Show Transcript

Very, very exciting, Miranda. We're live for our second ever SEO Community Campfire Chat. Before we dive in, I gotta say thank you, thank you, thank you to all of our amazing sponsors, because without them we really couldn't do this.

Our community partners, really exciting. I wanna talk about SEO jobs, Dry Fly Digital, Citation Labs, Bento.

which helps us with our email marketing, AirOps, which I think is just super baller. They do so much stuff with like agentic workflows and ModEx, which powers our community website. Everybody.

It's so exciting today. We're joined by Miranda Garman. She is vice president of organic performance at monster.com, which is really rad. Miranda, we've known each other for a whole bunch of years going back to our connections.

Miranda: Thank you.

Noah: on, on clubhouse and you've always blown away with your technical acumen, your ability to explain like crazy complex topics, but also your, your crazy specialty of doing web migrations at scale and cadence, which I is the craziest story that I've ever heard of. And so when I started thinking about campfire chats, I was like, we've got to have Miranda on, like, we've got to talk about this. I can't wait to like do a webinar with you. So.

It's been a lot of years in the building. Do you want to say hi and tell us a little bit about your background?

Miranda: Yeah, so first of all, thank you for inviting. yeah, so I have been in the SEO industry for the past 17 years. And at the time, I actually always focused on international SEO and enterprise SEO, so always multilingual and luckily also for lot of large clients like Philips, McAfee, Ronstadt, WP Engine. And currently as vice president, what you mentioned at vice president organic performance at

I also have my own ANC for which I work for some enterprise clients and that's basically what I'm doing at the moment.

Noah: And you're the really cool, your own agencies digital rescue Rangers right.

Miranda: Yeah, and I can tell you actually why and where the name is coming from. Yeah, so at the time, we work at a very large enterprise company based in Netherlands who had over 80 markets. And we always have a Chrome extension in our browser, which is normally green, basically means the page can be indexed. And so it's always there, and we don't really look at it. But then I was going through the website, and suddenly it turned red. So I was like, this is...

Noah: Yeah, cool.

Miranda: weird. then so basically what I found out is that all the or half of the site, all the e-commerce pages, so the pages that drive money, they were on index and that was one site and I was like, oh shit. And then and then I checked across all the 80 sites and we were like, oh, this is this is something really, really bad. And so we flagged the developers like this is happening, please fix it right away. So it got quite quickly fixed, but still Google picked it up.

Noah: no.

Miranda: pages were deleted from the index, we had to resubmit. So this whole process, basically, we were chatting with each other and we said, the rescue ranges of the companies, imagine if this would have continued for a longer time. So that's why basically the name Digital Rescue Rangers came from.

Noah: That's amazing. So I'd love to dive into the topic of the day, which is doing website migrations. If I go back to 2021 when we met, I think at the time you were working with Rantstat and you're in the middle of a project, the kind of scale that at the time I'd never heard of anything like this. Can you walk us through?

the remit that you were working through when you were doing that migration project with Ransat.

Miranda: Yeah, so basically, with ones that were operating in 42 markets, and then in some markets, they were having multiple sites. So basically, we build a whole CMS, a new platform, and we went from the old one to the new one. These websites have, as a minimum, 200,000, 300,000 year-rounds, and that was like, as a minimum. basically, like the whole flow of like,

making sure that the new platform is SEO friendly, the migration step, that was basically, and then the actual migration that we did basically repeated for all like 52 sites, just at Ransdad.com.

Noah: Wow. And were you going from a whole bunch of different tech platforms? each of the sites, was it possible to end for any of the sites to have a different tech stack or were they all built on the same stack to start?

Miranda: Most of them were built on the same stack and others, like because on the run, they had sub-brands, so a couple of them were built on different stack.

Noah: Crazy. Okay, so do you want to walk us through what some of the lessons that you learned and how to execute a flawless migration? Do you want to share your screen and walk us through that? Awesome.

Miranda: Yeah, sure.

So let me share my screen.

Miranda: I think you're able to see it now, right?

Noah: Yes, for sure.

Miranda: So I just prepared a few slides, but basically this is what I told already. So 17 years in the field. I'm also quite always trying to improve myself. So I'm currently working on my third master and then yeah, trying not to do it like a normal university, but trying to get the highest of the highest. So that I do as well. So it's quite hectic sometimes. And, but then about site migrations.

So especially for lot of SEOs, this is like a headache topic and there's a very good reason for it because it can go really, really wrong. So this is one example of WhoCommerce who tried to at that time move their pages to who.com and it basically deleted all the traffic and it was so bad that they just decided to go back to the old domain.

So with site migrations, as soon as you hear it up.

it's good to get directly involved because if SEO is involved on every step and very early in the process, it can go relatively safe. But if not, or if not with full attention, it can go really, really wrong.

Miranda: And basically what I find is that preparation is key. So the better I prepare in terms of roadmap, getting the details, getting the scope, who is responsible for what, getting the timeline as well, that really helps to get a better migration process.

And often there's a couple of key decisions basically or questions that come up. First of all, it's very important to know what is the platform that you're going to. If you currently have a slow platform and you go to a much more modern, faster, cleaner code, that sounds good.

If you go to, for example, JavaScript framework, which are very popular nowadays, React, et cetera.

That can be very fast, but it has an extra level of complexity. For example, making sure that you make use of server-side rendering, that the content is directly in the initial HTML so Google can very quickly see it. So can I?

Noah: As you started to talk about that, I had already written down SSR. Like, I was gonna ask you about it.

Miranda: Yes. So, and nowadays, like, JavaScript frameworks over the last years are very, very popular. A lot of enterprises where I work with, they are using React. And so that's something to be very aware of. And then specifically about this one, that's sort of question about server-side rendering or not. So anything static, content pages, content on the site, that should be like server-side rendered.

But for example, on the case of Monster, if we show related jobs, so we have an article and we want to show what are the related jobs for this article, then we don't want to have that static because it could be that this job goes offline because it's fulfilled. So we don't want to the experience that if user clicks on it and it says, okay, this job was gone. So that's why for anything that is not expected to change,

like a few days or maybe even just daily, that can be server-side rendered. anything that we need to make sure like it's very fresh and it depends, for example, on the user query or user behavior that is dynamic loaded in more appropriate.

Noah: So anything that deserves freshness, your client-side red earning, you're saying?

Miranda: When it's not the key topic of the page, for example, so we want that the content is also fresh, but if that is dynamically loaded in, then you get an issue with indexation.

Noah: Got it. Okay.

Miranda: And maybe on that perspective first. So Google first crawls the content in the initial HTML, and then it comes back later in JavaScript. And JavaScript takes nine times more time than crawling HTML. And large language models don't crawl JavaScript, for example. So therefore, also Google recommends to have all your content that you want to have indexed directly in the initial HTML, so directly what Google can see.

Noah: That's wild. The nine X number. We had a discussion last night on clubhouse about how important it was to have everything available in the HTML because of the cost and the time. But to put a number like nine X on the time is crazy.

Miranda: Yeah, I think that came both from research from Google as well as another agency who did a research on that and they also found nine times.

Noah: Yeah, wild.

Miranda: Then another question which comes very often up like, or basically if you change the domain name, you need to change URLs. And so that's more logical question. But when we do migration, we get a lot of times like, can we change this folder structure or can we make these URLs more simpler? And if you want to maintain SEO traffic, that's usually not a good idea. So it's the best to keep the URLs the same. And less.

your old site is really, bad. And so it's a clear improvement. Then it's fine. otherwise, I'll just get the tissue. I had allergies. But otherwise, it's recommended to keep your URLs the same.

the way, actually, about this topic on John Muller from Google, he's also saying that it has very rarely a benefit.

and that it imposes a high risk for the traffic. So he also ends some risk, usually no gain, and basically means not a good idea. So that's why Google also don't recommend changing URLs.

So the arguments that I then often use in my conversation is basically to give an estimation. If you want to change the URLs, it has an impact of 15 to 30 % of your traffic.

Noah: Totally.

Miranda: decent migration. These amounts are coming from the 15 % are coming from the original page rank document of Google. And Google was using first 15 % dilution as well on the redirect. Like more six, seven years ago, Gary Eels said that there's no dilution anymore of redirect value.

However, with the website we did and the migration even at Randstad, we see that even if we keep all the URLs the same, there's some loss of SEO value. that's why from our experience, we don't agree with the zero percent. And this percentage is also coming from Ignite Visibility.

They also shared in the blog their experience. instead of like...

you will lose some SEO value. It's easier for the higher management to have like a number because then we can do like calculations. Okay, this is the current traffic.

This is the potential traffic that is at loss. And is it then a good idea to change the domain or change the URL or not?

Noah: It seems like everybody gets so excited to change URLs too. Clients are like, Hey, we got to do this. I want to change this and that and this and that. It's like, no bad idea. The data backs it up too.

Miranda: Yeah.

Miranda: Yeah, even so, if URLs would be changed, readbacks need to be mapped, et cetera. But then also internal readbacks need to be updated. So although the user will land on the correct page, there's an extra step. And that's why Google also recommends to update those internal links. So basically, it's like, more risk for your SEO traffic, more work for development and for the SEO. And there's no positive ROI.

That's basically how I try to use those points in my conversations.

Noah: Love that.

Only risk, no reward.

Miranda: Exactly. And then from a preparation side, so it's basically the earlier you hear about immigration, the better it to get involved to make sure we have the maximum influence as SEO. So even in the initial stage, like what will be the platform, what is the purpose, et cetera. And then even before things are being built, so in terms of the wireframe pages, if there's a new design.

to make sure to already check, like, is there enough place for content included? So that's why it's basically good to get as early involved as possible.

Miranda: And then if the platform needs to be rebuilt, then what I usually do is I create a document for SEO requirements for the developers that before they start building, they know already what are our requirements. So there's less of a review at the latest date. Like one enterprise site became a bit extreme because I had document of 38 pages of SEO requirements and literally sped up.

Noah: my God.

Miranda: Yeah, it was a very long document. This specifies everything, how should an image URL be, like with hyphen separated, small case, the format, like try not to upload it as like a huge file. And then like, all the different aspects, images, videos, URL structure, folder structure, canonicals, hreflang, sitemaps, and so all these things together, having that.

Noah: You

Miranda: I'm not sure if other developers all really go through it, but in case they have a question that they can use it as a reference. Like this is what we have agreed on in the beginning. And then at any point later, if they have questions or they can use this reference to build their products. that hopefully then if you come at the review stage, go ahead.

Noah: Quick detour, quick detour on that. So do you demand in your dev workflow whenever they're like pushing through a sprint that they have to meet all of the SEO requirements or it's a failed build? Like have you been able to argue for that?

Miranda: So what we had at Rundstead is that for every component that was being built, there was an SEO, UE check. So basically, even if it's not SEO or for SEO purposes, we had to validate and say, okay, this is fine. Because in the end, everything on the website is influencing SEO. So that's why we had that built in.

When Teams became larger, it was a bit difficult to manage because then we had to say, okay, these are not stories affecting SEO, so we had to fill that a bit down. But basically making sure that when stories were developed, that we were for each story, that we had something to say as well.

Noah: Okay, cool.

Miranda: And I think, okay, I'll come to that later. Another very important part is making sure you have a list of all the URLs. And so that means combining different sources, Google Search Console, Bing Webmaster Tools, which is not here on it, Analytics, backlinks, sitemaps, site crawlers, log files, and just merge everything together, clean them up, remove the parameters, etc. And then have a clean list. That will help, first of all, to see like,

If pages have zero traffic, it's okay, they are less important, but at least make sure that all the pages that generate traffic, that they are included in the scope of migration. What we had in the beginning of migrations is that when we made these exports, and then when we started migrating, that the content team, because other people also work on the site, that they published extra pages and we were not aware of them. So we were like, hmm.

which that caused an issue after Go Live. So that's why we learned that before we basically start the actual migration, or basically the start of the content freeze, we need to repeat this process again, just to make sure that any pages that have been created, that they are included in the final list to migrate.

Noah: I love that focus on what actually drives traffic because that's the only thing that's part of your potential loss, right? So like focus on what matters.

Miranda: Yeah. And then sometimes there are pages with, it's like a legal requirement to have them on the site. They don't have a lot of traffic. So it's just good to double check like, is there anything that's like, is very important to move.

Miranda: And then the more in detail, the more laid out, like who's going to do what, which phase. We used to work with spreadsheets and then like the waterfall basically planning, where we specify this person is doing this. So we had that ready for every migration and that worked very structured. And then also having like weekly meetings about how are we proceeding with the plan. But I think it's very good at the beginning to make sure that

everybody knows their responsibilities, what is expected, for which date, and that should all be included in the project plan. So it's not necessarily a project plan for SEO, it's basically a project plan for the whole migration.

Noah: Who owns that? What role owns the plan?

Miranda: We had that it was the product owner, but then basically there were the different departments that were working together, like analytics, SEO, paid search team. So we all had our share basically in the plan.

Miranda: And then to actually prevent that, or when developers basically become very annoyed, is then when they almost build with, or they're done with building and they're almost ready to go live, and then comes SEO in and basically says like, oh, these are 20 things that are not SEO friendly, please do it again, while they are stuck with their initial deadline. So it's very important that instead of reviewing at the end when they already build their most of their things,

to actually see if you can review components or review templates. So as soon as one or two sprints are finished, they have something delivered to already review at that stage. And that will make sure that, first of all, you catch things earlier, but also that those things that we flag, are then also included for the next components.

So that's why it's better to review throughout the migration process and then still have your free audit, go live day audit, and afterwards.

but instead of doing everything at the end or when the site is almost ready to go live.

Miranda: And I added the immigration checklist as well. I will get back to it later. But in case you want to see what are the things that I look at. And this very differs per site because in this one, I don't think I included structured data. And this depends per e-commerce site or per job board. But if you want to have a base, you can copy this

Miranda: So couple of things to minimize the traffic loss is basically you migrate all the pages, and I must say pages that had traffic. And sometimes what happens is that people say, like, these pages have like 10, 20 visitors. They're not really important. Let's kill them. But then if you have like 100,000 pages of these 10, 20, then together, if you sum it up, it's still quite some important traffic.

So that's why I prefer to at least migrate all pages that have traffic and to minimize any risk on that. I did work with another client and they said, no, we don't want to do that. We have millions of pages and we want to move only with a clean version to the new site.

So there we actually switched it around. We said, okay, first and clean up the site like how you want, have then a five month break.

So basically the impact of that upgrade or that cleanup basically is stabled out and then we start the migration. So we still isolate the impact of the migration and not changing things at the same time.

Noah: Thank

Noah: So that's fascinating, that delay, the five month delay. So like how much time do you typically like to have to lead up to do a migration? And there's the ideal of, I want five month lead time or whatever. But then there are times that you get brought in when another company fails at any step along the way through this project plans, right? So it's like, that's best case.

What's the worst case you've hit?

Miranda: Yeah. The worst case is like, hey, we just had immigration. Traffic is going down. Can you check why? And then basically the damage is already done. So in this specific case, it's an enterprise client and it was like a two year project. So they told me, and I had like deliverables in the beginning. So that was the company where I made the 38 page document for.

Noah: Yeah, yeah.

Miranda: So I told them, OK, then first do these cleanup changes. Then in the meantime, I will create this document. We make sure that everything is detailed, the project plan, et cetera. And then by the time that we start, we have this five months break, basically. But that's not with all the clients. Some clients, indeed, are like, we need help now because we went live and traffic is dropping. that's actually, yeah. Then you need to go back to like,

archive or see what was the previous content on the website. If you could still find it.

Noah: Hmm.

Miranda: And another very important one is to isolate the migration from anything else. It's actually the same or like a follow-up. If you isolate the migration, you can see if traffic goes up or down, you can see, okay, this is due to the migration and spot any bugs or errors that come along. If you do content updates at the same time, which often the client wants, because they go live on a website, they want to fresh content, maybe we do some content.

And then it becomes really, really messy. That was actually with the first site that we had with Brønstedt and then live. The local team, they made still updates.

And then we had a migration, URLs were changed, which were not initially in my list. Existing URLs, they were not there anymore. Pages were shorter. So it just creates a mess and it leads to traffic decline.

it takes, for example, like three months break. So you can see, okay, this was the impact of the migration, went well or not. And then start doing content updates.

Noah: Quick question from the audience. We're gonna get to the first one in a little bit, but the second one is, how to find the old content if the site has already been migrated? What tools do rely on?

Miranda: Yeah, so I often go to archive.org. That usually works for larger websites. And yeah, that's actually my go-to point. But for smaller or less popular pages, they often don't have the content. so archive.org is my go-to.

Noah: Got it.

Noah: If they have, let's say they have the BigQuery export set up pre-migration, you can probably lean into that. Like, hey, if a page gets impressions, that's at least the, you know, that's a good starting point probably too, right?

Miranda: Yeah, as a data source. But then you still need to know what content was on that page.

Noah: Oof. It gets harder.

Miranda: Yeah. So that's why for now, it would be nice to know if there are any other tools out there. But for now, the tool that I only know that does this is archive.org.

Noah: Hmm.

Miranda: In terms of redirects, in case pages need to be redirected, it's highly recommended to do one-on-one redirects. If we have articles and we just say, OK, we redirect to this category, Google sees that the new page is not giving the same answer as the old one, and basically it will just delete the traffic. And also in the Google documentation, it also said it basically would do

to the dilute the SEO link juice basically coming from that page. So it's just when you create a one-on-one or when you create redirects, they need to be one-on-one going to the page with the same information. So that's a very important one.

Noah: Is that a hill that you're willing to die on as it were when you're talking with stakeholders about managing a future migration? Would you put it at that level of importance like, or is it like we're always shooting for that, but we get to achieve it 10 % of the time or is it like how?

Miranda: No, no, would go to do these sheets. I was working until my 2 AM, 3 AM, getting all those redirects done. to, in the end, at Ransselt, I was the global SEO manager. I would be accountable if the traffic goes down. So I was like, I want to make sure it's excellent and there's no traffic loss. So that's why these one-on-one redirects, if.

Noah: Yeah. Yeah.

Miranda: And suddenly you come to pages with like 10 traffic. Okay, and there are not a lot of them, 10 visitors, okay. But usually I try to use for everything the one-on-one video ads.

Miranda: These days, I know that people are using AI tools as well. I do find that can speed it up, but it's important to make sure that it's the same page or relevant content. One thing that we had, for example, with monster is that we changed the province or the region name. And that's very likely wouldn't have caught with tools as JTPT at that time that was not available yet. But...

If you have asked Chattpt or other tools to create redirects, make sure you review them very carefully. So to make sure you still have like the one-on-one redirects.

Noah: You mean AI sometimes can be problematic? Who would have thought?

Miranda: Who would have thought that AI sometimes gets wrong answers, right?

And then the big, big day basically on Go Live, there are some things very common. So first of all, common to go wrong. So first of all, as soon as the website goes live, it's important that the redirects need to be there.

You see what we used to do is implement the redirects already on the development domain, test them all there, see if it works, and just change, or basically developers push the site live, and then the redirects were already there.

But yeah, just to make sure that Google doesn't see the URLs as 404s. Those should be actually like instantly going live at the same time. And another thing, very, very important, small thing can be overlooked, but often on development sites in the robots.

txt, it says like, don't crawl this site. And then when you go live, it's very important to update the robots.txt. Sometimes we found it out one day later.

It's not so good. So this is something to be very aware of.

Noah: Back in 2018, I was at MozCon and I talked to members of the largest real estate website in the United States, what their biggest mistake ever was. And two issues were, one, de-indexing the entire site, which meant that Google like de-indexed millions of pages over a period of a week. And then the second thing was they pushed live an infinite crawl loop.

which had the opposite impact of indexing millions of pages that just went deeper and deeper into the structure. So yeah, that stuff matters. Robots.txt.

Quick question. What is your favorite and least favorite days of the week to go live on?

Miranda: Good question actually. We usually do Tuesday and Thursday. So... No.

Noah: Got it. And Friday is never, right? Never, ever, ever launch on Friday.

Miranda: Yeah, and the reason for that is if something goes wrong, you want to make sure that the developers are available. So when you go live on a Friday, it's weekend, you don't want to necessarily disturb them or they have other plans. So that's why we prefer we go live on a Tuesday because then we still have the days after for testing and reach out to any of the developers if needed. At the time when we had, for example, two migrations in one week, we were choosing Tuesday and Thursday.

Noah: OK, I interrupted you. think you're about to hit the sitemaps piece.

Miranda: Yeah, just in case the website changes, then make sure that the sitemaps are also updated in Google and Bing's webmaster tools. so before the go live, there's a technical audit. And sometimes it looks if everything is fine there. Links are going to the correct environment, et cetera. But then when you go live, sometimes the links are hard coded, which should not have done. So that's why immediately on the go live date,

it's important to do a crawl to make sure or to double check, are there not any links pointing to the dev domain. We had that happen a couple of times. And then basically the development domain was getting indexed.

And then it was like a hassle to get them out of the index again. So it's just on Go Live Date, make sure you check for the links besides all the broken links like normal technical audit, make sure to also check for links to the development domain.

Noah: Do you ever do a soft launch where you sort of have a two stage where you push rel canonicals in advance of the actual migration or you like have the new server live enough that you can set rel canonicals to point to whatever the new URL pattern is and then go live or do you always do just a hard?

or do you?

Miranda: So, yeah, so usually I do like a hard one because it's a change of, or basically we stay on the same domain name. So it's just like a replacement with a new migration that is coming up. There we probably going to launch or move like a subsection of the site. And that's also related to email marketing because for email marketing, if you have like a new domain,

like it's not so much information can land in the spam box. So actually they need to basically warm the domain up. And then we said, okay, also for SEO purposes, we'll move like one section of the site to that domain as well, and then move the bunch.

But usually because we stay on the same domain, it's just like a hard switch.

Noah: I love that. I had not considered the email marketing impact, even though like for the SEO community, we're about to launch our weekly newsletters or bi-weekly newsletters. And it's like, we've just spent the past month warming up our domains. So like, it should have occurred to me. I never have thought about email marketing impact as part of a migration before.

Miranda: That's actually the nice thing about user integration. You work so closely with all the different departments together. yeah, so it's... like with the impact of the cookie banner, slow... slowing down the loading time, et cetera. So it's also a nice collaboration project with the different departments.

Miranda: And after the Go Live, especially I was doing an audit, I think every three days. And then from two weeks, once a week, and then for one month. So I basically called it aftercare to spot any issues. Especially in the first days after Go Live, we want to make sure that we find issues before Google does. So that's why the frequency is much higher.

When we see the impact on the search results, that usually came in two to three weeks after we went live. So there's a bit of delay because Google is sending traffic, but then it doesn't crawl all the pages of the site, at least not with enterprise, at the same day. So before then, process the pages and adjust the rankings.

So usually it takes two to three weeks before we start the first initial impact of the migration.

Noah: Awesome. So this is a perfect opportunity to ask you, tell us about your worst migration mistakes that you learned the hard way.

Miranda: I think the worst one is basically when people are working on the same time on content and you're trying to do the redirects, cetera. And then things become really, really messy or people still change your redirect file and don't check the URLs and then they're redirecting to 404s and then you need to fix them again. yeah.

So that's why I have a way of focus on isolating things. Like, okay, this was your time that you have time for content. And now we start a migration. So no content changes basically.

Noah: Now, so let's rewind to that five month pause. Is that a content freeze for that whole five months? How do you think through that, I guess?

Miranda: It's not a full content freeze, but it's basically they wanted to clean up pages from the site and it basically, so they wanted to delete 30 % of the pages of a website that has more than three, five million URLs. So that's quite a lot. Basically means that they are deleting one to two million URLs of the site. And so that project, we want to have finished five months before.

They could still make changes to the website until the view content freeze, until we start moving the pages over. But that project about deleting the pages in large scale, that needed to be completed five months before we start.

Noah: Got it. Okay. Ooh, and you're already, it's funny. Your next slide was like totally in line with where my head was at about things going wrong.

Miranda: Yeah, so the very thing, question was like on the day of Go Live or the day after, like everything goes wrong, like all the traffic is gone. And then I was like, but have you checked your analytics code? Like is it measuring correctly? And so that was if you, if somebody says that traffic is down across all channels, it's like, okay, there's an analytics issue and there's not necessarily an SEO issue because

If it would be SEO, you would see SEO traffic going down, but the other channels still relatively stable. But if everything is down, then know, okay, there's something with the tracking. So that's one, making sure that the tracking, or basically a pitfall, that the tracking was not accurately done.

Another thing is I got like...

Search is going down, cetera, but paid search and SEO combined. And then I was like, but did you activate your paid campaigns? And then I'm like, So it's also like when a new site goes live, during the migration, we put it on pause to make sure that we don't send any traffic to maybe like temporary and other available pages.

And then a couple of hours after go live to make sure the site is stable, we boot the campaign on again.

but sometimes this step gets forgotten.

Noah: A lot of folks are not thinking about talking to the paid team and that can be a tragedy and a costly one based on how much money we spend on ads at scale. Wild.

Miranda: And I'm thinking about other comments. Just like small things like, sometimes the decline is in a small corner basically. So the robots TXT something, it's a very tiny thing, but if you miss that step, it has like a, it has a large impact. There's a bit of the comment pitfalls that I came across.

Noah: That's awesome.

Miranda: And I think, yeah, that was it actually for my end.

Noah: Ooh, questions. We've got some stacked up in our channel here. Okay. How do you justify the cost of maintaining 301 redirects indefinitely? Oops. Can you stop? There we go. Perfect. Nicely done.

Miranda: Yeah, yeah. So I try to refer as much as possible to Google sources so they know it's not just for me or from a SEO consultant, but it's Google who says it. And Google has their guidelines that the redirects need to be live for one year in order to properly transfer the SEO value to the new URLs. But for any SEO that, for example, any pages that have backlinks, that actually could be indefinitely.

Noah: you

Miranda: So in terms of like trade-off, because sometimes a lot of read-backs can slow down the site, is making sure that at least live for one year and then to check like for which URLs still have backlinks to keep those on, but for example, remove others.

Noah: It does seem like there's been a lot of chatter in the past year about how that one year threshold has been really clearly stated by Google and proven to be effective in testing. Like a lot of folks have tested that and have found that that one year timeline makes a lot of sense. Quick question, your migration checklist. Let's wrap up the migration part of our discussion.

I'm going to just show the link for everybody. Do you want us to just kind of wrap up high level the checklist again for us? And I'm going to show it again. So everybody can get the URL.

Miranda: Yep, I will actually share it.

Miranda: Yep, and I will share my screen at the same time.

Noah: What I found interesting so far is just like the scope of the other teams that you're integrating with is larger than what I'd considered. And then also one thing that occurred to me is that doing a migration is a huge opportunity for everything to go wrong. And that's what we always focus on is like the loss part of the equation, but it, what I was hearing from you was how large of an opportunity it is.

Miranda: Mm-hmm.

Noah: to build on your relationships with the other teams when it goes right. And that you're gonna make them the heroes in the process, right? If they execute all their steps right. And it's an opportunity for you to build fantastic relationships with other teams. just, that was something that occurred to while you were talking.

Miranda: Yeah. And I also think it's a great opportunity sometimes because especially if the website or the platform is a bit old and currently there are faster solutions available. So it's also an opportunity to make things faster, which of course is also good for the user. But I also find it nice because it's basically you do a project, it goes live and everything done, and then it's like, okay, next. So it makes it very

dynamic process. it's basically like with all of them, the product owner, developers, feels like you are together in a roller coaster. And to make sure that everything gets delivered good and to be sure you stay in the schedule for the next one.

But yeah, it's an interesting teamwork together.

Noah: Yeah.

Noah: Nice now on this checklist, just give us a quick high level. We don't have to go line by line on it.

Miranda: Yeah, so I basically had per category. So for example, site setup, URLs, navigation, what are things to keep in mind? So basically, what is the topic and what kind of actions should, would be good to look into? Here, for example, like URL contained lower cases only. If you have a block that gets automatically created based on the title of the article, that for example, words,

That's an option, the way, where it's all filtered out of the URL. WordPress is doing that sometimes, and we also use that to, in our custom platform, to make URLs a bit shorter. So it's just a nice, like, when you review a lot of components or websites, it's nice to have a template.

And at the start of the project, it's good to check, like, are all the items that I want to review, is it in it? But then when you're...

especially if you go in a fast pace, it's very handy to just go down the checklist and see, okay, have I looked at these items and make sure not to forget any important parts.

Noah: Now someone else asked what tools do you use to QA the audit checklist? Seems manual.

Miranda: Yeah, a lot of it is manual. Manual and Screaming Frog, basically. Screaming Frog is the ad hoc go-to tool that they use for the crawls. So, for example, URLs are small case. Those would be things or parameters. No parameters are being used. Those are things that would be good to check with Screaming Frog. Other ones like...

The menu is basically available if JavaScript is disabled. Those are things that I would test manually.

Noah: When you're setting up, I mean, lot of the stuff you're doing is for sites that have millions or hundreds or billions of pages. You're just doing small subsections of content, right? You just have representative based on template type or something like that.

Miranda: You mean in terms of the view?

Noah: Yeah, in terms of the crawl, like you're just looking at representative pages per template kind of thing.

Miranda: Yeah, in the first place I would review components. So the accordion component. Is it the way how I prefer is that when JavaScript is disabled, that the accordion is expanded and that the content is visible on the screen and that JavaScript is used to collapse it. That's the way how I usually prefer to have the accordion programmed. So...

Noah: Mm.

Miranda: By doing that, as soon as the component is built, or a couple of stories have been built, and then they say, okay, now it's ready for review. So that's the first level. And then when basically a page template is ready, then I review the template. Okay, is the clean code, is the header structure correct? Is it following the semantic order? So nicely from H2, H1, H2, if it's subheader, H3.

And then basically this one is as a final check before the site goes live because this basically includes all the items.

Noah: my goodness, keep going, how many rows are there?

Miranda: This is 88. But I must say, I actually wanted to update this for the call. And then I felt like it's different for every company. Monst is a job board. So we have the job posting markup. We need to make use of the indexing API. So I think it's good to use this as a base and then see based on your situation, what do I need to update or further expand. But hopefully, it will give you a starting point.

Noah: Yes, there's just 88 steps. I love it.

Noah: So you just touched on one of the things I wanted to talk about. So I started experimenting with the indexing API right when it came out. What was that? 2018, 2019 and. Oops, there we go. And what I found was that for the stuff that doesn't qualify, you know if it's not a job board etc Google will still go to the page when you submit it. So talk us through what it's like.

Miranda: Yeah.

Noah: to work with for a site that meets the guidelines? how much content can you, like what percentage of your site can you actually get indexed when you can leverage a tool like that?

Miranda: I actually find it still a bit disappointing because Google basically prefers for static pages, so pages which doesn't change a lot, like urls.com in and out, just the site maps. But then for individual jobs, sometimes they're live for a couple of hours, sometimes just for a few days. So it's very short. So if we would only rely on the site map, Google wouldn't know of these jobs in time and also doesn't know when to delete.

So with the indexing API, we say to Google, OK, these are new URLs that have gone live. Or we can say, these URLs have been deleted. Delete them from your index, and then crawl them again in terms of crawl budget optimization.

And we had from the jobs that we're sending, which is sometimes like, or at least the requests, like maybe it's like three, four millions a day, that not even the half gets crawled, basically.

Noah: my goodness. And there's so much flux that so that sounds like a really difficult challenge to stay on top of in terms of the volume of URLs that are going live and the volume that are falling out all the time. So how do you manage jobs that are? Let's say we filled the job or like talk us through the content removal process for job board.

Miranda: Yeah, so for any static page or page that has stayed live for a while, so not just normal pages, we use a 301 redirect to transfer the SEO value. For jobs, we use a 410, which basically means the page is gone. Google can delete it. And because otherwise, if we don't do that, then basically we would end up in like millions of 301 redirects. And so we thought, OK, we won't use that.

Noah: Mm-hmm.

Noah: Yep.

Miranda: So for expired jobs, we use 410s. So indicating this is not an error, it's not a 404, but the page is gone, you can delete it.

Noah: Yeah.

Noah: Can you share with us something that you never expected to be the case or something like that, that you, like what have you learned that's new and unique working at a site like Monster? Like what's your biggest learning in the new role?

Miranda: It's a different volume perspective, basically. So in Ransot, was like the outside of like 300,000, 400,000. But here it's like Monster and they recently have been merged with Career Builder, which is another like very large beast in the game, basically. So we have both websites have something like maybe like 8 million, 9 million pages.

Noah: Thank

Miranda: And so in those kind of websites, the crawl budget optimization, that's really an issue. So which jobs do we want to favor above other ones? And are we sending all our new jobs to Google? So that is kind of like different type of optimization than, OK, what can we do to improve the content on our pages?

Noah: I was wondering about that because based on the scale of content, prioritization is a must. And I was wondering if you like, hey, I'm going to focus on the jobs with the highest value. are you thinking that way? Like you're pushing presidents, vice presidents, directors. Like, do you go top down in terms of prioritization or how do you think through that?

Miranda: In terms of scalability, so if I do this, how many pages or markets are impacted? So one simple thing is increasing the number of jobs that are displayed on the page. It's basically one setting. It's impacting all the pages across all markets. So those more from a template perspective, those are more the optimizations I look at.

Noah: Awesome. Speaking of learning, you're kind of driven. I was looking at your, your personal website and you've been crazy into sports your entire life. it sounds like it's something you got from your dad. And when I was reading through that, I was thinking a ton about my own childhood. Like where does, where does that inner drive, where do you, where does it come from? Like.

Miranda: Yeah, yeah.

Noah: How do you, and what has been the impact on your whole life of that like development of discipline and how can others tap into that? Like some of us just by nature, hey, I love a good checklist. I wanna execute the checklist. And others of us are constantly like pushing ourselves to become better and are crazy, crazy curious. I wanna get at where that comes from. Can you make us smarter about it?

Miranda: So it to go way, way back is basically when I was like 16, 17, I was like delivering Chinese food on the scooter. And I was just like riding and I was thinking like, like, I can live like like this, but is this really like what I want? What I would be most proud of? And then I thought, no, this is something that I need to change. It's not that something that my parents were like to...

normal school or something, they didn't do masters or anything. So I thought if I want to change it for myself, I need to pursue it myself. So that actually changed a bit to switch and to say, okay, I'm going to try to reach the max and just try to get everything out of it.

And that basically, then when you come to a point and it's like, okay, what can I do next?

I saw a master, I was looking at the 10 best MBAs in the world. And then I saw in the top 10 was IE Business School in Spain. And I was like, that looks really cool.

And in Spain, I used to work in Spain for two years. I felt like, that's also really cool. So then I got into that master. At that time, I was a bit more impulsive.

I was working at a company and...

The master started a bit earlier than our... So normally in the Netherlands or university start in September, that master started in May. And I still needed to finish the thesis of the bachelor and everything.

So I was like, I don't have time to work. So I went to my director, I said, I need to resign. And I need to resign like next week, because otherwise I won't be able to start this master.

And she was like, you can't. I was like, I have to. So I was telling her the things. And then, okay, I resigned with a very short time.

And then I thought, okay, now I'm here, I'm doing my thesis, but I still need to have income. So that was actually the same time that I started my own company, just we're doing like freelance projects to make sure that while I'm doing the thesis, at least I can have a bit of side pocket money to get ready for that master. And then when I was done with IE, I was like, like I want to keep developing because I think if you keep learning,

Miranda: and that will make you a very strong asset, independent of how the market will go. So, and that was basically how I got Harvard on my radar. And I was like, that's really, really cool. We only see Harvard in movies. And yeah, so that's eventually how I landed there.

Noah: Hmm.

Noah: My dad also went to Harvard and his way of targeting it was, what department could I actually get into? And he did, he did his masters in public adminis, no, in higher education management. So he decided to go to the ed school because he felt like it was the only one he could actually get into. Which is super funny. okay. So.

Miranda: nice.

Miranda: Okay,

Noah: As you've grown in your career, this inner drive to keep improving has led you to improve your tech skills, your hard skills, but also your soft skills and working with other people and collaborating. And one of the things, it's not unique about your life story, but you've lived in so many places. Like you've lived in the Netherlands, you've lived in Greece, you've lived in Spain. Where are we talking to you today? And tell us a little bit about what it's like to live there.

Miranda: Yeah, so I'm currently based in Dubai. I moved nine years ago to Dubai. I was sitting at home in the Netherlands looking out of the window. It was very cold. And I was like staring and I used to do like I did the traineeship in Spain before and I was like, no, this is not how I want to continue. So then I applied on anything that had to do with digital marketing, analytics, SEO, paid search.

Then eventually one agency decided, okay, you can work for us in Dubai. So that's how I landed there. I also lived in Egypt. I stayed in Morocco for a bit.

So what I find is that actually a lot of the people are the same with cultural differences, but everybody like wants to...

meet people, wants to be respected, wants to have conversations, hang out. So I think it's also, I think it's very important to understand culture, what is culturally appropriate. And I think it makes it for, so for example, here in Dubai, it's not always common that a man gives a hand to a woman. So for me, when we go to a meeting,

We don't stick out our hand first basically. We just see like is the man doing that or not? So I think like being open to that, being responsive and know about these things will help in communication with a lot of different people.

Noah: Awesome, we're getting to the end of our episode here, Miranda. There's so much to get into, but I just want to tie this up a little bit. Any special causes that you want to shine the light on, that you want other SEOs to know about, anything that's really important to you on a personal level.

Miranda: On a personal level, it's... Yeah, I don't want to go too political. But basically making sure, like having respect for each other, seeing each other as humans and treat people how you want to be treated. I think that's... Like have empathy for others, being respectful. think that's...

Noah: I love that.

Miranda: like really from a human level, human connection. I think that's a really important thing to do and to also, yeah, emphasize with others, learn more about each other's backgrounds. I think that's something which I find important.

Noah: I think my thing is like lead with kindness, you know, and I fail at that a lot, but like, I'm really working on that as my own personal thing for 2025. It's just really being kind to people and really digging into trying to help and really putting everybody first. Awesome. There are going to be some questions in the live event.

questions channel if you can check in there in slack and get to them. There's a great question, but it's gonna take us another five to seven minutes to talk through it. So if you could get into that into that question in the slack community, that'd be great.

Just to tie this up in two weeks, we've got another guest, which will be pretty fun. I got to know Rand Fishkin in the past year and a half, a little bit. And so I'm really excited to have him on as our next guest.

February 5th and will keep everybody updated in the slack community about that in the SEO community. Miranda, it's amazing to spend so much time with you. I'm hopeful that we get to actually see each other in real life at some point at a conference and.

We're going to end today's episode, hopefully without a hot mic situation like last time. So at this point I'm going to stop the recording and we should go.

Miranda: Thank you for your patience.

Miranda: Yeah, yeah, that would be great.

Noah: and actually leave the studio room. actually don't leave, mute yourself. So for everybody else, great spending so much time with you. Be good and catch up soon.