Enterprise SEO on Edge feat. Nick Wilsdon
Hello, Nick. Hello, Olesia. Nick Wilsdon,
consultant and founder at Torque. SEO with 22 years of experience who can deploy
high performance programs across large organisations and global brands. Based in
the UK, area of expertise and enterprise SEO, edge SEO. Worked with such agencies as Rina Media,
Havas, Dentsu. And with such brands as Vodafone, eBay, Lavabeam. Search awards judge, speaker at
conferences and contributor to specialised media. You mentioned edge SEO, and many people…
like not that we are going to go into the basics of it, but very briefly, why
is it a thing? Why is it important? Yeah, it's becoming increasingly important now
as the CDN layer has been developing. So we've always used CDNs for distributing content. So when
you… CDNs like Cloudflare or Akamai or Fastly, they have a distributed series of servers
around the world. And when I say access your website in Mumbai, I don't own your
servers based in California, for example, then I don't have to go all the way to
California for that request. So I access a kind of semi-local copy or a cashed copy
in Mumbai. So we've kind of relied on these CDN networks to distribute content globally.
And
that's given websites much better responsiveness, performance and ability to deal with traffic
globally. Now those CDN networks have been developing and they've gone from fairly simple
sort of servers serving lightweight content, lightweight pages to having increased processor
and CPU capabilities. And that means that they can start to do more than just serve content. They can
start to run programming on that layer on those servers. They can start to manipulate the request.
And we've now added, in terms of computer edge, we've now added the ability to have kind of
storage in these areas. So whether we have set storage, static storage, or whether we have
dynamic sort of database or most storage with KV Store, we have the ability to store information
and do very much more complex logical processes on the edge. And that's really opened up a huge
amount of possibility really for us on web.
We don't have to rely anymore on computation on the
origin servers. We can start to do computation on the edge. And you can imagine that's a
huge opportunity for every part of digital. It sounds very exciting, but can you give some
real-world examples, maybe without mentioning any names or brands, but something like you've
done something and it worked this way. And in terms of ROI, return on investment, it was like
that if possible.
Yeah, absolutely. You know, trying to get it down to understand it's ROI,
but I can tell you that a very simple use case for edge might be redirections because normally
with the redirection, we take the user in, they come through Akamai and they come to the origin
server. And when they get to the origin server, we usually then redirect them to where they're
meant to be going. And that means they can sometimes get bounced back out through the CDN
and back into the origin server. So if you think about it in that way, it's very inefficient as a
journey. That's not a good way of transporting the user. So if you have those redirects on the edge
and you redirect the minute the user touches the Akamai network or the Cloudflare network, the
minute they interact with that network, they go immediately to the destination they're meant to go
to. You can see that this is a much more efficient way of dealing with those user requests. And some
of the advantages we've had on that in real world examples is I've reduced some journeys from 900
milliseconds on a response to 250 milliseconds. So you can see there's a dramatic increase in web
performance by doing that.
So this is great. It's good for the user. It's really good for Google
because Google doesn't want to crawl unnecessary journeys that when it's only getting redirected.
And we can start to do a much better use of our redirects and we can start to manage redirects.
And another use for this might be kind of one to one redirects. When you work with a company,
they're very resistant to put a lot of one to one redirects on the server. They sort of their IT
department will complain. They'll say, how are we going to manage these? This is going to cause a
lot of performance impact. So when you're doing this on the edge, you can literally have millions
of one to one redirects without any impact on performance. So from an SEO point of view, where
we want generally one to one redirects, we don't want rules that send all of those redirects to
the home page or all of the redirects to one of the category pages.
We want to manage lots and
lots of very specific one to one because we want to maintain those journeys really for the user. We
want to send something that's got the same intent, the same product. We want to manage this process
much more in much more detail for Google. So if you get to our eye on that, it's quite difficult.
It's difficult to separate that from all the other SEO activity we would do, but internal and
inbound your URLs are incredibly important for SEO. So yeah. What are the costs involved, both
not only money, but also you need time, maybe some other resources? So what do you need to start
doing that and offering it and making it happen? Yeah, I mean, it can get quite complex from an
in the beginning.
It's fairly simple. I would recommend that people look at Cloudflare because
Cloudflare is accessible. A lot of people are using Cloudflare as a CDN. So it's very easy to
get into that area as a developer or as an SEO, technical SEO, really who wants to
explore how to do this in Cloudflare. So that's, you know, they're very, very supportive
in terms of edge and edge compute and giving free access to some of this capability to the users. So
Cloudflare is a fantastic place to start learning this. It's much harder to get into Akamai, which
is more enterprise. They don't have as many or any options for developers or people who just want
to learn it. It's going to be big enterprise platforms. It's quite hard to get access to
that, but you can certainly learn a lot of this on Cloudflare. In terms of costs, I mean,
there are costs associated with this and this is obviously why the CDNs are doing this in the
first place.
This is a way of them upselling and selling computation in the same way, you
know, computation on demand, essentially, which is what a lot of, they've realized there's
a huge amount of money to be made in offering this kind of computation. So those costs vary a lot
between Cloudflare, Akamai, Fastly, but they're not as high as you might think. So that's good.
I mean, this is, if you're taking something like, you know, maybe you're doing something around AB
testing, for example. You wanted to AB test your entire website and not a submissional AB test
where we would test one version simultaneously with another, but an SEO kind of AB test where
we would test before and after.
So for SEO, when we're doing this kind of testing, so we wanted to
replace a menu structure or something like this, we'd want to insert that menu into our new site
very easily. And we could do this via Edge and then we'd want to make sure that all the traffic,
you know, bots and user traffic was all getting served the same response, the same page and the
same new menu. And we would do this at a fraction of the cost that it would be to do this through
optimizely or one of the other platforms that traditionally you'd use for AB testing. So it is a
very, very cost efficient way of doing large scale changes on sites. So yeah, not as expensive if you
would think. And you're kind of asking what the other costs are, I mean, developing, learning to
develop this. Yes. And do you need some developer like JavaScript or is it my JavaScript level
enough for that like or someone else's? Yeah, you can start very, very, very basic level to be
honest, Olesia.
So JavaScript is really what's required. Some platforms are kind of getting
into WebAssembly. So you go from Java runtime, which can just run quite basic JavaScript. You can
get into WebAssembly and Rust and start to kind of build much faster scripts that are compiled that
will work better at scale for larger sites. But yeah, certainly getting into it, it's JavaScript.
And there are slight differences between the edge platform. So there's differences in the way
that you code for CloudFlare versus Fastly, versus Akamai, they're all slightly
different.
But generally it's JavaScript. Is it hard for big enterprises to understand
the idea of edge SEO and make it adopted and proceed with that or the integration and the
whole process of starting it over? Is it hard? Yeah, I think they find it. I mean, they find it
an opportunity, to be honest, at large enterprise. So they're very excited about this because there's
a large challenge with big enterprise in terms of getting things done. I think a lot of SEOs look
at large enterprise sites and they can't quite work out why seemingly basic things haven't
been done. And I've done this as well in the past. You look at it, wow, surely they must have
known that. They must have seen this. I've run an audit with Screaming Frog and I found these
things are wrong. And surely they must be able to fix this.
And there's a kind of problem with
large enterprise sites where it's very difficult to push these kind of changes through.
And whether that's just the CMS concerned, whether it's a developer team that's dealing with
that CMS, whether it's even that CMS management is even in this country. I mean, it can be based in
offshore teams. It can be based. It's sometimes much harder to get things done at enterprise and I
think people would realize. So the opportunity you have with Edge is huge because we can come in,
directly with the Edge team, do something like redirects or do changes very quickly with a single
point, a single team and then roll these out in a period of months instead of you can get something
done in one month or two months versus things that might take eight or, you know, 12 months normally
through the Dev team.
So it's significantly faster and I think big enterprise kind of likes that
velocity. They like the speed at which you can execute these tests and these pilots. So generally
they're very, very positive really about Edge. And I think even if you don't use Edge to permanently
fix something, even if you use Edge to maybe prove the concept. So you, for example, you take Edge
and you completely optimize a webpage for page performance and you dramatically improve it.
Google will still retrieve that page as the optimized page, which is perfect. And you may then
choose to keep that layer there or you may say, well, now we proved it, we're now going to
get this done with the developer sprints. So now we're going to feed this back into
our development cycles to get it done. But that's still proving that in a matter of months,
still pretty dramatic and it's pretty fast moving really for a large enterprise. So they tend to be
pretty, pretty happy about this kind of work, to be honest.
How do you find these enterprises? How
do they find you? So it's not like Upwork case, yeah, when you go and find someone there. But that
involves maybe lots of negotiation and talking and looking at each other. And how do you do that? How
do you start like that? It's, I mean, it really is largely word of mouth and reputation that get you
these jobs. That's what I've always found. I've worked in enterprise for many years now, I mean,
obviously I've been in SEO for 22 years, but I've worked in Vodafone for eight years. And you work
with very senior people and those senior people moved to other companies and they become senior
people in other companies. And your reputation kind of precedes you really in terms of work that
you do. And I'd say that all my work pretty much comes through networks and people who either I've
worked with or know somebody who I've worked with, that's generally how. I think that's, you're
right, you would struggle to get this kind of work with upwork or these kind of sites because
you wouldn't have the, probably the reputation or the credibility to be able to walk into a large
enterprise and then get that work.
Even if you actually in reality have the skill, and I'm not
saying that people on Upwork don't have the skill, they probably do. But yeah, it's difficult to
get those kind of jobs. And then you're right, it's a lot of negotiation. I think the, I've
recently got a client and we negotiated with them for four and a half months to get the project
started. So that's a lot of effort, it's a lot of work. You're securing a master services agreement
that takes a very long time.
You have to have the right insurances in place. I think we have, you
know, five million US dollars and cyber insurance covering us. So you have to have all these
various things in place to be able to get this work. And you have to be able to get through
procurement, you have to get through legal, you have to understand how to secure these kind of
jobs. And it's definitely not easy for a sort of single person or a single freelancer, have to be
able to secure them. And then once you do secure them, then you're probably going to deal with, you
know, net 60 or net 90 before you get paid at the, so yeah, it's difficult working with enterprises.
It's definitely not easy. So before you start working with someone like that, before you
even got contract, you need to have how many, how much for your investment, like $100,000 or
how much before you get first client to do that? Yeah, I would.
Yeah, obviously you need to be, you
need to be thinking about that kind of area. Yes, definitely. I mean, you can, you can find clients
that are easier than other clients and you may, you know, I love it when we find a client who
is net 30. It's fantastic. It's, you know, it makes such a big difference. It's quite,
it's quite difficult sometimes, I think, for people to understand that we celebrate a big
client win, but the, the win is always kind of offset slightly by the fact that we now have to
cover everyone's payments that are based on 30 days, maybe for up to 90 days.
So it's so every,
every win comes with a sort of slight tinge of, you know, how are we going to, we've got to bridge
all these loans to cover these costs to get this, to get this in. So it does, it does become quite
a complex thing to navigate, but it's certainly difficult if you didn't have some funding
built up. So whether you do that, you know, through investment or whether you do
it through work on other projects, but you can kind of move up to these larger
enterprise projects in that way.
I've never taken investment really for myself, but I've always
done it through self-investing with my own work. How fierce is the competition inside this Edge
SEO, enterprise SEO? So are there really many people who are hunting after one client
or is it more diverse and more complex? No, I'd say it's, it's a very new area right now.
So I wouldn't say there's a lot of people working in this area. So it's still fairly new.
If you
imagine that Cloudflare and Akamai and the other CDNs have kind of developed this amazing platform,
they've developed amazing kind of play area for us all to create this in. And it's almost the same
as someone's just invented Apache. They just, they've created it. They've got, you know, so
what's everyone going to do about it now? How are they going to work in it? And that's, there
aren't enough people to create the, the script, the edge workers to architect the solutions. They
don't exist in this space. It's still very, very, you know, it's not very contested. I think that
will change over the next few years. But certainly it's a, it's a weird blend of skills to get good
at edge.
And I think that's why I quite like it because you're developing scripts, but you're
keeping one eye on SEO that, you know, you've done it. You're keeping an eye on performance because
if you're going to start changing things with edge scripts and you add, you know, 90 milliseconds
or 120 milliseconds to the request, you may not have had an upside to the work you're doing. So
you need to be able to understand performance, you need to be able to measure performance
in everything you're doing. You need to have understanding of networking and, you know, you're
checking routes on the command line. You've got sort of idea of kind of, you know, an idea of
CDN architecture, at least at a sort of basic level. So all of these things can come together in
one place to be able to do edge work.
And that's made it a very interesting area because often, you
know, CDN people generally would be more working on security and configuration of the CDN. They
wouldn't necessarily be thinking about coding and even the ones who would be thinking about
coding wouldn't be thinking about the marketing opportunities of doing that coding. So it's a
very, it's a very interesting area to be in, but it does require a kind of combination of
skills. And I think that's probably why there aren't many people in the area right now. Yeah,
that makes sense. How much time does it take, for example, so you've started the
contract and you have this kind of idea, which you are proposing to the client.
How
long does it take before it is implemented? So I'd say we kind of caught the plan. We worked
very much in agile scrum methodology. So we would go into a spike sort of period where we would try
to discover kind of how we're actually solving it, writing scripts. But this is something that
you would generally get done within two months. Really, that seems to be the kind of timeline
that we've been working in. We can work out the problem we can call it a solution and we can
start to deploy that all within two months. But it's still, despite it is edge and so easy
and so forth, you sometimes need to negotiate with clients, like all the time you need to, the
changes are sometimes they make you wait really long before you can do something. Like with that
added, how long does it take with the clients? I know what you mean, so you're thinking more
about you kind of make these. Suggestions and the client's deciding whether or not you can
do them.
And I think it's slightly different on edge because you kind of come in with those
prerequisites in mind. So you're sort of coming in to say we are going to do this on the CDN layer
and your prerequisites are all about access to that CDN layer. So you'll be negotiating with the
front end team, with the edge ops team, they'll be checking your skills and your ability to be
able to give you, because they're the gatekeepers to the CDN. They're not very willing to give
most SEO teams access to that area. I mean, rightly so. It goes against everything that they
work for. Their job is to keep the websites up and running 24/7 and the idea of letting in a team
to release code on that CDN layer that could potentially bring down the site is something that
would be quite scary to most front end or edge ops teams. So there's a lot of kind of prerequisite
negotiation in terms of getting you access and being able to set up the testing areas to be able
to set up the staging that you'd want to do for these kind of changes.
And then we would have
to assure them with the proper amount of testing that we would do, which are what is our testing
methodology, what are our unit tests, how are we rolling this out, what's our rollout plan, it's
very, very structured. So most of the negotiation kind of happens before the clock starts pretty
much and that would maybe take one or two months and that's kind of part of the negotiation
you're doing really to get the contract. So if you have all of that done before you start
the project, then you're pretty clear. I think one lesson that I've learned a long time ago in terms
of delivering this kind of projects is get the prerequisite done before you start the project.
Otherwise, you're right, you don't want to be negotiating access to the CDN one month in into
that kind of project, it would just be a disaster. So you know, you deal with this very much as a
sort of enclosed development project and get those prerequisites done as part of the contracting.
How do you reassure them? What do you use to make them convinced that this is required and
you are talking something sensible and like that? Yeah, it's it's previous experience.
And again,
this is why it's a difficult area to get into because no edge ops team is very willing to let
a team come in and do this kind of work and let's they can prove that they've already done this kind
of work. So it's a self-fulfilling ability. So I was incredibly lucky working with the clients that
I did originally where I was doing this very much as a proof concept pilot work. So I was allowed
to kind of explore and learn in this area and that was that was very unique. But generally clients
now would want to see exactly what you've done, what the results were from previous use
cases, what they'd expect, what you know, they need the detail and that requires that
you've done this kind of work previously. But that's that's everybody
to get their assurance. Oh I see, what do you use to measure the
results and show them? Maybe some tools, maybe something. So what's the environment of this
testing and how do you do that report on this? Yeah, I mean, in terms of performance
improvements.
I think we would see what the client had set up already. So we've
worked in RIGA, we've worked with SpeedCurve, you would use obviously Google Search Console
as well for Core Web Vitals for those kind of results. In terms of SEO performance, again,
we'd see what they had in place. You know, any of the SEO tools would help with visibility
changes, which would be good. So and then yeah, I mean, you'd get into the analytics. So what have
they got from Adobe Analytics to Google Analytics, what have they got set up? But yeah, we would
use all the analytics that I'd suppose really to prove the use cases. And I think you need
to have a good understanding of analytics. I've got several people in Torque who are very good
analysts.
We can use pretty much all of these platforms really to get the data that we need
to prove the outcome of the work we're doing. So for example, you do something and you tell
something, how long does it take? How long do you measure it to make sure it's that result
and that it was this influence that impacted it? So yeah, it's very, I mean, be exactly what
you're used to with SEO work. So performance, performance improvements are brilliant because
we love them as SEOs, because we do, we do XYZ, we release, we have charts, we can see
very clearly this is when we released the edge work or this is when we released the code
and this was the change, it's very immediate. Some things are trickier than that. I mean,
if you're doing something that affects the crawling of a large website, you'll know this,
it can take several months really for Google to relearn or adapt to that change. It's something
that doesn't happen immediately. So do some things do take considerably longer to get the evidence
on.
So it's SEO results are very rarely quick in my opinion, especially on very large websites. So
it is something where you have to kind of let the client know this is going to be something. This is
what we'd expect. This is a hypothesis and then we would work on the analytics to prove that out and
prove as you quite rightly say, what's that due to this change or was it due to other things that
happened or was it just one of the things that they needed to do to make everything better?
This is a problem that all SEOs I think face is how do we bring commercial value and proof to
what we're doing.
But the good thing about edge is the speed at which you can make those changes. So
you can do lots of very large changes very rapidly and that does help considerably. Usually
when you're just making these quite small tweaks on a large website, it's even
harder because unless you're doing changes that are quite very significant to the
templates, it's very hard to see the differences. And the changes which you want to do how do you
decide on them? So you just make an ordinary audit or does the client have some problem and
they refer to like, are refered to you to solve it? So what's how do you do it? Yeah, a bit of
both. Certainly in terms of performance problems, the client tends to come to us because they know
that they're getting a very poor performance score and they're trying to work out how to fix this.
And it's performance is a very tricky area to really fix in because you have to dedicate a
lot of resource within the company to get it fixed and you're never quite sure if that's
going to pay off.
And I think Google went very hard on performance improvements and they
linked those performance improvements very much with SEO improvements and with commercial gain.
And I think they kind of had to walk that back a little bit over the last sort of year or so.
And I'll probably get into trouble with John Mueller of saying that. But they kind of had to
walk that back a little bit because there were there was a huge effort on performance, which is
fantastic and it's great for clients, but it may not have delivered into the commercial success
that some of the teams thought it would. So yeah, some of those things are a little bit trickier.
But yeah, it's trying to find value though. With SEO fixes generally and I come to this from an SEO
point of view, I know that there are things that we really want to fix. So we want to maybe fix
things with the crawl or we want to fix things with parameters or we want to fix things with
schema or deployment in this kind of.
We know that they have an effect, you know, we know
from our own SEO experience and we know from our own audits at least to have an effect.
So those recommendations tend to come from the SEO side more than the client. So you are
building your own experiments not based on your previous experience with other clients, but you
have your own websites where you test things. Yep, we have our own websites and then you
build up experience from doing these projects on clients. So a bit of both really. So you very
much have to test on your own websites as well, definitely.
But you have very, and again, I'm
very lucky with some of the enterprise clients that I've worked with that they're very happy for
that kind of test and learn scenario to go on. So and if you take, I mean, in many respects, eBay
is probably one of the most complicated websites for SEO that I've ever seen. And I can say that
with my hand on my heart and after 22 years, it is a very, very complicated website and it's
huge. And from an SEO point of view, it's a real challenge is it's and that's why I love eBay
for that. And I know the SEOs work on eBay, love it for that.
It's an intellectual challenge
because you're trying to steer something that's so large and you're trying to steer it in the right
direction all the time. It's very, it's made me feel a lot more sympathetic towards Google
because when you're working on a very large site, like Google is a very large site, you sort of,
you're steering a juggernaut and you're trying to get it in the right way and you're kind of
nudging it back into line, back into direction. And you know, someone comes along and points to
the specific result in Google and they say, oh, Google, why is this result wrong? You know, this
is terrible. And they expect Google engineers to jump on it and start fixing that result. And
that's not how it works enterprise. You need to look at that problem, that issue and
then work out what's the solution that we can start to roll out that starts to fix it for
the entire site. You can't fix individual small issues.
If you start to fix those individual
small issues, then you're going to have an increasing amount of tech there that will catch
up to you later on down the road. And you know, that's what working on large sites is about. It's
not about fixing the small individual issues, but using them as use cases to fix
the overall direction of the site. In terms of reward, except for that you're
working with a very big and nice website and doing some of the book to the industry.
Is it
somehow also expressed in some maybe like real money terms? So do people in your edge SEO and
enterprise SEO get paid more than like other other SEOs? I was thinking about you mean by other
SEOs? I mean, I certainly I know some affiliates who get exceptionally well paid. So payment in
SEO is an interesting area.
I think it's, and I know I've talked to you about this in the past
in between different countries from markets, what do people get paid? I think the entry level really
into SEO has kind of stayed the same really, which is slightly depressing for me, that I think it's
kind of stayed the same in the UK for the last 15 years. And people come into SEO, especially
into agencies and they're getting paid around 22,000, 24,000 UK as an entry level into UK. But
it does move fairly quickly after that. So people move jobs, they jump jobs every year or so and
they manage to gain you know, two or four thousand increase each time they jump.
So they can kind
of increase that quite quickly. I'd say a head of SEO in the UK in an agency is usually getting
around 70 or 80,000 a year. That's kind of normal. There are better jobs out there. I think it gets
quite tricky at agency to get over 120. That's kind of quite rare. If you go into in-house, I
would say that the entry level tends to be higher. So you tend to get paid more at the lower
levels. And senior levels is kind of equivalent or slightly more than agency for in-house.
Consultants get paid more though. And if you're coming into kind of what you get paid for edge, I
mean, it's like with everything in digital if it's in demand and there aren't many people that do it
and it's scarce, you're going to get paid more. So yes, you would expect much better salary
for that. So for those who are looking to go into edge SEO, so maybe make a competition for you
or work with you, maybe, where do the better start in agency? Or in-house or freelancers? What's the
best way for them to make a quicker introduction into this whole business? Yeah, I think either.
But edge SEO, as I say, it tends to attract people who are very technically curious because
it's very new and I think that's what attracted me to it.
I got into SEO because I found it very
interesting and it was great doing really fun things that had big changes. And I think that's
what I like about edge SEO is being able to make a change in edge scripts across an entire
website that makes a real material difference that website. That's very exciting. But yeah, I
don't think it really matters whether you start an agency or freelance or in-house really to get
into this area. But you would have to be fairly happy to teach yourself really this. And I think
that there is material out there and you have to run your own tests, you have to understand
how to run an edge worker in Cloudflare. But you'd have to teach yourself quite a lot of
this. I don't think it's going to be very easy at the moment to go into any environment
where they're going to teach you this. So I think that's going to be the difficulty.
If you want to get into it, I would absolutely recommend it because I think it's going to be
a fantastic growth area for those people.
But it's going to be fairly self-taught. But if
you self-taught to a degree and you come to someone like me, you know, certainly I might
be able to make good things happen for you. Okay, so you found one of the contracts. What do
you put into the contract as a deliverable and do you guarantee anything and how do you manage
this in the legal terms if it's not a secret, not in big details, but how do you
promise anything there or do you at all? Oh, I think you have to. You have to
have deliverables and you have to, but you know what you're delivering with an edge
project because you define the activity very carefully.
So, are you delivering a solution
to handle the redirect? Are you delivering a solution that allows you to add schema onto
all the requests in this certain format? So, the actual deliverables are very easy to define in
an edge project. So I think it's harder to promise the commercial outcome of it. That's the same way
it's always been an SEO. You know, I don't think I'd ever want to be in a place where you know, you
would, you can forecast what you think is going to happen and you know from experience we've done
this on other projects, what's likely to happen. But promising a client that this will absolutely
happen is not a very sensible approach to do.
You need to kind of let the client know that this
is kind of the outcome that we would expect. But yeah, that's kind of how you would
propose it. But each website is very unique and sometimes the changes you need to make
to this website or that website, you've never done them before like in half, in my case it's
sometimes half of all the cases. So how do you tell them that it is really required and it isn't
how do you forecast anything if you don't have prior and prior experience with that particular
change? Yeah, I mean, this is where exactly. I mean, this is where you would and this is
probably why we initially started doing edge from an SEO perspective because we know SEO. So
you audit site and you know what a site has to do to improve their technical SEO and you prioritise
these issues because you know some will be more important than others. Some are affecting 10% of
the site, some are affecting 90% of the site. So you have an idea of prioritisation and so as a
technical SEO audit, you kind of want to execute that SEO audit.
You want to do all of those fixes.
And if you're simply doing those in edge, then all you're doing is kind of saying that if you fix
all these things in your SEO audit, then your technical SEO will be improved and we would expect
to get a positive outcome from that. I think it's easier to do that with enterprise sites
because they already have the link equity, they have the authority and generally if you improve
the technical SEO on a large enterprise site, it's a very, it always has a positive outcome.
I've never yet worked on an enterprise site and fixed all the technical issues that I found
and not had a positive outcome.
But you know, would I do that on a smaller site and you never
know, you know, is what they needed on that small site more authority? Did they need more content?
You know, where there are the bigger factors in play and you can fix all of the technical issues
on a small site as you know, or even on a medium site. And it doesn't have the dramatic effect that
you would hope because these other bits aren't present.
They haven't got the right content, they
haven't got the right authority, they haven't got the links, even. So yeah, it's easier to predict
this for an enterprise site. I guess that's, you know, well, it's difficult to get these
contracts. The easy side of it is that I know fixing things on these websites always brings
positive effect. So that's the kind of benefit of working on these large enterprise websites,
which maybe makes it easier. It's made my life as an SEO is very much easier than other SEOs.
I don't know. It swings and roundabouts, really. It was my other question, so how do you cope
with this stress? Because when you're working with this very, very big website and you know
how many people depend on it and on you and if something breaks or works unexpected, so you
are stressed most of the time.
So how do you cope with that and what do you do to keep yourself
in good, sane mood? Be able to love your life. Well, yeah, it does get stressful because you
have a lot of people involved and you're right, the scale at which these websites operate is,
you know, makes you very, very careful about the things that you're doing. And I think the way that
you kind of offset that stress is you reduce risk mitigation, which is a really big thing of working
enterprise.
So you work out how can we do this thing with the least amount of risk? So if I'm
going to test something on a large global website, I'm not going to do it on the US version. I'm
not going to do on the UK version. I'm going to pick a very, very small market and then test
things on that scale and then scale it up, roll it up as I've proved that it's been successful.
So you kind of, you learn kind of risk mitigation techniques in terms of how you deploy these
kind of, these programs and these, these pilots and that certainly helps. In terms of sort of my
life, though, I think you have to be quite happy to work in a stressful environment and I love
my work. So I kind of don't mind that too much. But yeah, it's, it's not always easy. But if
you have good people around, you good teams, you know, it, it works. So yeah, but it's probably
take more time off, Olesia, you're probably right. Can you tell what's the future of
edge where is it all going? What the trend is and what can we expect in the
near future and in the distant future? Yeah, absolutely, Olesia.
Yeah, I think I've had
some people who kind of talked to me about edge and they said, is this sort of, is this a new
trend? Is it something that's going to last for a year or a couple of years? Is it like, you know,
VR? Is it like voice search? We're not mentioning voice search anymore, but is it something like
this? It's sort of happening and it's going to go away again. And I really don't think it will. And
the reasons I don't think it will is because this is fundamental architecture for the web. We're
able to now do computation on our origin server, but we're able to create lightweight web services
now through the edge that are doing different parts of that computation.
We're able to develop
in layers and I think that's hugely powerful because we're going to do all these various bits
of logic that we can now stick on the edge that would have been very hard to incorporate into our
sort of main, you know, into our origin server, into our sort of main application. We can say,
you know, we can have a separate web service that maybe kind of helps deliver a different
site maps for different parts of site without having to develop anything on the origin server
at all.
And we can do very quick, very agile kind of development using this environment. And
I think that's hugely powerful. And you can see the way that the CDNs are developing this,
they are increasing kind of like Moore's law. We're kind of increasing the capacity and the
ability to do this work, this logic on the edge, it's increasing, you know, every year. And
the, you know, when I first started this, we were able to do simple policies for
redirects in, in Akamai sort of 5,000, 1 to 1 redirects and then 10,000, 1 to
1 redirects.
And now I can do, you know, a million or two million, 1 to 1 redirects.
So it's, it's increasing very, very fast. And the database storage in terms of KV store is,
is increasing dramatically and the ability for us to take data and attach it to the edge as well.
So we can attach our, our databases on the origin and we can basically pull data from them into
the edge. So this isn't something that's going to go away very anytime soon. It's going to be a
major part of, of internet development. So yeah, I really think it is, is the future.
Usually
they don't, but if someone has any questions, can they ask you where they can find you, do you
like to be contacted or anything like that? Yeah, feel free, contact me through Twitter's
hugely the best place to be honest. Yeah, to your point when you're kind of relaxing,
again, jump into SEO Twitter a little while and see what everyone's complaining about. So
you can find me on Twitter. Nick Wilsdon is my username on Twitter. So feel free to send me a
message there.
Okay, good. Thank you so much..