message not work with this audience?

221

The computers are not going to be able to reason about that with human level intuition.

222

So humans are really great with coming up with the ideas

223

and they're really great at connecting the performance ideas with hypotheses about why a particular message worked for an audience, resonated with that audience, drove conversions with the audience,

224

and why a particular message may not have.

225

And then once you have some hypotheses from looking at the real world performance data, you can create variations on your message, on your audience, and try again, right?

226

So really this is a long-term human

227

and machine partnership where the humans are doing

228

that ideating about the initial messaging

229

and then reasoning about the performance data with human intuition

230

and insights and then starting the process over again

231

while the machine handles all of the computational optimization that they are so good at.

232

I'm going to start using computational optimization in my daily language.

233

Can we expect massive changes like this or do we generally feel like, hey, this big new era that we're in in the algorithm, this is here to stay.

234

I know some brands feel like they don't have enough time to actually adjust before the next big thing actually rolls around.

235

This is a really interesting question.

236

And I think if you look at the broader trends in the computing industry, algorithmic advances are moving faster than ever before.

237

The pace of AI innovation is absolutely astoundingly rapid.

238

New papers are coming out literally every day about breakthroughs in specific components of artificial intelligence

239

and machinery models and how that drives performance in real world applications.

240

We, of course, are focused on delivering the best results for our customers

241

and the best kind of experiences for people that see the ads.

242

And so we're trying to incorporate as many of these technological breakthroughs as we can, as quickly as we can, because that produces better results for our customers.

243

So I think it's safe to say that the pace of innovation will continue to increase,

244

will continue to drive as much innovation as possible into real-world performance benefits for advertising partners and for people that see ads.

245

And that's going to come with an associated need for advertisers

246

to understand what the opportunities are with these new algorithmic advances

247

and system changes

248

and how to really take advantage of those to get the best return on your advertising spend

249

and the results for your business.

250

I wish I had some better news that things will slow down, but I don't think they will.

251

I think they'll continue to accelerate and hopefully that produces massively better results for your businesses, but it will come with a cost of needing to keep up with all of the changes

252

that are happening in the world.

253

Yeah, probably good news for me because I'll try to keep updated on everything and keep folks updated on everything.

254

But it is challenging, right?

255

There's no doubt about that.

256

And I think just like a reminder to everyone watching this is that from my perspective, I won't speak for you, Matt, but from my perspective,

257

all advertising platforms, Meta and anywhere else you're advertising, the objective is quite simple.

258

I think I think Meta's job is to get you to want to spend as much money as possible.

259

And that want is the key word.

260

You want to spend as much money as possible

261

when you're getting the business result as closely as possible to your target or exceeding.

262

Right.

263

So any change, like I would be stunned

264

if Meta ever made an algorithmic change or a major sweeping change just

265

because it feels good or to try to move things forward.

266

Like everything is rigorously tested.

267

Everything is meant to literally put more dollars or confirm more objectives for whatever the advertisers are trying to do.

268

Right.

269

So that's exactly right.

270

Every change is A, B tested and we only launch things that move our customers metrics in a positive direction.

271

cool matt thank you very much for a rapid fire of five

272

or six questions here i do appreciate what you do i'm

273

loving the andromeda era right now i'll admit it was a

274

little tough to get used to a few things on some smaller brands

275

that we work with but um you know for those that

276

that we've battled through it's it's nice now we have uh what i think

277

and what many advertisers out there have experienced now are more comfortable in what's working

278

and how to approach things so yeah keep rocking on.

279

Thanks for helping us make more money.

280

Great.

281

Thanks.

282

Thanks for your time.

283

Great chatting with you, Sam, and have a great day.

284

Thanks, Matt.

쉐도잉이란? 영어 실력을 빠르게 키우는 과학적 방법

쉐도잉(Shadowing)은 원래 전문 통역사 훈련을 위해 개발된 언어 학습 기법으로, 다언어 학자인 Dr. Alexander Arguelles에 의해 대중화된 방법입니다. 핵심 원리는 간단하지만 매우 강력합니다: 원어민의 영어를 들으면서 1~2초의 짧은 지연으로 즉시 소리 내어 따라 말하는 것——마치 '그림자(shadow)'처럼 화자를 따라가는 것입니다. 문법 공부나 수동적인 청취와 달리, 쉐도잉은 뇌와 입 근육이 동시에 실시간으로 영어를 처리하고 재현하도록 훈련합니다. 연구에 따르면 이 방법은 발음 정확도, 억양, 리듬, 연음, 청취력, 말하기 유창성을 크게 향상시킵니다. IELTS 스피킹 준비와 자연스러운 영어 소통을 원하는 분들에게 특히 효과적입니다.

쉐도잉 연습: Meta Ads Insider Exposes How The Algorithm ACTUALLY Works - 영상으로 영어 말하기 배우기

맥락 및 배경

일상 커뮤니케이션을 위한 5가지 핵심 표현

단계별 영어 쉐도잉 가이드

쉐도잉이란? 영어 실력을 빠르게 키우는 과학적 방법