This is the first episode in a series of podcasts from Skittle Digital. In this episode, James is joined by Jimmy Hartill, of RAJA UK and JimmyHartill.com for an exploration into working in SEO in-house vs. agency-side, selling in SEO to stakeholders, the importance of links in 2023 and how to leverage AI to benefit Google’s QDF feature.
This is the first episode in a series of podcasts from Skittle Digital. In this episode, James is joined by Jimmy Hartill, of RAJA UK and JimmyHartill.com for an exploration into working in SEO in-house vs. agency-side, selling in SEO to stakeholders, the importance of links in 2023 and how to leverage AI to benefit Google’s QDF feature.
This is the first episode in a series of podcasts from Skittle Digital. In this episode, James is joined by Jimmy Hartill, of RAJA UK and JimmyHartill.com for an exploration into working in SEO in-house vs. agency-side, selling in SEO to stakeholders, the importance of links in 2023 and how to leverage AI to benefit Google’s QDF feature.
This is the first episode in a series of podcasts from Skittle Digital. In this episode, James is joined by Jimmy Hartill, of RAJA UK and JimmyHartill.com for an exploration into working in SEO in-house vs. agency-side, selling in SEO to stakeholders, the importance of links in 2023 and how to leverage AI to benefit Google’s QDF feature.
James: (Laughs) Fantastic. So it’s fair to say that you’ve had experience working both in-house and agency side, which isn’t isn’t as common as you as might think.
I think it tends to be that people go into an agency, and then they spend their career moving up the agency ladder, in a way specialise in working within the agency setting. And similarly, it works the same way for in-house people. So what do you think the main differences are between working in-house and in an agency?
Jimmy: Well, in an agency I don’t want to say there’s a lot more work but, I feel like when I was working agency-side, I had to do a lot of educating, and it was slightly more of an uphill argument because when I’m talking to a company or a new employer, I’m going to try and explain the benefits of what we’re doing why we should do it, and lay a lot of the groundwork down. To explain what we’re doing and why, and how it’s going to work.
And there’s also some element of buy-in, like you need to be charismatic enough for someone believes what you’re telling them. Whereas when you’re in-house, there’s still an element of that, but you’ve been hired to be the “font of truth” and meant that this for me particularly has created a nice dichotomy. I’ve got an agency trying to pitch stuff, and I get it immediately, and I can rationalise that to our bosses, and a lot of the “having to win the argument” or gain the trust is already there, because they know, I know what I’m talking about.
It just feels like it streamlines a lot of decision making processes, at least on the local level, because you’ve still got to go higher up the chain a little bit for the biggest national companies. But for a local company I imagine it’d be quite helpful having an in-house SEO specialist (or at least well-educated individual) working there.
James: So do you think that other people in your role who haven’t had that agency experience still have that same authority to kind of carry forwards on on behalf of their agency?
Jimmy: I think it depends on what they’ve been told, because it’s kind of a problem of being like a general who got through the academy versus the soldier who was on the front line getting promoted. Because when you’ve done the actual work you understand it a lot better than when you’ve only been told what works. So a lot of the guys who just stay in-house their entire career don’t always have that. This is my perspective, by the way. They don’t always completely understand the nuts and bolts of what works and why and like some of the nitty-gritty of the finer details, but they understand what they’re told and they can just repeat it.
So if someone were to say tell you that buying links is totally fine, or you haven’t been through anything like the Panda update, or the Penguin update, you won’t understand why [the thing you’re being told] is not the thing you should do – having to do a recovery and go “oh man that was terrible, that was an awful time, please never again”.
James: I mean it’s kind of like a theory versus practice thing.
Jimmy: Yeah exactly
James: The amount of disinformation out there in the SEO sphere! I mean it’s still so prominent. I can’t remember the last time that I loaded up an SEO website of some sort, a news portal for instance, and found advice I disagreed with.
You know you read something and you think “well that was outdated about five six years ago” or “that hasn’t worked in however many years”.
Jimmy: Yeah it’s kind of nice having the wild west element. I think there’s a lot of people drawn to SEO because they like how it is the wild west. You can kind of carve your own route.
And we haven’t got an established Academy or a University of SEO. Like, a Google certification (they were talking about it a while ago) and I kept thinking “how can Google certify, when the whole point of SEO is that you are going around Google in some way”. You’re trying to find a way to make the most of the algorithm, while they (Google) are always like “oh we love SEOs”
I don’t think you do Google…I think you wouldn’t keep punishing us if you thought we’re doing the right thing.
James: How do you keep up with SEO Trends?
Jimmy: The same way I think an agency consultant does. I read new sites. I know which ones are the good ones. I know which guys to follow.
I went to Brighton SEO and I got to meet John Mueller and he threatened me with more core updates, so I had to leave him alone.
James: [Laughs]
Jimmy: [Laughs] I know right, we’ve already had two this year, can we like cool it down? And he said: do you want a third? I had to say: Bye Jon. I feel I was on brand for the industry in trying to get a selfie with a guy.
Yeah so it’s the same kind of thing really, I read stuff.
We test things when we can. I feel like when you’re an agency you have a wider pool of knowledge to test from but when you’re in house (not so much).
One particular problem we’ve got is that we’re testing stuff, but it’s across different European countries, so I can’t know what works in one demographic (like in Slovakia) is going to work the same in another (like Italy). And it won’t work the same here, so even if we get success stories (we can’t always use the replicate elsewhere).
We have much wider pool, even dividing by sectors in the same country, so we’ve got good consistency.
But the way Google treats things in like different areas of Europe is so different. It meant information I obtain has to be passed on carefully.
James: Oh sure we’ve always seen some versions of Google be somewhat ahead of other versions
Jimmy: Definitely in the EU.
James: Yes, not always for the best. But you know we’ve certainly got a more advanced algorithm.
Jimmy: Yeah I think it’s just for language rollout because they’re all out in America first, and they can just roll it straight to UK, and then we end up getting the updates really early. And then it’s almost like we’re the guy at the front getting shot. Then the guy at the back and go “oh man bullets are coming this way, we can duck now”.
Meanwhile, that’s great…I’m still getting shot. Can we please just give me a flak vest? Anything?
James: It’s interesting to me what you say about tests. Because one of the things that we always tried to do in our agency setting is run tests. You can create dummy websites and you test different link building techniques on those sites. You can never really have a control (site) because of the Google environment, and so many variables (but you can look for correlations).
But I guess you can’t can’t always get the resources? Or budgets are too tight to do that in-house?
Jimmy: yeah it’s particularly tricky when you only get one website. You can’t really AB test unless you’re doing it on that site. So you can’t do anything really “fringe” because your bottom line gets hit so hard (especially if it goes wrong).
(For example) I want to move our top performing category, and change it around a little bit. They’re going to say right that’s going to cost you something in five figures (just to test). I can’t really afford to pay you back for that so I guess we’ll let that one go.
But I’m still testing in my own time obviously because every every SEO tests.
Because until we get the entire algorithm leaked to us somehow like happened to Yandex, which is still hilarious to me. We’re not going to know what Google’s actually ranking for so we can’t really check it. Although the Yandex algorithm leak was really interesting, did you see that?
James: Yeah, I think we read through what was effectively “the formula” but the percent of traffic from Yandex for our client base was minimal to say the least.
Jimmy: Yeah it didn’t affect anything for us because we haven’t got like a RAJA Russia, but I did quite like looking at some things that they had, like (I think it was) bounce rate (or something) in there. Something in there I didn’t expect to see. I was like that’s interesting…is that in the Google one, or is that like a Yandex original?
James: One thing I spent a lot of time doing (back in the day) was reading through the Google patents. Because there’s a lot of stuff that they’ve thrown in there. I think (some of it) just for the sake of it.
So they’ve come up with an idea a ranking factor, shoved it in a patent and got it pushed through. But actually they’re probably not using it. Or if they were using it might have used it for a short time while testing it, and decided that it didn’t work.
Jimmy: If I was Google I would definitely do that.
I’d put a patent on Google Eye Track.
And then all the SEOs would (ask) “They track eye movement? How do they track eye movement?”
And then (they would all be) trying to work out how that works.
James: And (Google) would sit back and wait for the impending the impending Twitter storm…
Jimmy: Yeah, and all the LinkedIn posts about “What Eye Track is”
James: There’s always a lot of speculation in the SEO industry but I’d be curious to hear your take on what you feel the challenges are for 2023 and beyond?
Jimmy: I mean AI is the big one everyone’s dealing with.
Especially with content. Because everyone’s realised with AI you can create a level of content immediately. And for e-commerce sites that’s fundamentally what I am doing at this point.
Merchandise copy at speed and in bulk. The way is can scale is kind of a game changer.
But if everyone does it all at once using Chat GPT specifically, we’re going to end up with everyone having “not exactly” duplicate content but close enough.
If content is king, and suddenly everyone has access to the same level of quality content, it’s going to mean content is no longer king.
And we’re back to the age links.
So I’m kind of curious to see how it pans out.
I think the fact that there is more than just Chat GPT out there in the marketplace now is very helpful, but I’m also Unsure how many LLMS have genuinely got to that level that quickly and how many doing something different so…I don’t know.
We’re experimenting with AI, everyone is. Everyone’s working with AI content in some capacity.
James: I think that’s interesting about the AI. I think my kind of my kind of approach to leveraging it has been to use it more like a almost like a personal assistant.
So if I need to write a code snippet… I can describe the code that I need and hit go. And it’ll write that code for me, and then all I have to do is sense check it.
Whereas it might have taken a couple of days to write what I needed. AI can produce it in seconds, or minutes. All with just a little bit of help and not a nudge in the right direction.
As a tool, it’s really down to how you use it I think.
Jimmy: I’ve been using mine for editing quite a bit. We’ve had existing content copy we need to refresh, so I’ve been putting into Chat GPT, and asking: rewrite this in an active voice retaining the the headings and subheadings, and it usually outs out something which is similar, but different enough that I’m hoping it gets QDF.
We’ve had mixed results from it because we’ve had some crawling errors occasionally, now and then. But it’s generally been pretty good for uplift.
James: QDF – that’s really key. It is something that I remember designing training packs to teach new starters way back (like 2012).
Jimmy: What I was realised (sorry) [to camera] QDF is Query Deserves Freshness, for anyone who’s not super up on all the SEO terminology.
James: I was coming to that [laughs]
So yes, I used to put that in training packs, and it seems like the industry has only just caught up with that as “a thing”. It wasn’t something that I discovered (by myself) by any means. I had people teach it to me when I was starting out. It does make such a difference if you can wait a period of time for your content to bed in, analyse its behaviour in SERPs, and then tweak it and see what impact it has.
I mean it doesn’t take a lot to propel several pages up the rankings (especially) if it hasn’t been performing well.
So yes, I think QDF is still a massive thing. I’m intrigued to sort of see how Chat GPT could help enable that because it’s certainly not something I’ve deliberately used it for.
Jimmy: I took a big project where I had to refresh all the copy everywhere. I guess eventually I saw how this was just eating into my time, day-to-day. So I decided to get the AI to do some. I can test it against what I’m doing and see if it works out all right. And it was functionally the same result, so I thought: alright, let’s just keep doing it.
So I did the whole thing with AI and it just made it a lot quicker to do it at scale. I didn’t obviously get a meteoric rise for everything, but most things that have been there for a while were looking a bit tired or just weren’t the best English (because you know it’s a big website) and for a while it kind of improved everything.
So I felt it was a really good tool.
That being said I also have been trying to push to get my own copywriter producing bespoke stuff which I know wasn’t written by an algorithm that literally everyone has access to…
So hopefully that’s going to be a bit of a thing for us.
James: What do you think about AI integrated into search?
Jimmy: that’s a good one talk about. We’re talking about Bard and how I don’t like Bard that much?
James: Yeah there’s Bard. There’s Bing’s AI as well.
Jimmy: As always Bing got there first but no one talks about Bing. Which is really kind of sad. It was like when they rolled out Bert for Google, and then Bing had been doing it for like several months which is hilarious. Because hey Bing had done it for ages and no one noticed, and they had to tell us they were doing it because no one cared. It’s just it’s kind of sad.
Bing has got Chat GPT baked into it.
I say no one is using Bing but it’s actually a big big channel for us.
So Bing is kind of out there, but Bard has got major teething issues. So I feel like until Google (the market leader) has got something adapted, I don’t think it’s going to be a huge thing yet.
James: And what about Google perspectives, and Google Search Generative Experience.
Jimmy: Honestly, I don’t know. I don’t know what I can do about it. I feel like they’re going to release something saying that AI content, as long as it’s helpful will rank. They say a lot of things.
James: That is effectively what they’ve said. As long as the content is helpful to the end user (there’s no difference) and I think that’s probably right, but sometimes what Google don’t explicitly say, is where the important message is. I think when you take that into account alongside all the other ranking factors around uniqueness and search intent, I think AI is probably not ticking all the boxes at the moment. At least not in its raw format.
Jimmy: It’s like I said, if everyone’s got the same informative content how do you weight who’s the best. Then it’s links again and we’re back to everyone scrounging for links.
Which I am not opposed to, I still think links are king personally. Which is like mild heresy in this day of SEO.
James: I agree, everyone could create equally good content by copying each other or having access to the same you know the same tools. You could all create equally accessible websites. With equally good technical foundations. So it comes down to links really.
Links are in essence a form of kind of user feedback. Votes of confidence is how we always kind of describe them. I think that whilst Google have made strides to say publicly that links aren’t as important or that building bad links doesn’t matter and doesn’t negatively impact…
Jimmy: They don’t “factor it in” (in quotes)…
James: Yes, when you actually really look at it, things like including a paid links in the Google spam report (there’s a section for sites buying selling links)…if Google didn’t care about that they wouldn’t include it in the spam report. And that’s been refreshed recently, and had a lot of coverage in the SEO industry news, but not a lot of people saying actually this indicates or should indicate the links are still a major factor.
The fastest way up the rankings, at least from from my perspective and my experience, and even over the last 12 months has been building links. Not at speed, but at a pace that builds traction against the acceleration of competitor link profiles. That’s kind of the the single biggest influence on rankings once you’ve got the content anyway.
Jimmy: I would argue about link velocity for ages and I still don’t know if it’s a real thing or not. I think the amount of links you get and how fast you get them (must be important) because we ran PR piece ages ago which did really, really well…it did crazy good and it won some awards. And you can find it on Google so I can talk about it. I think it won a Drum award for the agency that worked for us.
I explained it on an internal call to people at one point. We talked about how acquiring too many links too quickly can look suspect especially if it’s like from certain areas of the website. And I had to explain that acquiring 60 newspaper links all at once is totally fine. It was all the major newspapers and a bunch of extra spin-off sites but acquiring 60 of those all at once doesn’t look suspicious.
But if it’s 60 links from places where it’s unclear why they are linking to you (that are the issue). It’s those kinds of conversations. Which again you can still do totally naturally. It just looks weird.
James: And it ties back the QDF thing. I forget exactly which one, there’s a Google Patent that talks about comparing link velocity to real world search trends. So for instance if it sees “a story” starting to build traction, and the search trends are increasing, and the volume of people looking for it in real time (is increasing) then it’s not going to flag an increase in links as unnatural.
So I think link velocity is important. I think spikes are probably less indicative of spam in isolation.
Jimmy: Yeah
James: I think there’s plenty of other factors that you could kind of take into account quite quickly to in order to identify a spike in in back in individual links as a spam signal anyway.
Jimmy: I think that’s kind of the key part about it. Like when the natural links are acquired, and in line with what you’re saying and in line with everything else like with the current trends does help to really see where things are going.
James: It’s maybe difficult in-house, but are you still using the disavow tool?
Jimmy: I use it for the essential ones.
At work, I describe like flossing. So it’s like you don’t need to floss technically to keep your teeth clean. But I feel better for it.
(My favourite is) when you get the spam keywords. I had to add a few of these – the usual gross assortment of rude ones.
My favourite one was the actual words “bad keywords”…there were these links with the anchor text being the words “bad keywords”.
I could see in my mind’s eye exactly how it went:
They plan to spam attacked this website, and the spammer says: alright boss, what do you want the anchor text to be?
And the boss says: you know, just bad keywords.
And then they actually literally did that, and now I’ve got evidence, because I disavowed it. I was so proud that they were that dumb.
James: That’s a great story. I love it so much.
Jimmy: Right?
James: I did think about kind of collating all the crazy stuff that we’ve seen over the years and putting it into a something (it’d probably fill a book).
Jimmy: (There are the ones that have) got the receipt number at the front where they bought it from “the globe” and you have to go “DISAVOW”.
Can we single out that site?
James: I think considering the amount of spam that that one domain generated, yeah…
Jimmy: Free tip, if it’s got a big number (at the start of the anchor text) it’s from the globe. You can disavow it. It’s a spam one and someone’s trying to get you (with a negative SEO attack).
James: Again Google coming out recently and saying you don’t need to worry about the disavow tool…you know…yet, I still see uplifts following disavow activity.
And we always have since that tool came out, what back in 2013.
Jimmy: I kind of treat it like changing your own car battery. Theoretically you can do it yourself. But you can avoid being electrocuted by getting a professional to do it for you. So it’s like maybe don’t disavow yourself unless you’re super confident you know what you’re doing.
Get an agency to do it because it’s definitely still helpful. But I would say get an agency for it or know what you’re doing.
I might try it for us, but I wouldn’t say (to anyone in-house) to freehand or do it yourself.
James: Tools like Semrush (and plenty of others) that deal with link toxicity scores…
Jimmy: Hmm, not so useful…
James: ..yes, well similarly, I haven’t found (these to be particularly helpful). There’s a number of ways that I found that we can speed up that link auditing process. Some I’m quite proud of because they’re methodologies that we developed or some proprietary tools and software that we use to speed up the process.
But I haven’t found anything really that is a good substitute for a manual review. Because you can have some perfectly good anchor text, perfectly good Domain Authority score or whatever metric you want to use and then you do a check and the site’s been hacked and it’s got adult content on it, or there’s hidden links in the background and you think okay…(why wasn’t that picked up by my enterprise level tool)…that can’t be allowed to stay.
And you kind of evaluate (each link) in context of your own your overall link profile.
I think well actually it’s not a link that particularly carries much relevance to me or is going to pass much authority compared to what I could be building, or what I could be earning.
Jimmy: At the end of the day, you can always go back and delete the line if you want to try and put it back in again.
James: Yeah I think it’s processed in “real time” and has been for some time.
Jimmy: I wouldn’t advise anyone did that, but you know if you do make a mistake you can fix it. It’s just maybe not for a bit of money website.
Jimmy: I think having a variety of content is probably quite a big thing for Google now.
Like having video content, infographics, alt text and images…everything to make the content more accessible.
It’s just like a key way for how they rank stuff now.
James: I agree completely. I think especially embedded video. Including an audio transcript. Like you say, an infographic. As many different media formats that suit different learning styles.
Because not everyone digests information in the same way.
So you know can have a video version of a podcast. You can have an audio version. You can have a transcript. Then you’ve kind of ticked three boxes there.
I do think it’s really I do think it’s important and also Google trying to match to some degree as it develops more intelligence, the ability to match you to your preferred content format and the medium that makes most sense to show you.
So you search a particular topic and Google knows that you tend to watch more videos, that is what they’ll populate the results with.
Jimmy: It’s also like answering the question isn’t it? If I’ve got multiple choice answers where (Google can say) here’s the thing you want to find out in three different formats…this one’s a video…this one’s musical…this one’s in the form of an interactive Flash game…
James: Then you know it wouldn’t be hard to measure the results of that on a pseudo-individual basis.
Jimmy: I feel like flash games are showing my age a little bit. Their life is dead for a long time.
James: So here’s another thing I wanted to get your your view on really. Site speed was such a a buzzword for a long time. The initial wave of (even) Google released tools, or (tools from) ex-Google employees to measure individual parts of the page load…
Jimmy: Yes, it was the “crypto” thing for a bit for SEOs where everyone was a page speed expert for a couple of months, and then just vanished again.
James: It was always really hard because there was only so much you could do, and you could spend weeks and weeks refining page load times and see little to no gain. And then Google rolled out Core Web Vitals which gave us three or four areas to focus on and we found that by just going after those areas. you’re able to boost rankings pretty quickly if you ticked all those boxes. Whereas if you didn’t, you actually kind of had a decline.
Jimmy: It was kind of awkward to explain to people because it’s kind of like explaining internally (the question was put to me) “is this a new ranking factor?”
I was (trying to frame it as) more of a contingent. Because if your page isn’t quick, and you’ve got all these loading problems they’re not going to rank you well (regardless of your content, links). But it’s not like it’s a direct ranking thing where your first input delay is factored into the algorithm (where faster sites will always perform better). It’s just if your first input delay is kind of crap, they think you’re not great, so they won’t up-rank you. So it’s like it’s not a ranking factor. But they’ll down-rank you for not having it, rather than up weight you for having it.
That was kind of my point of view on it. I was trying to explain to people across the business and there was really a lot of conflicting information around as well.
I was saying, it’s not like H1 tag importance for your ranking. But if you don’t have it we’re going to get penalized effectively.
James: Yes it is more of a minimum standard isn’t it.
Jimmy: Yeah I think right.
James: And it’s not just site-wide as such, because you can investigate groups of URLs within search console.
But if you tick those boxes you’ll rank, if you don’t you quickly slip down.
Jimmy: I feel like it’s less like you get up-weighted for it. This felt like it was only a negative downright kind of thing. Like it’s just because you don’t meet these requirements so we’re ranking you lower versus (the previous perception) “ah you have your speed of 97 and there’s this one at 96, so you’ll go higher than them”.
It was just like you’re failing the test, and they’re passing the test.
James: I think that’s better in a way though. Speaking to web development agencies (they’d tell me) one of their biggest challenges was when (SEOs would demand site speed fixes) and they’d say “we’re spending all this time on improving site speed but (the stakeholders) are not seeing the improvements (in rankings). But they’re telling us what we have to do, and you sort of think okay well at least now we only have these these three/four factors, and you only have to worry about meeting those those requirements – everything else on top is a bonus from a usability perspective but as far as ranking is concerned it’s it’s made things a lot simpler
Jimmy: I feel like what I should have done was go to the web dev team and just give them a modem and say you’re only allowed to use the internet this fast, and if you can’t edit the website it’s too slow. And that would have been enough to get across how you should build the site. Instead it’s just everything is baked in JavaScript and I’m like “nope” because I get why you’re doing it because it makes a really nice friendly used experience…except it takes like actual stopwatch time to load the web page.
James: Do you think that’s partly as a result of Google better understanding JavaScript over the last few years? It’s becoming a bit more of an acceptable thing (it’s always been acceptable but something that is not necessarily entirely to your benefit or detriment).
Jimmy: I’ve always had the mentality of it needs to be in HTML. It’s quite like making a meal, it’s like you can have your carbs (HTML), your CSS is your meat, and then your JS is like a sprinkling of salt on top.
And if your meal is half salt it’s gonna be hard to eat…
That’s how I treat the crawler…if the meal is half salt, you can’t eat it.
James: [Laughs]
Jimmy: But it’s gotten a lot better with the whole Second Sweep thing…it does mean that you can actually have the meal be salt and that Google’s going to just knuckle down and get through it.
But I’m like “just make it easier on him just put some of it on HTML”.
It hasn’t all got to be dynamic. Do a few things so it can digest this quickly.
I used to just toggle JavaScript off and say if the site doesn’t work we have a problem, but now it it does still index stuff. We’ve tested a whole bunch and there were some links in Java and Google did manage to crawl them.
It took it a long time but did crawl them in the end, and it did rank them properly so it did get there.
It just means any big changes you had to wait a lot longer to get the results. But I don’t know, it’s an ongoing argument happening with my Dev team around how much JS we should use.
James: You’ve had experience working on relatively small sites, and now impossibly large sites.
Do you find that the index lag is particularly a hurdle?
Jimmy: It’s an ongoing discussion. I’ve got a bunch of projects on the go but one is to improve the site layout because the deeper you have to go to get to stuff the more awful it feels.
If you bury stuff on a small site you’ll still get to it fairly quickly.
Now if things are 4 tiers deep and it’s not a top category…I have to submit this (manually to Google). I’ve got to go to Search Console and manly submit URLs when new products launch so they get found quickly.
I’d do that anyway to make sure, but it would be a lot quicker if I just knew Google was going to get there.
James: If you have the cruel budget, and not that I’m saying that you should sit there and manually submit everything…if you’ve got the cruel budget then it should level out.
You know when you’ve got a new, or smaller site and a low cruel budget, you create a new piece of content and have to wait a couple the weeks for it to be indexed and ranked.
Whereas on a large site you should be able to publish something straight away, and have it appear (in SERPs).
Jimmy: I would really appreciate if Google could tell you what budget you’ve got. I know they can’t but it always feels like I’ve not got enough. Like I’m running out before the new stuff gets found. So I have to make sure…so link from the home page…link at the very top of the home page…and once it’s found then we can put it in the taxonomy properly.
Then it’s fine, but all your new products need to be on the home page just so they get found quickly.
James: I mean that’s very that’s very good advice, but you can kind of approximate your cruel budget can’t you? Because you can look at the stats for the last however many how many days, weeks, months…and then take some sort of average.
Jimmy: Occasionally you’ll get stuck with things like…if someone gets stuck in a loop somewhere. You have to work out exactly where it is very quickly and it could be one category (of thousands). It’s got like a broken link, and you don’t know where it is. But it’s going to get things stuck in a loop and you have to fix it or else (the crawler) just won’t get down to anything beneath it.
Yeah big sites are big sites are crazy. I think I saw a graph on LinkedIn for this. I can’t remember which agency it was but they had a graph where they had content versus technical, and importance. And it was, as the site gets bigger and bigger and bigger, content gets less and less important, and as it gets to the huge size, technical becomes super importance.
It’s how you structure it, and they had a thing about links as well.
I think links are super important the whole way through because you need them for everything.
James: I suppose when you’ve got to push a technical change through on a big site and there’s already a development queue and that resource has already been taken up, you might not have that change made and impacting the SERPs for quite some time?
Jimmy: We’ve got a big platform for every subsidiary. So you have to make sure everyone’s got the same if you make one big systemic change.
You’ve got to say “right so I need you to do this 21 times”.
And they will say “right it will take a couple of months”.
And I will say “I was hoping for this afternoon but all right”.
So that’s why they don’t like me very much sometimes the web team.
James: I mean that’s always a challenge I suppose. It’s partly prioritisation but also there’s so many different stakeholders all pulling in different directions, and they all want the resource for their own projects, or you just can’t win, so it becomes about what you can affect and I think content if you’ve got access to the site to put content on…great…you know you can publish content.
Otherwise we do come back around as you said to links.
Because you can do that without having to necessarily change anything on the site.
Jimmy: It’s the easiest one to do. It’s the hardest one to actually get material for but the easiest one to do without permission.
If you’ve got full bureaucratic control for technical, content…until you’re perfect…great, then do links.
If you’ve got loads of red tape then just do some links in the background because no one’s going to say you can’t build links.
Just don’t buy them from some random guy on LinkedIn sending you messages because that’s a bad idea. I don’t want to out anyone particularly but if someone’s on LinkedIn offering to sell your links and they start a message with “hello dear” just don’t… just don’t…
James: Well they’ve become known as hello dear links haven’t they? I think there will always be some kind of transaction in link building. It’s better if it’s not purely money but a transaction of value.
You know, if you can offer a piece of content that is going to interest the audience of the publication or website that it’s going on, then I think that’s the main thing. There’s no uh you know there’s no reason why that kind of transaction can’t continue.
Jimmy: I’ve always said make the rule don’t buy links. Then you have exceptions to that rule in certain circumstances. I try to make it the rule so that the exceptions are exceptions and not the baseline.
James: The most successful link building campaigns I’ve ever been a part of have not relied on buying links. There have been PR campaigns (I won’t say stunts) but we’ve had PR shots that we’ve put out that hit really, really, well.
You get a couple of nationals, and then suddenly there’s 200, 300 or 400 smaller sites taking up versions of that content (with no extra effort on our part).
You can’t plan for those necessarily. If you plan to make something go viral it inevitably won’t.
I think when they do happen, you see the benefits…and long-term benefits as well.
Jimmy: I feel like the hardest one to get buy in for, in link building, is digital PR. Because it’s it’s the best method. It’s the most eco-friendly method of building links without accidentally going to a toxic landfill.
But it’s also the least reliable.
So I ran about three for a different company where the first two did OK, they did acceptable. They got like one National and a few local links. 10 or 12 links total.
Then the final one hit the Zeitgeist perfectly. It was during Covid (the great content creator for everyone)
It got 15 Nationals which I think is an actual number off the top of my head. It was loads and they did really, really good job.
I’d love to say I can do that every time, but I definitely can’t.
James: I don’t think it’s wrong to have (link volume) as KPI…but to have it as a hard deliverable…where you have to get this many links from a PR campaign (that isn’t right).
The stakeholders have to understand that you’re going into this with an element of “this might not work”.
But if we’re going to do, say, four campaigns over the course of a year (one a quarter) then there’s a good chance that one of those four is going to hit, and it’s going to hit big. And we’re going to get enough links from that campaign to have made up for the investment in the others.
And you just have to have some sort of experimentation I don’t think you can really guarantee that every PR campaign is going to land.
Jimmy: I feel like SEO deliverables are always an awkward conversation anyway because it’s like nothing’s ever guaranteed.
I always said treat SEO like a philosophy
Someone would call the copy on the page “SEO copy” and I would say “no we’re not calling it that, we’ll just call it content”.
Just call it product copy, because it’s not just for SEO. It’s for the users, for everything else.
Just make sure it it is optimised…which is the SEO part…make sure your website is built right…get these links to get the get the brand out there and if we get links back to the website that’s brilliant but if they’re reading about you in the newspaper, that’s already a win. You’ve got a dub out of that, if you get a link out of it and you get link juice and you rankings are up that’s brilliant you’ve still got a bunch of readers looking at it, so yeah I feel like SEO and KPIs are a strange blend of stuff.
We need something to track it with but also it’s not like PPC where you just buy money.
James: I do agree with most of that.
I think you have to position SEO in some regards as a performance marketing channel.
I think if you’re having to sell in, to say a board of directors and you’re telling them actually there’s no guarantee, we’re not going to work to any particular KPIs, and you’re going to be investing for 14 months (they’re not going to hire you).
But if you’re going in with data, and you say okay here’s the lay of the land. Here’s the things that are influencing the rankings as they are now. Here’s how your competitors, you know, the top 10, have got their rankings AND here’s our action plan AND we should be able to get there within a reasonable time frame then I don’t think that’s kind of at odds with the SEO discipline.
Jimmy: I agree with that. And the way basically I present stuff internally is “here’s what I’m expecting to get out of it” and “here’s the thing it’s not a guarantee”.
I think it is important. I like to caveat it because every time I like to say “we’re going to do something” I don’t want to say “we’re going to definitely make x amount of money”.
It’s like this is my projection because I can’t give you a hard ROAS on some things
James: We create models and things. Not designing an attribution model as such, but say “here’s what we think the market opportunity is”
Jimmy: Yeah, yeah and then like the currency to the conversion rate?
James: Yeah so you kind of build a good model and then you can overlay the data that’s publicly available, so how the competition have got their rankings, how they’ve been sustaining them…and what the biggest influences are on the shifts in rankings as well. Obviously take into account of algorithm changes, updates and things, but you can build a good case.
And say OK – this is the scientific part. The artistic part is how we’re going to achieve this, because that’s content and digital PR campaigns…links…all the other stuff.
And then the reporting bit is the business head isn’t it? Tying that all back into the original vision, making sure that you’re still on track, and adjusting course as you go along.
That’s great, I think we’ve delved into a lot of different topics in a very short amount of time. I really appreciate you coming all this way today.
Jimmy: No of course!
James: You know it’s been great to see you again
Jimmy: Good to see you too buddy, always a good time.
James: I’m sure we’ll have you on again hopefully soon.
Jimmy: Hopefully that was controversial enough, some of that stuff, that will get someone coming in refuting what I said at the end, so that’s all good
James: Oh SEO’s love an argument
Jimmy: Yeah it’s great isn’t it? It’s fine I love it, it’s always about arguing, it’s like who can get the most volume out to say you’re the most correct. It’s beautiful.
James: Excellent well thanks again!
This week, I presented a webinar giving my thoughts on Google’s March 2024 core update. If you didn’t catch it, [...]
New Google core update for March 2024 will take approximately one month Multiple core systems will be updated – including [...]
New Google Ads feature makes reporting automation easier for “non-power users” Reports will summarise performance against top-level business goals Simple [...]
There is no doubt about it: a failed content marketing campaign can sting, and with good reason. It can feel [...]
There is a lot to think about when running a content marketing campaign, but the most important consideration is how [...]
In order to comply with the EU’s new Digital Markets Act, Google adds new search features Rich results enhanced to [...]
Worried we'll send you crap? Don't. No crap. No spam. Only the best insights.
Skittle Digital
China Works
100 Black Prince Road
London
SE1 7SJ
+44 (0) 207 730 9958
careers@reodigital.com
press@reodigital.com
© 2023 Skittle. All rights reserved.