Unlocked: The State of Search Engine 2025

25:55It's such a bizarre, funny outcome. We're such fans of theirs. We're just really, like, stunned and happy. And, you know, when you make things, they don't feel very real. They feel like, you know, we're in a room together, the search engine team, trying to amuse each other, trying to come up with things that are worthwhile. And then we throw them into the vast void that is the internet, and sometimes it feels like a vast void. and sometimes the void is like, we're making a movie out of that one. You're like, oh, that's weird. Okay. So we're very happy. Is that a good description of our mindset?

26:35Am I missing anything, Garrett? Beautiful. Absolutely beautiful. Someone asked in the chat, will we be involved with the production? I don't, what we said to them, what I said to them is, we have gotten to tell our version of the story. You should tell your version of the story. If we can be helpful, we've offered. But, like, yeah, my hope and expectation is that they do something we can't imagine with it. We do get a credit, which we're super excited about. There will be somewhere, you know, in the movie, a based on this podcast episode credit. It's very exciting. Okay, we're going to turn to some incognito mode statistics.

27:13But once again, we just want to say thank you. Before we do that, we get to make the show because of everyone on the Zoom. So thank you. Incognito mode, the premium subscription, about 2.5 % of our total listeners subscribe to be a paid member of the show, which is a really exciting number for us to see. And those 2.5 % of our listeners represent like a pretty significant chunk of our budget. And so we really do get to make the kind of show that we make because of the small but mighty number of listeners who are contributing to the thing that we are making. Yeah, I did want to add that it was a big, mildly stressful year for us in terms of our paid subscriptions, which is we switched subscription platforms.

28:02And so we used to work with, we launched our subscription program with one company. And this year in March, we switched to another company called Supercast. And I know there were some kind of, as happens with these sorts of switches, some like technical issues, some folks having trouble, you know, accessing their new premium feed because the old premium feed, the old platform had not been removed. And so anyway, thank you guys for sticking with us. Apologies for anything that was annoying. I also had trouble connecting my feed. So I personally experienced what you did. But yeah, all in all, it's been great.

28:43We're super happy with where we're at. Yeah, and it's funny. It took us a while to sort of troubleshoot the source of the problem. And it meant that for months, every weekend on Sunday, I would just sit down at my laptop and try to tech support listeners who couldn't get their feeds to work. It has been really nice having that not happen anymore. Not because it wasn't nice to talk to listeners, but because one, I actually didn't know how to fix the problem at all. And then two, I just had such a sense of catastrophic worry that, for the most part, a podcast is a thing we know how to make. And we had sort of invented a second product.

29:23And the idea that we were causing dissatisfaction was such a bad feeling. And I think there's always going to be stuff at the margins, but I think that it is now working. And that is a vast relief.

29:40we're going to take a quick break when we come back we open the floor to our listeners and we talk about questions we failed to answer stories that crashed and burned and yes yes no

30:00at sierra i discover top workout gear at incredible prices which might lead to another discovery. Your headphones haven't been connected this whole time. Awkward. Discover top brands at unexpectedly low prices. Sierra, let's get moving. This episode is brought to you by Indeed. You're ready to move your business forward, but first you need to find the right team. Start your search with Indeed Sponsored Jobs. It can help you reach qualified candidates fast, ensuring your listing is the first one they see. According to Indeed data, Sponsored jobs are 90 % more likely to report a hire than non-sponsored jobs.

30:37See the results for yourself. Get a$75 sponsored job credit at indeed.com slash podcast. Terms and conditions apply.

30:48Ford Blue Cruise hands-free highway driving takes the work out of being behind the wheel. Allowing you to relax and reconnect. While also staying in control. Enjoy the drive in Blue Cruise-enabled vehicles like the F-150, Explorer, and Mustang Mach-E. Available feature on equipped vehicles. Terms apply. Does not replace safe driving. See Ford.com slash Blue Cruise for more details.

31:25And with that, we're going to transition to your questions. And so, Emily, is there one we should start with? Sure. I saw kind of a bit of a conversation in the chat about our America versus China episode, specifically about titles. I was wondering if Jocelyn wanted to come off mic. Hi, can you hear me? Yeah, we can hear you. What's up? This is really fun. This is the first time I've joined one of these, but I'm a huge fan of the show. I think my question was, do you ever test titles ahead of publishing work in marketing? So we do a lot of pre-testing and it seems like a simple way to maybe optimize to get people to download.

32:04That is such a good question. It's funny. I mean, there's like an official answer and an un-invisual answer. The official answer is no, only because unlike most, like if you worked at BuzzFeed in 2015, my understanding was everything people published was at least A, B tested and often more. They would pitch a bunch of headlines, they'd put the thing out, and then very quickly they'd get data on what was getting clicked more and that would become the headline. RSS publication, as far as I know, doesn't offer anything like that. And you kind of can't tell clicking because for the most part it's just going to people's feeds.

32:42That's not true, actually. You can tell. What we have done a couple times is we've published an episode that we really liked and we felt like it wasn't getting enough listeners. like all of our episodes tend to be within a pretty tight band but we've been like that really felt like it deserved more and then i think twice in the past year like two weeks later we've gone back and just changed the title um the there's one about teens and ai use for homework that had like a kind of like cutesy nonsensical evocative title that i'd been into and people weren't really listening so we changed something that was sort of like here's what it's really about and then I feel like the one we did about weighing cats versus birds got changed a few times um yeah awesome uh I wanted to call on Jared you had a question about using short clips if you want to come off mute and ask that sure hey PJ hey Jared you were talking in your opening statement about keeping the podcast long form um is there an opportunity or or a downside to potentially like clipping the current long form um podcast format and using it to reach out to um a wider audience or people that are maybe like stuck in that um short form area of the internet to maybe bring in new audience members.

34:10I think it's a great idea. I mean, we've been talking about it a lot internally. Like my feeling right now as someone who really does, and I hate when people are like, there's a good type of media and there's a bad type of media and the good type's the type we're making and the bad type's all the other stuff. But like, I do like long form stuff. I've noticed that there are some long form shows, video interview shows that are figuring out how to clip and exist on the short form internet. I want us to do it. I just want us to do it in a way where it feels like it's a true advertisement of the thing it's advertising.

34:42We're not grabbing something out of context where when you come back, you're actually disappointed by what you find. But also where it feels like something that is cool and entertaining if you encounter it on TikTok or on Instagram Reels. Not something where it feels kind of slapped together and we were just trying to get to minimum viable video. But like something that looks entertaining, beautiful, thoughtful, whatever.

35:09Cool. I saw a question from Chris about AI. If you wanted to come off mute. Hey, yeah, I was just asking questions about the equipment and the tech and the platforms and how you guys are looking at it, both in terms of increasing the efficiency and the productivity as well as where you've seen it not work? Oh, just like AI as a tool? Mm-hmm. It's interesting. So AI weirdly, obviously AI is both a useful and highly, highly, highly polarizing technology in America. People are worried both about job loss, but also just like are these tools that, used maybe correctly can help us research and think and used incorrectly can swap out critical thinking skills we desperately need.

36:03So we use it, but we use it carefully. The place where almost every podcast is including ours has started using it over the past, I don't know, it's been like six or seven years now is that Descript. The way it used to work is you would conduct like a two, three, four hour interview. And then afterwards you would sit down, listen to your tape, and just manually type all the words from the interview, which was like cool and also immensely not cool. You got to sit deeply with your tape and your brain dried out and rotted while you did it. Descript auto-transcribes using AI. So it will like take the audio, turn it into text.

36:40Everybody uses it. Everybody likes it. It seems relatively not controversial. And you can with Descript make like, you can do the audio editing we used to do in Pro Tools, largely in text, which I really like, I really prefer. The other place I use AI a lot is like, similarly to how I would use Wikipedia as a first stop but not a last stop research tool. Like I'll ask it to spin something up for me and then I'll go to the primary sources because as I think a lot of people know, AI can have a tendency to deeply hallucinate. Awesome. I was wondering if Dylan wanted to come off mute to ask us about the questions we've failed at.

37:24Hi, I'm also my first time joining, huge fan of the show. Yeah, I was wondering, I imagine you get a lot of questions that you aren't able to answer for whatever reason, like you aren't able to get the right interview or you weren't. but there might be like interesting components of those questions. And I'm wondering if you've ever considered like, I don't know, creating some kind of mashup episode of like, here are all of the questions we weren't able to get to a final answer to, but there was this really interesting nugget that came from it. Cause I'd be interested in knowing not only why you failed to answer that question, but also to hear some of those like nuggets that you have in the can that you weren't able to share.

38:02Truly that's such a you question. What do you think? We are such a tiny show that I honestly, I feel like if there's one really excellent moment in an interview that we will try our damnedest to save it. There's an episode we put out last year, which sat in just like sat on the shelf for ages because we just couldn't figure it out. It was, if you guys have heard the why would you buy luggage at the airport luggage store, there was a moment at the end of that interview where you hear PJ like interrogating our expert on the fact that he himself had bought luggage at the luggage store. And it was really like it made me laugh every time.

38:46And I was like, we have to find a way to do this. And so we kind of created this extensive history of commerce at the airport. And then the whole thing kind of just fell into place. And really, I love the episode. So I feel like a lot of things, it's the beauty of being able to write on the show and not just interview people is that if something is missing, we can fill it in with our own words. So there are certainly episodes that have been sitting on our docket. There's one that Garrett's going to give me a look when I say this, Ticketmaster, where there's an interview we did with an expert, with a historian who I just really enjoy it.

39:27But for us, it's how do we elevate this beyond just like a history, a fun history of a company to like answering a question that people have right now. So sometimes we, you know, we're like looking for the right hook or the right extra little shiny bit to add to the thing. I'm trying to think, is there a question where we just dumped it, but we loved it, we just couldn't. We were talking about one yesterday about volcanoes. The volcanoes one. I was thinking about the arson one, a lot of fire related. Oh, yeah. Well, let me tell the volcano one because I'm determined that we found a way to do this.

40:02Somebody asked us, like, why can't you just throw trash into a volcano? Like, why can't we just deal with our trash that way? And I, yeah, I find that question so charming. I think most of our team does. And it's like, well, what are the pieces you need for that one? Yeah. We're also just still young enough as a show where very few things have we, like, fully relinquished. We have a lot of things that are sort of like, we've put it on the shelf. Maybe we'll go figure it out someday because it's charmed us enough to, like, keep it in the mix. And I think we still have enough hubris to suspect that we'll figure out a way to make all of them work eventually.

40:36The arson one, though, it did die. It was a fascinating story where it was a very deeply personal one where somebody told PJ about a case of arson in their town. It was like a very interesting, essentially true crime story. but they realized that there were people close to the story that just didn't want the story on air. And so it was a thing where it's like, if we were like ruthless, like tabloid reporters, we would have gone after it. But it's like human beings who would prefer their story not be on the air. Like we don't, there's other stories we can do. Yeah. Cool. I saw quite a bit of love for the Colossus series in the chat.

41:16Ron, do you want to come on and ask your question? Sure. Hello. I was just curious how long you spent working on that two-part series and comparing that to the lead time, how much time you're spending working on, say, the stories, the episodes that are based on a question where PJ is mostly just doing a two-way with a guest. Great question. PJ, I'm going to out you here. So I started getting interested in data centers about a year ago. I had a really interesting dinner with a good friend who works for an engineering firm. He does like acoustical engineering. And he usually works on, you know, I've known him for many years.

41:53And he does like gorgeous concert venues and like incredible, like just these really exquisite sound focused spaces. and he told me that like all of his work, it went from being like a couple data centers here and there to like all of his work was just on data centers, like big tech companies hiring them to minimize the amount of sound that was coming out of data centers. And I found that to be such an interesting, surprising tidbit. This is before data centers kind of took over on the news and I tried for a good several months to get PJ interested. I mean, to give PJ a lot of credit, it's like there's many stories that he ends up doing, not necessarily because he was the kind of enthusiastic person, but other people on the team were enthusiastic and he's like going along with it.

42:43But, you know, it's on him to make it good and we all work together. But, you know, he has to kind of like summon up the curiosity to make it like his own. And with data centers, he just couldn't do it. And so I had to take, I was like, fine, I will report a story then. And so it's been a year, I would say, concentrated work effort, maybe closer to like four or five months. It is difficult because I am editing every episode. Of course. So, yeah, it's kind of trying to work on this, trying to work on other things. But then once I was really deep in the reporting, that Rust's team, the story was delayed.

43:24We went to Memphis quite late in the process and added an entire section to the story. And so the Rust's team would kind of really come in. And, like, that Gunnar episode saved us a lot of times. I love it. But it was really one where PJ's like, you need something. I can rustle it up in days. and it was fun just to say like as a person on the team trying to support on colossus like you know the challenge of a relatively low budget hybrid narrative show in 2025 and i'm sure also in 2026 it's like you're constantly looking for ways to ease production you know is there an idea for a really great conversation you could have that can buy you more time so oftentimes the question is like how do we make this easier but then you also have to have a little part of your brain that's like yeah but how can we make it harder like what are the things that are worth um you know the production time we're buying ourselves and shruthy who like a great editor pushes certainly me i think the entire team with her curiosity like you don't want to come back with less than she wants to know about you don't want to not be able to answer a question you don't want to have to say like oh we didn't call that person who could have been great you don't want who have like not designed like the most perfect sound cue like we're all sort of like building to her vision and it is really fun like when truth is in the reporter seat like her desire to really understand and really go there and really push it is like it's like both challenging and like you know one of the reasons we make a show like this is to be able to really like go after stuff in a serious way so it's fun to to lean into like serious reporting ambitious ambition on a story just as it's also fun to like do the breezy stuff.

45:07Okay, we're going to do one more question. Ryan had a question about segments on Search Engine if you want to come off mute. Yeah, sure. Yeah, I don't mean to compare Search Engine to Reply All for obvious reasons, right? I mean, it's a totally different show, but there's some constants between them. It just always felt like you had this segment, yes, yes, no, that was sort of intentionally different than the core investigative stories that was the base of Reply All. And they were like neat to have just pop up in the regular feed. Like I know you guys are doing them for incognito mode, but it was a nice sort of mix.

45:44And I was wondering if you guys had considered something like that for this show as like a regular segment. So this is a question really close to our hearts. I'm going to answer with PJ. I'm sure you have something to add. So we obviously, PJ and I came out of Reply All. There were things that worked beautifully about that show, some things that were much harder just in terms of production. And so I think coming to Search Engine, I mean, the very concept of Search Engine is how do we do some of the things that we did at Reply all but much easier? like, which is how we came up with an explainer show.

46:18So it's like, you know, something that is very question-driven, but an answer that uses a lot of archival, maybe just one interview, but if we really need to, like a couple interviews, that was the original concept. And so Yes, Yes, No is really just that. As a person who produced a lot of those segments. You know, you try to pick a question that has, or you try to pick a tweet, you know, but just translate that to question in the search engine world. You pick a question that has like layers to it, like different, you know, the ability to move from one like world to another has like some characters ideally and archival.

47:02So if you go back and listen to Yes, Yes, No, it's like the ones that I think really sung had like humor, had like funny tape, had stuff that we were finding off the internet to build into something that felt like more than a conversation with just, you know, three dudes in a room. So yeah, that is Search Engine. It is, when we can pull it off with just one voice, like something like the Gooner episode was Daniel talking to PJ about an interview that he, I'm sorry, about a story that he reported for, you know, several months, and he's a great talker. It's funny. There's some archival we were able to use because he had interview clips from some of the folks who gave us permission to use it.

47:43We were able to use the music from the PMVs. That was really disturbing and great. So you get what I'm saying? Once you see that that's the template, you kind of can see how a lot of these episodes are, how we think through them. The problem, and this was the problem we ran into with Yes Yes Knows 2, it's really hard to find the right ones and often things will go wrong. There's no such thing that we've ever really figured out that is just like an easy, replicable, always like, you know, no production hitches type format. I agree with all that. And like, yeah, I think like, I think one of the, first of all, obviously, yes, as soon as we're randomly chosen by Alex Blumberg.

48:27I don't know about this like big into everything. I think the place where this show actually tries to be an homage that as truthy was saying is like we're doing a lot of topical explainers and trying to pick ones that uh are sort of like like one of the modes that i think the show is in is very recent history like we're a history show about things that happened six months or two years ago um i think the other part though that you might be pointing at is one of things reply all did i think well uh was you could feel like you were learning and hanging out at the same time that it would be sort of like you'd be hearing conversation but the conversation would be narratively structured and we've leaned away from that with search engine some of that is a function of i don't have a co-host um but it's also like a pretty intentional production decision just because sort of the the the thing that repile was a little bit unfortunately like legendary for was being a show that required a lot of production hours to make.

49:29And some of that was about a bunch of particular people being particular, but a lot of it was just about when you're trying to both preserve the funny, improvised, real things that happen in a conversation, but wed them to the best possible narrative structure for explaining something, the way you do that is by working lots of hours. Like it's just, it's immensely hard, not on the performers, like the on-mic talent, It's mostly hard on the production staff, the editors and people behind the Pro Tools monitor. So we've done it a little bit. I hear that hybrid sometimes in our episodes with Manny, Noah, and Devin.

50:04When Garrett's on mic, you'll hear stories and conversation being told. But it is hard. It's really hard. And we want to make this show so that it's sometimes hard, but not always hard. I think it's funny. there's one thing we're working on that is going to go out in January that is like, um, you know, to label a somewhat benighted genre of podcasts, it is three dudes talking, but the two other dudes who aren't me are brilliant and funny and really laid on their feet. Um, sort of conversationally, like they move between analysis and comedy very quickly. And it's about an internet experiment. Like I'm to see, we learn a lot about the show by the emails we get and the feedback we get.

50:48It's not like reply all redux or anything, but the one thing I hear people missing in questions like that is the feeling of an active hang that you're learning from. And while that's not a place we want to drive to, it's a place we want to visit sometimes. And I think that story might visit it a little bit. Can you hear me trying to make the gentlest promise possible? I mean the buy the theory of the buy chocolate was absolutely a yes yes no like there's no it's just PJ and Ryan you know and Ryan's explaining to him but I just want to be careful I don't want to say absolutely because it's just like I want to be respectful to like I don't know like I made that with Alex and Alex they are like singular weird talents like you sometimes people will be like I miss this segment I'm like well people made that segment and like those people are working on different things now if I if I called Alex Bloomberg at like the place he's like trying to save the world with like green climate technology.

51:41I was like, we got to get the gang back together. He'd be like, no. But yeah, we're trying to make sure that we honor the spirit of things people might miss. Absolutely. I just meant like from a production standpoint. How dare you? I'm so sorry. Just erase what I said. You can't, it's on the internet. Garrett, you got anything else? I have nothing else. Does anybody else on the team have anything else? PJ just Take us away PJ I know I started this way but really truly thank you we are counter trend right now we are making an audio show with listener support we may as well be like in the 1970s and we are I just we all are aware of how lucky we are and thank you for creating that luck thank you for being here We're immensely grateful for you all.

52:38And with that, we'll see you at the next board meeting. Bye, guys. Bye.

53:14Thank you.

53:47There's phishing. And then there's phishing. Out here, you're after dinner. Online, hackers are after your data. They'll bait you with fake emails, lookalike sites, anything to lure you in. But with Cisco Duo's end-to-end phishing resistant access control, Every login, every device, every user stays protected. Looks like no one's biting today. Cisco Duo. Fishing season is over. Learn more at duo.com.