Dan Petrovic is the director of DEJAN SEO, an Australian data driven agency that's recognized around the world. You might have caught him on one of Moz.com's Whiteboard Friday presentations, or seen him at SMX.
The agency itself has been recognized all over the world for what it's accomplished by gathering data and using that data both to strategize and to innovate for clients.
Dan is free with his insights, and has mountains of them. He's easy to talk to, because he's thought deeply about just about every aspect of his work, and is happy to walk you through his process. Catching up with him was a pleasure.
If you're wondering how you might move the needle for your clients, or even how to develop better relationships with them from Day 1, then join me in spending some time with Dan.
On delivering a transparent data driven process for clients and the discovery phase
Raney: So, I understand your SEO agency is significantly data-driven. Your message highlights a good SEO will have a clear process of work transparency. What do you feel that clear process entails, and how do you work with clients in a transparent manner?
Dan: It is difficult to be data-driven for an SEO agency today, because there seems to be a layer of defense and protection in the initial phase, when you're only just start dancing together.
So, for example, an inquiry comes through, and they say,
"Can you give us a quote? An idea of what you can do for us and give us a rough idea of how much it will cost?"
To which, I say,
"I don't know!"
As simple as that.
"I need to have a look at your search console data. Here's an email address. Add it to your search console, I will have a look at your queries, any technical issues, and so forth."
And most of the time people say,
"Oh, sorry, that's a bit too much at this stage. We're just sort of looking for a rough idea."
So this is what sets the tone, and a lot of SEO companies, a lot of digital marketing agencies give up at this stage. But I tend to be a little more persistent. I say,
"Can we sign an NDA?"
Typically, with any prospective client, I try to engage in some form of work to start with, that's to do with looking at actual hard data rather than guessing.
So, everything starts from that. I think that's a super important point to establish the working relationship. Because if in the initial stage you say, "Aw, it's okay, we'll make a few assumptions," you have to remember whatever you say in the initial stages will be kind of the anchor point in the negotiation.
Whoever mentions the first price, that's what everything is going to rotate around.
So if you make an unfortunate assumption about the deliverables, traffic targets, or certain KPIs, they will always remember what you said even before you looked at the data.
We have to be very careful around that.
So, obviously, my first spot is, get access to search console data and analytics. If they have any tools I can have a look at. For me, the initial stage is the stage of discovery.
I don't even try to think strategically, or do anything clever at this point.
I'm literally just trying to get into the organization and draw upon any source of data that I can. This doesn't involve just the numbers and things we can draw in by an API. It could also mean talking to their stakeholders, management, business partners, list of suppliers, anything like that.
Stage 1, learning! Absorbing the data.
And really what I said, don't really think about it, just try to get as much as you can. Then, I allow myself 4-6 weeks to look over this and come up with something that surfaces up as an opportunity.
On delivering a transparent data driven process for clients and the discovery phase
Raney: What do you do next when something surfaces as an opportunity?
Dan: We can't quite skip over that part. That's the critical point. It has many substeps.
For example, an opportunity might be something along the lines of digital PR, or outreach. Or it could be something to do with completely missed query opportunities that should be there, that are not.
A missed opportunity could be a terrible click through rate (CTR) on a query or a page, or a set of queries, that is just abnormal in comparison to the rest of the website. We have an entire process. I could talk for two days about just that one process, CTR optimization alone.
Going through all that, and the final step once we get all the data, we try to get some fresh data as well on top of all that. We've built frameworks about how to service these insights.
The one I mentioned about CTR optimization, in short, involves looking for anomalies and deviations with regards to the averages for the site.
So for an average position 3 in Google for this particular website has a click through rate of 15%. We detect anything that has a 10%, 5%, 1% click through rate, and we go, "Wow, why?" We try to learn from the bad and learn from the good.
Some queries will outperform that rate.
Of course, you do this to scale. You build tools, you use smart ways of running spreadsheets to surface these insights.
Once everything floats up on the surface, you still have an opportunity to do a lot of market research. I think a lot of agencies forget to do this part. It's super easy. One way to do it is, let's say we're doing query research.
What are people typing in to find this type of business?
The problem is, most businesses live in their own information bubble, and most agencies live inside their own information bubble.
They all have, for example, financial institutions. They call their credit cards, saving accounts, and mortgages, "products."
We never call them products. A credit card is not a product for me, it's a credit card, right? So they have their own internal jargon that gets into their copy.
So I to listen to the actual public.
So let's say we run a survey of 1,000 respondents in a particular area. We ask them an open-ended question. We don't give them ideas with A,B, C, choices.
We run an open-ended survey and say:
"If you were looking for this, for this type of product or service, what would you type on Google?"
Or, "If you were on this website, what would be the first thing you'd look at?"
An amazing amount of queries are actually volunteered by the general audience to a particular research piece we have never considered before. This is not the long tail.
I actually wrote about this. I referred to it as the "new tail." Queries that have never been typed in before, or queries the business has never considered in the past.
This is quite exciting because Google has also confirmed "X" amount of queries, every day, are queries typed in for the first time. Google has never seen them before. They happen on a daily basis, and they represent a significant part of their search queries on a daily basis.
Users always come up with new ways to look for things. And a percentage of those queries have never been typed in before. And that's quite interesting, I think.
So once you have all this data, the inside stuff, the consultant sits down, and really thinks about it. This is where human ingenuity comes in.
You forget about the technical stuff and the process, and you really think. You have to have time to think about the whole thing. You start forming some theories, hypothesis, ideas about what would be the most beneficial page to start with.
The worst thing an agency can do for a client is just dump an unsorted, unorganized list of things to do to the client and block their development time for the next six months in no particular order.
Our job is to structure this advice and prioritize in order of impact, or in the order of effort.
So, for example if something carries a heavy amount of impact and is easy to do, then that will surface up right to the top of our to-do list. If something is just SEO hygiene, and it would be nice to have, and we don't expect much impact, and will take two months to implement, that goes right to the bottom of our list. Or may not be implemented, ever, depending on the development capabilities of the client.
So, once we have this list, it's time to go back to the client and present our findings. Because what we see as an opportunity, they may say,
"Oh, actually, we're about to phase out that product, it's not really important."
And we'll go,
"Ahh, okay, well, it looked like an opportunity, I see it's not, let's wipe it off the list."
So bringing that information to the client, getting that little bit of back-and-forth feedback, at the same time, we've touched with the client and team already in the past, because we've been speaking to their brand manager, PR people, technical department, people who are in charge of the content or loading products in the database or things like that, we've already had a relationship with them, examining the business.
Now we're going back to them. This is the second touch point where we're clarifying our findings and bringing that preliminary information to them. Before we actually form a strategy, and after we iron those bits out, we go back to our labs and finalize the strategy.
So the strategy is a comprehensive document, a road map of everything that will be done. Content, technical, outreach, everything that falls into the strategy, with the calendar of the events.
We'll do this first. We'll do this last.
Then, the big meeting happens. Presentation. We agree on everything. And we're ready to move forward.
So we already have a game plan that's structured and ready to go, but there comes the third touch point, which is essential, is that you have to gather the team of your client and get everyone on the same page.
This is the organic search strategy.
This is how it aligns with your PPC strategy.
This is what we've learned from your click thru rates in PPC.
This is what we've learned from your social department.
Your brand manager can help you in this and this way when we're doing outreach for link acquisition when they're doing the PR.
When your CEO plays golf with this and this guy, we want them to do x, y, and z. We get everyone into the same room and try to align all the channels together.
In a sense, the SEO company kind of glues together different departments in an organic strategy and tries to get everyone working in a synergy.
It's a good idea to have a little training session, planning one right at the start, where everyone is brought up to date with a new strategy. Then having 1-on-1 session with specific people or departments, more specific to the types of things that they do. In my experience, once you do that, everyone is super-charged, ready to go forth and implement.
An SEO company cannot do all the work without the input of the team and the buy-in of the team they work for.
I see a huge amount of enthusiasm arising from that point. But in about three months, everyone tends to go to their own default activities and kind of forgets what is being done in the organic part.
So it's always good to have quarterly campaign reviews to see if they're on target, if there are any bottlenecks in the process, to try to unlock those pain points and friction items.
It's always better for the SEO company to bring the bad news to the client than the client questioning the SEO company.
So, if there are any problems, if for example, a recommendation on the technical level hasn't been done by the development team, we need to get to the bottom of the issue. What's the problem? Ok, CTO disagrees with the strategy. Alright, let's work around that.
That's why it's a good idea for us to have an advocate on clients, and someone who will liaise and negotiate things internally within the teams.
So the process is complex. There's a lot more to go into. But in a nutshell, we're looking at data, drawing section, and discovery section, and then after that, strategy building, client input, slight adjustments, and presentation, after which team training, briefing, and execution of the entire strategy.
On finding the key client side stakeholder
Raney: Nice! I'm curious about this liaison. How do you identify that person, who are you looking for there, and how do you formally tell them they're going to be filling that role for you, or get them into a place where they're working for, and with you?
Dan: Typically, in 9 out of 10 cases it's the person that made the initial inquiry and organized everything that stays our primary contact. That's the person we've built rapport with, and they are our advocate within the company.
Sometimes they are just the introducer, and then we are handed over to the right person who will manage the actual account.
What I've found is the one who makes the inquiry who we have our initial meeting with is the one I recognize can be the key advocate for us who can make things happen on their end. So, I think having a bit of a different approach, rather than sending a quote and saying,
"Well, okay, this will be $5000 a month, or $10,000," and we might do these things...
A better approach is to have a more engaged relationship with the key person that made an inquiry. Working back and forth with them. Establishing a bit of a relationship with them. Working through the data.
I know it costs money and time. It engages a large amount of staff on the agency end and represents a commercial risk. You're doing all this, work and they might just go and pick a different agency.
But what I find is that if I actually do the work, before we actually get paid or engaged in a formal sense, there's a sense of accomplishment on both ends. And it's very hard to break up that relationship. That tends to result in a healthy relationship for the long term.
On controversial SEO myths and link building
Raney: Gotcha. To pivot just a little bit, let's talk about SEO itself for a second. What would you say is the biggest piece of SEO advice that's being peddled around out there right now that you just severely disagree with?
Dan: There are many pieces of SEO advice that are, I guess, controversial.
But one that is particularly bad, I think, and one that I'm guilty of spreading in the last, well, maybe three years back? Is that link building is dead. Don't link build! It's a waste of time.
I used to get up on stages at many conferences and say, "Link building is dead. All you have to do is write extraordinary content," and then links will happen organically. And then you don't have to do anything!
It sounds really nice, doesn't it?
Dan: You make the content. Build it and they will come. It was the single most devastating commercial decision I made as an agency boss in terms of our own bottom line.
Link building was a significant revenue stream for us, and when we switched to link earning, a lot of people were put off by that. They didn't believe in it. Both internally and externally, I tried to advocate link earning.
Work on the content. Work for the reasons for the link to exist, and then the links will happen.
There are different reasons this does not work for clients in reality, they expect deliverables. They pay you money, and they want deliverables.
Links are easy to quantify. "Hi, I got you ten links."
So I got a little more clever with the whole link metrics. BA, PA, the whole quality metrics for links. But to me, that didn't matter. A high-quality link, for me, is the link that has a place, that has a purpose.
So if somebody from Google's webspam team looks at this link, they won't be able to say, "Oh. This doesn't fit."
Arguably the best link you can get, whether it's organic or if you pushed it into existence, is the link that brings you a lot of clicks. It's not the BA or PA or Page Rank or whatever metrics we have.
The best link you can get is the one that sends referral traffic through.
I think if people focused on that, we'd have a lot healthier SEO industry today.
Now from the agency end, I really believed you could build the quailty and expert content to actually earn its own links. Why? Because I don't link build for Dejan. Dejan Marketing earns its links.
Which seems quite extraordinary, but not so much when you think about it. You have a couple of experts working on stuff, experimenting, probing, giving information. People appreciate that, and they share it in the industry.
The problem is, on the client end, they're busy with their projects. With their own clients. With business development.
There isn't always a brand advocate.
There isn't always a Rand Fishkin of Moz figure to drive the whiteboard Fridays. To drive the blog posts. You know what I mean? So it's really difficult for an SEO company to replace an internal brand advocate.
Because we only have limited time and resources to dedicate.
We will never know and understand their product to the extent they will internally.
So we can provide advice, and guidance, but we don't have the brand advocate, the evangelist, the marketing person, the PR person who can really do the TV interviews, the blogs, the social media interaction.
An SEO company can not come in and write an article the way that person can.
So, the reality and the fictional vision of what we think link earning can be, they're very different.
So what I've found is of course, if you have no quality content on a site, that link building is pointless.
But if you try link earning alone, without link building, most of the time it's not going to work. In reality, in practice.
So what really needs to happen is a bit of a hybrid approach. Have that super quality content. Something new. Something newsworthy.
Give it a bit of a nudge. Give it some effort. Because the links that happen as a result of high-quality content are a lot easier to get than links you're just flogging towards ordinary content.
The true reason for the link has to exist. Then we do link building on top.
In fact, there was an interesting SEO Fight Club Session at SMX Munich earlier in which I was going against Marcus Standler from Ryte. We were trying to sway the audience opinion as to whether link building is worth it or not.
It was a bit of a surprising statistic in the end. I won by 1% only after my argument, but at least I didn't lose!
It was interesting to see how few people actually still believe in link building.
I don't know how to think about that, so I decided to think of that as an opportunity. I thought,
"Well, great. You all stop doing link building, you just work on great quality content. And I will work on great quality content, and I will do link building on top of it. Let's see who wins!"
On content promotion and giving content "the nudge"
Raney: So, now I've got two questions! First, when you're giving this great content its nudge, what are some of the methods you hit on to give that nudge that maybe you weren't doing before, or people don't think to do?
Dan: It's super simple. You give it a nudge before you publish it.
If you publish the piece of content, you can beg for the links to happen afterwards. It's just hard for it to happen.
So the best strategy to get a piece of content link is to create the reason for the links while you're building the piece of content.
I have an anecdotal example. There's one of my tools. It scans the log files of my website, and it looks for new referrers. So I use this tool to detect first-time linkers. So today I can connect to my analytics account, and the tool I've developed scans for referrers on a daily basis, and every time it reports a first-time click from a domain or a page we've never had before, I get an alert.
It's a nice, elegant way to get a fresh link report.
So, one day, I get a link report and it looks like a Wordpress URL, but it's a bit of a messy URL.
It looks like a draft. I can't access it.
So I contact the owner of the blog.
"Hey Bob, I see you're writing about me, because you linked to me, from your draft, and I'd like to perhaps offer an opinion, an image. I'd like to know what you're writing about, maybe I can contribute."
After he finished freaking out that I knew about him writing about me...
He actually reached out, and I was able to contribute.
Obviously you don't have to do guerrilla tactics like that.
Let's put this in reverse.
I'm working on a great piece of content for my client. Right at this point is when I'm earning my links. I'm getting the opinions and the input of the influencers I want to share and link to my content before the content is finished. I'm embedding them within the content.
Now, SEO industry does this the wrong way. They do the whole listicle thing and the whole influencer interviews nobody reads. "Top 27 SEO Experts."
Who reads those articles? Nobody.
Those experts check whether they're on the list.
We've flogged that tactic to death in the industry.
But outside of the industry there's a good opportunity to engage the influencers we want to share and link to our content on a meaningful level. You can go as simple as asking them for a single statement. Ask them to contribute a paragraph or two. You could be asking for permission to quote them. Or you could be permissioning them to write for you, an entire post.
There are many combinations of this, but my point is, engage the influencers in the content. So they feel like they're part of it. They contributed. It becomes something they evangelize. It's something they stand behind, whether it be a cause, or an agenda, or something they're trying to push.
Once the content piece is published, you email three or four influencers you've engaged during the content, and say, "The content is up. Thank you very much for your contribution."
You have the credits there, and perhaps a photograph...so. Ego bait, in a sense, but on a lot more personal level. Rather, do a more meaningful piece, a more personal one that they feel their heart and soul was in that contribution.
So, my advice to a lot of my clients is, link earning happens before the content is actually published.
That's probably the most ignored tactic as far as proper link earning, link building, average tactics go.
It is quite powerful if done right.
You don't always have to do things that are agreeable. You can use a controversial approach.
Here's an example. I did work for a competitor to Airbnb. So it's a big challenge. We needed something quite extraordinary to publish. We wanted to enrich their landing page content by mapping the location of the rental place to its proximity to some of the best restaurants around.
So we wanted to add value to that content page.
So we did surveys, and we were asking people, "What's the best place to eat in this particular city."
We had a very surprising answer, because most of the respondents of the survey answered, "McDonalds," as the best place to eat.
I was expecting fancy-schmancy, five star, Michelin restaurants, whatever. No! The general audience says McDonalds. I was taken aback, because it's been awhile since I've actually eaten at a McDonald's.
And I think: what is going on here?
So, I go through a drive-thru. I order something. It's an Angus burger with truffle oil. It all looked a bit premium. I walked into McDonalds. It's looking very nice. The interior, the exterior, there's table service, there's touch screens.
I'm thinking: Wow! Something's going on here.
But we couldn't do anything with this story, because obviously you're not going to put, "This rental place is like walking distance from McDonald's." That's not really a selling point.
So we decided to repurpose this discovery. This is one of those insights from the discovery stage which we needed to decide what to do with.
"How's this related to link building?"
You're probably wondering. But what we did next is we did research on McDonald's and what journalists and financial bloggers thought about McDonald's.
The consensus was that McDonald's is crap, the franchise is in trouble, the franchisees are complaining, it's all going downhill.
That's what they're writing about. So I approached a number of those journalists who wrote about the decline of McDonald's and everything else.
"I completely disagree with your story. You just said McDonald's is in trouble in Australia, but people love it. Here's the data."
But I didn't attach the Excel spreadsheet or my PDF to my email. I published it on my client's website, and I made it like a little data sheet. Not the news story, because that's the job of the journalist. I published it as a data sheet. That was the source of news.
Two days later, client calls me, super happy. "Wow, did you see this? We got a link in Business Insider, we're in Yahoo News and so forth."
Second paragraph of the article.
According to (Brand Name), link, survey results.
...McDonald's in Australia is doing really, really well, here's why...
Then the journalist did their own opinion on why it's happening.
They actually took pictures from the McDonald's Instagram channel and displayed all the fancy interior and the design and stuff and had their own theory.
So let me wind back to my initial point. I had a complete disagreement with the journalist, and I brought some new data that changed their mind about something. So, the tactic of engagement of the influencers, while building the content, doesn't have to always be agreement and everyone's happy.
You can approach someone and say,
"What do you think about this? It's completely contrary to what you were writing about."
It can often result in news stories.
The other thing we did in terms of outreach on high-quality content we didn't consider before, or seriously consider, was the social boost.
Most people boost their content for the purpose of traffic generation, in their ads. But we did it to a very, very micro audience.
We did it on LinkedIn, which was quite expensive, we did it on Twitter, and we did it on Facebook, which was even better for targeting. We created custom audiences and targeted to a handful of users, rather than a broad audience.
We targeted bloggers and journalists with ads, with sources of news content. We basically replaced the human outreach via email, and phone, with ads.
I have to say that bloggers and journalists are not used to it. There's a little bit of friction there, but some will pick up the news story and run with it. Particularly on Twitter, if it's a promoted tweet. If the data is interesting enough, at that point in time, to them.
Those are kind of interesting link building tactics we have discovered that weren't utilized in the past.
On creating influencers for clients when possible
Raney: That's awesome!
So let me back you up, because you were talking about how a lot of industries don't have their own Rand Fishkins, and I think probably our industry is...we all love to talk, and we all love to talk to each other. But not all industries are like that.
So have you ever had the experience where you went in and identified somebody who could play that role? Basically created it for that industry or that company so that conversations could start happening?
Dan: I have identified a number of such people, but they were always within my client's competitor's ranks!
Raney: Oh no!
Dan: Yep! So, literally right now I'm working with one company in the FinTech space that has a competitor that's a bit more established. They have a brand advocate that is doing just that, and we simply don't have the resources on the client's end to replicate that.
There has been situations where we've had that, and it's been great, but it's not always our main point of contact.
You'd be surprised. It could be their CTO speaking at a geek conference. Linux. Or developer conferences. CRMs, WordPress conferences, WordCamp. Having influence.
So the client has nothing to do with the WordPress industry, or Drupal, but their CTO is a notable person in the industry. Goes and speaks. Becomes a brand advocate without realizing it, and earning links at the same time.
So yes, we've had clients who are influencers, but not like a central marketing person. It's usually someone in tech or other departments, which is, I find, quite interesting.
On the impact of CTR on search rankings
Raney: Carmen: So let me pivot you just a second. You mentioned CTR and I know you've done a lot of great studies about it. Google has been kind of coy about whether they use CTR as a ranking factor. What has your research shown you? Does it impact rankings at all?
Dan: Well, you shouldn't be looking at CTR as a direct ranking signal for a particular page, for a particular query.
Getting an army of mechanical Turk or SERP click users to try to skew the results...I have seen results with some of my colleagues, but all the experiments I've done haven't yielded any results.
There's theories as to why. They're all from the same platform, there's reasons, Google managed to spot it.
If you wanted to influence the CTR, you could run a billboard, or have a radio campaign. It's not that hard. But you have to reflect on the benefits, for the effort to be worth it.
Let me share something I've done that's super crazy in order to increase my CTR and inflate the search volumes in order to manipulate Google suggest.
So what I did is I ran a search query for Google Australia. I was looking for bank account numbers that people publish on their websites for their clients to pay them money.
So I paid them all money. I transferred one cent to everyone. In the description was a query I wanted them to search. So for one dollar I could send a message to people's online bank statements, apps, printed statements, CFOs, CEOs, directors, accountants, bookkeepers, everyone who has access to the transaction sheet goes and Googles it to see "what's with this?"
The line in my transaction, one cent to everyone, was "SEO Dejan."
The idea behind this, let's call it what it is, bank account spam, is that a volume of people start typing in, "SEO," and then complete the query with Dejan. The idea is for Google to start suggesting Dejan, when the general public starts typing in SEO. Would Google start advertising my brand in Google suggest when people start typing it?
What I learned is, it did work, but it worked only for the short term. Once I stopped the campaign, the effect faded away. I also got a call from the Communal Bank of Australia investigating the whole thing. Also I received some legal letters from companies who were questioning why we transferred one cent into the wrong bank account and they had to refund that one cent to us.
A ton of other complications I would rather not go through again!
Raney: It was bold!
Dan: Yeah. Good fun! But guerrilla tactics tend to backfire from time to time. I think I'd be a bit more careful in the future. I'd get some legal advice before I run stunts like that in the future.
Why am I saying all this other than it being a fun story?
The point is, no matter what you do, no matter how genius the idea, if it's not genuine interest in the page to boost the CTR, or boost the suggest, it's just going to fizzle out.
So the effort to boost CTR is better spent on thinking about,
"How can I increase my traffic?"
So let's consider the ways of bringing traffic, organic ways. I would argue there are three:
1. Increase the rankings of the queries that are there.
2. Rank for more stuff! You didn't rank for these things, now you rank for them, now you're getting more clicks.
3. Well, you are ranking for these things. You can't increase the rank, because you're selling Nike shoes and above you is Nike.com. You're never going to increase the rankings.
So what can you do in that case? That's the answer. CTR.
If you think of CTR optimization as a traffic generation thing, rather than a result or rank manipulation tool, I think you're on the right track. So, as I see it, CTR optimization is a sub-discipline of organic optimization of SEO that deals with traffic generation and increase.
It's a wonderful process, and it's something we have control over. Think about that! Nothing in SEO is truly controllable. You can fix all the technical stuff, but at the end of the day if you have two websites that are both technically sound, one will always have a bit more authority.
Authority is always so misty and vague and it's determined by an algorithm that is driven now by unsupervised machine learning. What does that even mean? How does an SEO know what brings the authority to the page.
People just default to link building. Link building, technical optimization, but why not focus on something that you can really control?
I say to a lot of our clients,
"We are the weatherman. We cannot change the weather. We can predict it. We can tell you to bring your umbrella that day or stay in the house, but we cannot make it rain."
That's not what SEO companies do. Google makes it rain, not us. We can only prepare and predict certain things.
But what we can do is, we can change the attractiveness of the search snippet by using schema. By optimizing for featured snippets, by taking advantage of new and exciting ways to structure our titles and descriptions, running CTR experiments, we can improve our rankings.
Once you detect the deviations, you detect all the negative ones, what do you do?
You run a report company-wide.
"Please stop using this term or this format in your titles and descriptions when you're loading products into the database. Because it decreases our click through rate."
Likewise, if you find something really boosts the CTR on average you can send a similar memo saying,
"Start, or try experimenting, with this."
Recently within the tool we use for the CTR optimization process I introduced a "CTR experiments" feature, which looks at the previous period and benchmarks it against the current period of time.
Kind of like a date comparison in analytics. It can tell you which snippet, where you got more clicks. You take a period of time, you change your snippet, you go to submit it to Google and you run the experiment. You can see if it runs better or worse than the previous period, and see what works and doesn't work.
I think it's an interesting opportunity for the SEO world and in-house marketers to run mass-scale CTR optimization experiments. Not just for a certain term or a certain page, but you can take 500 products and run it against 500 products in a different category. Or similar category. Or split the category in half and put '$' in 'no $' in the other. Put 'free' in one and 'not free' in the other.
Continually benchmark and optimize.
Doing things like that are one of the few ways we can literally take charge and have some control over it.
Most SEO companies don't do it, which is interesting because PPC people do it all the time. You're paying for that click and you're trying to boost that click thru rate. It's a significant factor.
But in the SEO world, we're all kind of focusing on the rank, which is ironically not something we can control entirely.
On the April Fool's Day joke Future Rank
Raney: So to close us out, speaking both of being a weatherman and being a little crazy...I understand you played an April Fool's prank called Future Rank? What can you tell me about that?
Dan: Well, I was in Munich with a couple of friends. We were going to have a little brainstorming session, like a think tank session with regards to the future of SMX and what to do next year.
It's a bit of a tradition.
John Mueller was there. And we were cracking jokes about different tools and things. One of the guys says,
"There's this tool that can look at the historical SERP volatility data,"
You know, like my tool Algoroo.com, and MozCast, and all the others.
"They predict when an algorithm change will happen before it happens!"
We all started laughing, because it sounds ridiculous. But I think he was talking about an actual tool! I don't remember what it is, and I probably wasn't going to flag it, but it was just a day before April Fool's and I said,
"Well let's prank everyone."
We'll say we use quantum computers and flux capacitors and this and that, and we'll look at historical data on Algoroo, and we'll predict when Google will have the next core update!
We did and it was one of my most shared posts. People had a good laugh about that.
Tools like Algoroo have a lot of criticism. A lot of people don't understand what we do. They think we're just kind of guessing what happens. Michael Martinez, for example, he said, "Your tools are flawed because it's based on traffic..."
"Wait, hold on, it's not traffic. It has nothing to do with traffic. We are a volatility index. We compare the rankings from today to the previous day. That's all we do!"
If the order of the results are very different today than they were before, the bar on the graph would be very high. If it's very similar to the previous day, it would be very low. That's all we do. It's not rocket science. It's really really simple.
Tools that track Google's algorithms have all been showing consistent results, because it's easy to track that. There's no magic in the whole thing. Of course, there's no way for us to predict future events at Google. They're not strictly determined by statistics like the stock market or Forex trading. Sometimes a Google engineer says, "Alright, it's a good day to push an update!" And they just do it.
There's a human element to that, and you can never predict those things! So, it was just a bit of innocent fun. Some people got pranked for real, started sharing it in things. I hope they didn't feel betrayed, but yeah, we had a lot of positive response. It's a bit of lighthearted humor about the whole obsession of algorithm updates. That's why I put the focus on CTR, one of the things that doesn't depend on factors we can't control like ranks, and updates, and living in fear of the next update.
If you know you're optimizing your CTRs and you've increased all the traffic you can out of that, then you're doing a good job. You're not going to increase it by 200%, it's usually 5% or 10%, but that's 5% or 10% of traffic you didn't have before.
Want to get more of Dan's insights? Follow him on Twitter at @dejanseo.