You want your website to rank high on Google, right? You’re in a constant battle to bring in organic traffic and provide a phenomenal user experience.
It might even feel like you have to abandon a beautiful design in order to improve your website for higher conversions or search engine results.
In episode 71, I teamed up with SEO and conversion expert Nick Eubanks to bust these familiar myths. Nick shares his knowledge on how to use SEO techniques to design a conversion-focused website.
Everyone explains that making your business different is vital — but NO ONE (not even experts) explains how to actually do it... Until now.
Just click on that big fat red button, answer a couple of questions, and learn to stand the f*ck out in a no-bull, super-practical way:
"A terrific celebration of marketers and marketing in all its forms."
"When are you going to do something in French so I understand it?"
"You're literally the only marketer I can stomach."
Louis: Bonjour, bonjour, and welcome to another episode of EveryoneHatesMarketers.com, the marketing podcast for marketers, founders, and tech people who are just sick of shady, aggressive marketing. I'm your host, Louis Grenier. In today's episode, we're going to talk about designing beautiful, yet conversion-focused website without having to sacrifice on SEO or speed, which is quite a heavy topic.
To clarify, if you're a listener of this podcast, you know that we're not going to talk about the so-called best practices that will become obsolete in the next six months. Instead, we're going to really try to identify the core principles behind this very topic so that you can listen back to this episode in two year's time, or even five year's time, and it should still be relevant.
My guest today is an SEO and conversion expert with 14 years of experience. His first business was actually a lemonade stand, which is not original, except it wasn't lemonade it was his mom's mint iced tea, which is pretty cool. Now though it's a bit more successful of a business that he founded a few years ago.
He's the founder and CEO of From the Future, which is an agency that helps you build pages that rank and convert. He's also one of the three partners at trafficthinktank.com where they help
digital marketers level up their game in digital marketing and SEO in particular.
His two compadres, they are actually Matthew Barby, who's the director of acquisition at HubSpot, you might know him, and Ian Howells, the director of SEO at Learning Tree. Nick Eubanks, so happy to have you here. Welcome aboard.
Nick: Thank you so much, Louis. Very happy to be here as well.
Louis: It almost feels like the intersection of ranking well on Google when you do website and having a pleasant site that seems to convert, doesn't exist. It almost feels like you have to compromise one of the two.
Nick: It certainly feels that way.
Louis: We're going to try to debunk that and try to understand how to design something that is quite the intersection of the two. Something that works for robots like Google and also something that works really well for humans like us. Before we dive into this step-by-step together, can you describe and explain what SEO is very briefly?
Nick: In its purest form, SEO, or search engine optimization, is the process to try to rank pages
of websites in Google.com search results as high as possible.
Louis: Can you describe what you mean by conversion?
Nick: The goal of getting a user to your website to take an action that you want. That could be something as small as sharing a page on your site on a social media channel or giving an email address to something more directly attached to revenue, such as placing an order or filling out a contact form to become a lead.
Louis: Great. Now we've defined the basics and everybody should understand what we're going to talk about next. Let's dive in into a step by step together because I very much like the topic that you're going to talk about which is designing those beautiful websites, yet conversion centered.
Let's say that you have a client that you want to work with. You might pick an industry if you want if it makes it easier to explain or you might just want to get general, but when this is this objective of building such a website, where do you start? What is step number one?
Nick: Step number one is usually to become familiar with the client's personas. Who their target audience is? What the demographics are that are represented within their audience? How old they are, are they male and female? Where do they live? What's their education level? What is their income level? What are their interests? Where do they spend time on the internet?
What is their phraseology or the nomenclature that they use to describe the problems that they have, the pain that they have that they're looking to solve by finding the product or service that our client would provide or the way that they characterize a solution. That's a very important first step so we can begin to understand what keywords ultimately that these people, this audience, might be typing into something like Google to try to find pages where we want to make sure our clients, it's where we're showing up for.
Louis: How do you find that out? How do you particularly understand the phraseology and how people explain the problems they are looking to solve?
Nick: The way personas are most often built is using surveys. There's a lot of really cool survey tools that are out there now, so you can do ... in the olden days, we would use Google Ad Words or potentially even Facebook ads to try to target our audiences and get them to fill out surveys in exchange for some sort of monetary. If it's something super simple you can enter them into a drawing for a chance to win $100 Amazon gift card.
If it takes a little bit more time, and there's some higher value, you may have to trade some monetary vehicle every single time for those survey responses. For every survey response, you may have to pay $5. There are other ways to drive these a little easier. Amazon has a service that combines humans and machines called Mechanical Turk and now Google has a survey function where you can get closer to .80 cents to $1 per survey responses directly through Google's infrastructure.
Once you have those personas dialed in, then there's a couple of different pieces of software you can use to start feeding in what you believe as the marketer, as the SEO, that you believe those keywords are that are going to have search volume, to get actual search volume data back.
You can use something like Google's keyword planner, which is free, which uses data directly punched into Google in a number of different sources that they use to collect the data, or you can use a tool such as Keyword Keg, Ahrefs, SEM Rush. There's a lot of these SEO tools out there that leverage quick stream data to give estimates on how often they believe these keywords are being searched each month.
Louis: Let me take a step back because you are talking about something quite interesting with surveying people. You're talking about actually, let's say you work with a client and they understand, more or less, the type of customer they want to reach out and the most profitable customers they want to reach out to and they want to understand who they are in a demographic level. Then you would use a paddle type of service, so a way for you to reach this audience by just saying I want to send a survey to people between 20 and 40 living in the US and who are marketers. Right?
Louis: Without necessarily going into the details of every single question you use, but perhaps you can identify for us the top three questions you like to ask when it comes to the phraseology and the way people talk about their problems and stuff like this?
Nick: That's what it is. You want to ask leading questions that give you the opportunity to collect that nomenclature that you're after. You want to ask broad questions in the sense that they serve a specific function. You're not asking for use cases as much as you're asking for, "If you were in this scenario." Let's say that we have a client that sold a specific type of cleaning product.
Their cleaning product was organic and they wanted to find out what a potential consumer, what they would be looking for in a product and what words they would be using to describe that. Maybe the average executive in that company would be like, "Oh we'll just call it organic. We'll rank for organic kitchen cleaner."
Maybe that's not the terms that their customers or potential customers are actually using. They may ask for, some of those questions might be, "What is very important to you in a kitchen cleaner?" You would get the respondents describing, "It's able to kill germs, it's not going to kill my dog, it's safe for my kids, I'm able to prepare food on it immediately without having to wait or having to re-clean it. I don't want to have to do multiple steps where I'm disinfecting first with one cleaner and then have to water it down or wipe it down with water so as not to contaminate any food I might put on my countertop."
All those questions are getting at the benefits of an organic cleaner, where you don't have to worry about any of those things, but the people who are responding to those questions, not once did they use the word, "I want an organic cleaner, or I only use organic cleaners."
On the SEO flip side, you're realizing that the terminology that your potential customer base is using to describe your product, while it is an organic product, they're using benefit statements like, "Kid safe, pet safe, does not contaminate food, zero additional prep time." There's other terms that are going to come out of those surveys that will allow you to capture a lot of those value statements from a potential SEO perspective.
Louis: I very much love this first step because it is something that I believe is not going to change anytime soon, in the next five, ten years or even 20 years. Search is there to stay, whether it's going to be voice search or any type of other search, but people are always looking for answers for their problems, their day to day stuff, so that's fine.
The second thing that you're relying on right now, is you're relying on how people actually talk and the way they think, which is a second pillar that you can really leverage. That will always be the case. You're trying to, instead of coming back from a very ivory tower type of mindset where you decide, "This is how we're going to call this and this is the type of keywords we're going to try and rank for and try to rank for," you're using people's way of talking and their own vocabulary, not to just match that?
Nick: To your point, right now it's estimated that by 2020, 56% of all searches, at least in the United States, will be done via voice. Really understanding that nomenclature, the cultural representations that are going to become more and more important as more and more of search switches from being typed in into being driven by voice search, where people who potentially maybe would be, they're formalized the keywords they're using a little bit more because they're typing them, are just going to be having these
very natural conversations with these voice assistants, whether those are Alexa or Siri or whatever that one is that Microsoft uses that nobody's heard of yet but who knows?
It's changing very quickly and it's funny to begin to see and understand patterns where nomenclature changes. What we call a phone in the United States, people in the UK call their mobile and people in Germany call a handy. Not to be too inappropriate, but a handy in the United States means something completely different.
Louis: Yeah, which is why it's important to really target your search and think about per country and all of that. There's also a funny story, it's aside but I want to think about it, I want to talk about it. There's this Canadian chocolate brand recently that's spelled S-H-Y-T-E. Okay. It's caught on and even in Ireland and Scotland people say Shyte instead of saying shit, they say Shyte. The slogan is Eat Shyte. Their slogan is Eat Shyte. Seriously.
It's so funny. It's a side note, but it really goes back to this. Understanding people to a level where you understand how they talk. You understand how they search, you understand how they think and therefore you can match that with a proper SEO strategy.
I very much love that. I asked you the top three questions, but I think the number one that you mentioned is, "What is it you're looking for in this product?" Is there any other briefly, that you like to use? Type of questions.
Nick: Yeah. Questions that are polarizing tend to drive the best answers. If you're going to support, you're going to open it up to allow for people to give you polarizing answers, you need to create a polarizing question. You need to use language that's going to evoke emotion from the responses.
You may use in your question the word love or the word hate. Very polarizing as opposed to prefer to dislike. Using that strong, emotionally driven language. What are three products that you hate and why? Then suddenly your core audience is going to be explaining to you, "These are the products that I completely hate, that I would never buy, and here's why I hate them."
You're getting not only insights into what are the words that they're using to describe the things that they don't like, because your product may share many of those same characteristics.
You as the marketer need to essentially decide, "I need to make sure that I'm not bringing that to light or I'm not focusing too much on that or I need to reframe it, I need to find a different way to message these components." Sometimes you may need to make mention of some of these components for compliance reasons depending on what your industry is.
Again, given the opportunity, leaving questions open-ended enough where you're providing the carrot as the bait, but you want to leave it open-ended enough that you don't get too specific of an answer that it's not as useful as it could potentially be.
Louis: Yeah. You might run the risk of screwing completely your data if you're just asking leading questions that leads to the answer you want to hear.
Nick: Exactly. One of the absolute worst things that I see some people who are running surveys do, is they ask questions that are ... multiple choice is not great. I would prefer not to have multiple choice, but what I think is the worst is binary questions. When you ask people yes and no questions in a survey, I think that's a waste in my opinion.
Louis: From my experience, in terms of survey questions, a good thing to start with is always open-ended answer and an open-ended question so that you really don't screw any data and you just let people talk. Then let's say you ask the same question over and over again and you know that 90% of people usually answer those four type of answers.
Then you can use multiple choice question with a fifth possibility that says, "If this isn't the case, then what it is?" That's a good way to spot trends then going forward. I completely agree in terms of open-ended question. That's the step number one, which is quite already loaded with value. Once you know how people talk and the type of way to describe your product and all, what do you do?
Nick: Then it really comes down to the process that we would call keyword research. That differs pretty tremendously depending on who you talk to about it. It's something that we really put a lot of time and energy into. We at From the Future, bring a tremendous amount of data into that component of the process, so we scrape Google for who's ranking. We then go scrape each one of the individual sites that's ranking.
We look at the HTML on those pages, how many words are on those pages, what's the keyword use and all the different meta attributes. The page title versus the URL versus the headers that are on the page. The word use on the page, not just within the count itself. Then pulling a number of other third-party data sources to really triangulate, "What is it about this page that's allowing it to rank, but also what are the total number of keywords that this page is ranking for."
I think a lot of people that aren't really in the weeds in SEO don't understand that one page, anyone given URL on a website, can rank for thousands and thousands of different keywords. It's not just about setting up a page to rank for one keyword or ten keywords or even a hundred keywords. We call those, internally we call them bigfoot pages. Pages that have a large footprint where that footprint is comprised of lots of different keywords that that page might rank for.
We try to get that dialed in as much as we can and align groups of keywords to topics and then use that topic to create the roadmap for the website in terms of content, whether that's a sitemap or editorial calendar or content calendar depending on what you like to call it, but using a lot of that information to drive what should the content requirements be for that site based on who the audience is and what the goals are from a search engine optimization and traffic perspective.
Louis: We obviously don't have time to drill into the exact process you use for keyword research and this isn't the topic of this particular episode, although I'd love to talk to you again about keyword research in particular and dive into your process there. You've already given us a few steps within this keyword research realm, but perhaps you can give us one of the resources that people should check out if they want to know more about this particular topic?
Nick: Let me just double check here, and this is going to be different in each country, but if somebody wants to do a search for keyword research now on Google, you should be able to find an article that I wrote, that I updated just a couple of months ago for this year on the website, on ImFromtheFuture.com that really goes into a step-by-step process for how to start, how to affirm the terms that you should be looking for with data, what to do with that data, how to format it, how to think about organizing it and
prioritizing it. It's a good, free blog post that can get you started.
Louis: Okay, so folks can just maybe search on Google for I'm From the Future, keyword research, and they might find it, right?
Nick: If you just search "keyword research now", like N-O-W, we might even, and I haven't checked so I'm doing this now, is we might even rank for keyword research 2018. We rank number two for keyword research 2018, but that might be another easy way to potentially find it.
Louis: Okay. At the end of step two, we then have keywords categorized by the core topics that we want to go after right?
Louis: You took the example early on about this cleaning product of some sort.
Louis: Maybe just give us an example of a typical topic with the keywords that are within this
topic for this particular product that could be.
Nick: For a cleaning product, topics might be all the various use cases. It might be cleaning products that are safe for kids versus cleaning products that are safe for pets, cleaning products for the office, cleaning products for the kitchen versus the bathroom. Everyday cleaning products versus commercial cleaning products. This is where that ideation should be really driven by the data that you get back from the keyword research process.
Once you cast a wide net and you collect all this data and all these different term variations and you see how much they're being searched for each month and how competitive they are based on a bunch of these different competitive scoring metrics that are provided from some of the software tools out there, you can then start to group them based on patterns.
Nick: For kitchen cleaning products there is about 600 keywords that have a monthly total search volume between them of 100,000 searches per month. Kid safe cleaning products, that's got maybe 400 keywords, pet safe is 900, so that's a much larger group of terms but maybe that doesn't have as much monthly search volume. There's more total keywords that people are searching for but they're not searched for as much individually, so the sum of that population is going to be lower.
You can use these to create themes and based on those themes, the best way to take that and use that to inform the design of the website from the information architecture perspective, is to figure out what are all the themes that I need to support and then how can I build an architecture that allows me to support these themes in a way that would still not only work for search engines but would also allow me to attract potential searchers at each stage of that conversion funnel?
Nick: The general conversion funnel, more times than not, is different in different verticals, but for consumer verticals like if we're selling a cleaning product to Suzie Homemaker, there's going to be four stages. There's going to be the initial information stage, whether she's using shorter keywords, two or three-word terms because she's not really sure what she wants yet.
Then as she does some reading, clicks on some results starts gathering some information, she moves from informational into more of the investigational phase. These are going to be three, four, five, potentially six-word keywords where she's gathering more information. She's starting to refine the keywords that she's doing. She's starting to reflect some of that nomenclature that we learned about from the survey process that we may have built specific pages for. Then ultimately, she's going to move down into the transactional phase, where she's going to be using specific modifiers.
Nick: A modifier is a one or two-word keyword that is usually added onto a larger string of terms that are representative of the keyword that changes the intent. These are going to be the ones that are going to be most representative of buying behavior, are going to be modifiers like buy, best, coupons, discounts, where to buy, and other things like zip codes and cities if this person is looking to make a purchase locally versus online.
Louis: Right. Let me deconstruct what you just said because it's so valuable. First of all, you
mentioned a term called information architecture. Can you define what it is briefly?
Nick: The best way to think about an information architecture for a website is the URLs. The URL architecture is what ultimately creates the information architecture of the site. There's a little bit of nuance to that because you can support an information architecture based on the content you put on the pages, but it's really the best way to think about it is how links flow through the site itself, internal links.
Your average eCommerce site probably has a home page, an about page, a contact page, and then it's
going to have a shop page, some top page that shows all the products, from your total product catalog page, the standard information architecture for an eCommerce site is going to be categories and then those categories will have subcategories and then those subcategories will have the products in them themselves.
Louis: Okay. We were at the stage where we had the themes classified by keywords, and what I like about this method that we talked about a few times in this podcast, but that is still very nice to repeat, it's customer research. We talk in this podcast quite a lot about how to understand people, how to make sure that you understand how they behave, how they don't behave, how they talk, how they don't talk.
Your way of describing it--because you're in the world of SEO and this is what you do day in and day out and you're an expert in it--is really to use an indirect way to understand how people talk by just scraping data from the biggest search engine in the world and understand how people type, how people search, and you also ask directly surveys to understand how people search and all that.
I very much like that, because it's seen very much a core principle of marketing that's not going to change. Even if voice search takes over, you can still do that, even if there's another type of way people are going to look for information, you will always have to look back and how people think and how they search, so that's all good.
You have those themes, those keywords, and then you start to talk about how to transition and use those themes and keywords to build the information architecture. Then the next thing that you said that was really interesting is the conventional funnel or the way people are looking for information in those different stages. For consumer type product you said, can you repeat the four stages again for us, please?
Nick: Yeah. For a consumer product, it's really three stages. There's one additional stage in there that doesn't necessarily fit into the common model, and that's the awareness stage. The reason the awareness stage is not always going to fit in there is because an awareness is usually driven by brand exposure. There's four general buckets of keyword intent, which is what we're talking about right now.
We're talking about the intent behind these keywords. There's informational, navigational, commercial investigation and transactional. Navigational means that somebody is searching for the name of your
brand or the name of your product by name because they have some awareness of you already. They've seen you, somebody mentioned them to you, they saw an ad, they heard a radio commercial, they saw you on TV.
It would be, in terms for you guys, it could be whether people are searching for the acronym, EHM or Everyone Hates Marketers or they're searching for your name or they're searching for specific podcast titles, these would all be representative of navigational searches. If we remove the navigational component and just look at the core, straight line funnel, it's going to be informational, investigational and then transactional.
Louis: Okay. What you do there is you link your research, your themes, to a specific stage in the funnel right?
Nick: Yes, correct.
Louis: Okay. In the phase where we describe keyword research, not in a crazy amount of details because we have to talk through the different steps, but you need to keep in mind that those themes need to be related to a stage in the funnel that you can't just sort out keywords that would be in different stages. You talked about those different stages and you talked about short keywords versus long-term keywords or very specific keywords need to be careful not to group them together, right?
Nick: Correct. You can map intent for keywords. People who are searching, they're showing you what they want. It's not too hard to infer what their expectation is from the words that they're using to drive their search. We work with a lot of software companies. The beautiful part about working with a lot of companies within a specific vertical is there's a lot of commonalities between them.
When people are searching for solutions ... A great example is we worked with a whole bunch of different accounting software companies. There's this big theme right now around accounts payable and all of these would be opportunities for automation within accounts payable.
You've got people looking for AP automation, that's very top of the funnel. It's somebody who was at ... They heard this term somewhere, they read a blog post about it and they're curious. They type in AP automation. They're not even sure they're looking for software yet because the modifier that would show intent for them potentially looking for a solution, which is one of the most common use synonyms for software is, they would either search for AP automation software, AP automation solution, AP automation companies, AP automation vendors. There's a small handful of modifiers that would move them from top of the funnel, AP automation, into the investigation phase, AP automation solutions, AP automation software.
What's interesting about software is the bottom of the funnel is not as obvious as it would be with something like eCommerce where you would start seeing all of these buying intent modifiers. People aren't usually searching for demos. That's the end of the funnel as a contact form or a demo request for a software company, if it's B2B.
If it's consumer, it's a different story because it's a much lower price point, but in the B2B space, you want to get that demo. When was the last time you were looking for a piece of software and you went and searched for a demo of that software? I just can't imagine people being like, "I want to search for AP automation software demos." It's just never going to happen.
Instead, what they're going to do is they're going to look for integrations and specific features. People are going to look for AP automation for QuickBooks and it's because they have an existing accounting platform and they want to find solutions that can plug into their existing tech stack and extend their existing systems to offer additional functionality that they know they want or they know they're interested enough in to start prospecting for vendors, but they're not sure who those vendors are yet. Does that make sense?
Louis: It does. It makes sense then when you're at the grouping stage for those keywords, I'm visualizing this huge spreadsheet where you basically have the list of keywords as a column and you just have those buying stages as the next four columns, and you just try to match each keyword to a specific intent?
Nick: Yes, yep, yeah. We match the intent. Exactly.
Louis: Okay. Let's say we have that now. Let's say we have those core themes. We know for each theme which one is relevant to which stage. Then how do we start turning this very raw data, because it's still data in spreadsheets or in the software you use, into a site that really converts, that delights people?
Nick: The first part is converting it into requirements. We would design the sitemap, and then from the site map, we would get down to page level requirements. What is the use case for this page? What is content that we need to have, based on the persona? How much of the content do we need to have on this page? It's really marrying the two of them together. It's what do we need to speak to from a use case and from a sales perspective?
How do we make sure this page does a good job at selling the content on this page, selling the purpose of this page, but how do we also marry that against the SEO requirements? SEO requirements tend to be word use, word counts. We need to make sure that we get this group of topics and these synonyms onto the page in this volume and we need to have this number of headers on the page.
To get more technical for a second, without screwing this up too much, is one of the things that we do that's very different from most other technical SEO firms, is we're actually starting to now take deep considerations for natural language processing, which is how information retrieval for search engine works. It's a core piece of Google's underlying technology and algorithm. Understanding what are the terms that Google is expecting to see on this page that will represent the themes, the contextual relevance that Google's expecting to have created on a page that they would reward with a very high search engine ranking?
That comes down to measurements of things like salience. Salience is a measurement of term frequency or how many times is this term being used relative to the total amount of terms and topics on this page? We want to make sure that we get that dialed in to make sure that this page is representative for the keywords that we want to rank it for, but not too representative for keywords that we need to use to tell the story on the page, but don't really have an impact from an SEO perspective.
Louis: Right. Let me try to deconstruct what you just said because it's critical. That's the way technology is going, starting to act as close as possible to a real human who would actually read a page and try to understand the context. It's trying with the technology to act as a human reading the page, and it's trying to infer that by looking at the volume of keywords, that it's not too salesy, it's not too much, it's not too less, yet it's using connected terms so if we talk about cleaning product, you might expect connected terms like sponges and whatever else.
You're basically trying to reverse engineer the reverse engineer almost. It's like trying to understand how Google thinks, but ultimately how humans think. Do you have any tools or any resources that people can use to get started with this particular step?
Nick: There's a couple. I wrote last year about how to design an SEO content map, and that's essentially how to take all the keywords, once you've got all your keyword research together and you've got all the buckets, how to think about organizing them to inform a site's architecture and create essentially your sitemap and your content plan. On the NLP side, there's a guy named JR Oaks, who's on Twitter, and he's brilliant. Every day he's building more and more SEO and analysis tools based on natural language
processing and where it's going.
Then from a tool side of things, there's a company based out of Germany called Ryte, RY-T-E, they have a really cool beginner's level natural language processing toolset called TFIDF, that stands for Time Frequency Versus Inverse Document Frequency. It's sort of an initial approach at natural language processing.
Then there's two other tools that have come out recently that both have APIs. If somebody wanted to try to start understanding how natural language processing works and how the process for information retrieval begins to score groups of text and topics and identifying things like entities. One is called Aylien, spelled just like the word alien but with a Y, it's A-Y-L-I-E-N. They have a text analysis API. Then Google just put out their own this year and that's their own natural language processing API.
Louis: What is the outcome of these two? What do you do, let's say you have your sitemap ready so you know which pages you need to create and you're trying to figure out the type of content you need to produce in order to answer people's queries or to match what they're trying to do. You said that very well with the intent and the buying stages and all of that. For the natural language processing technology, where does it fit into this equation? What do you use this for when it comes to building a page?
Nick: We use it more for fine-tuning. We made the mistake that I've learned of trying to create the requirements and getting the content portion of these pages dialed in before we ever bring the pages live. I've learned that's just not smart. As much as we believe we understand what Google wants and we have all these great tools to lead us and guide us, it's never exactly right.
Instead, we'll have conversion-focused copywriters or we'll have copywriters that are very familiar with writing to generate leads to drive sales. People like Joel Klettke is a great example, from Business Casual Copywriting. He's one of my favorites. We'll have these guys and we'll have them write based on the personas. We'll provide the personas to them, we won't give them any SEO data, and we'll have them write these pages based on invoking emotion and driving conversion.
Then once we have all the content from the pages from our copywriters, then we'll go back and run either the staging URLs or we'll run just the big blobs of text through these text analysis APIs. They'll come back and let us know, "Hey, these are the entities that were identified, based on the sorting and scoring algorithms of these tools. If these are the entities you want, good. If there's entities in here you don't want, you need to sort of thin them out. You need to make them less representative." This is where the
technical approach gets really fun.
We just did this last week for a client int he financial space. They have a loan product. When we ran their pages through Google's natural language processing API, we found out that they had high salience, very high term weight, for a bunch of keywords that they needed to have on the page from a compliance perspective, and from a sales perspective, but they're terms that we don't want Google to have any relationship of this page. They're not important for the search at all.
We came up with a strategy to instead find a way to get those words off the page, so keep the messaging on the page for the users, but get the text itself off the page. What we did was, we created images, we created a horizontal bar, and we used it to really call out these requirements from both a legal compliance standpoint, but also from a sales perspective, that allowed us to meet that requirement, present that information to the user, but not have it be present in a way where Google's crawlers are able to find it
Louis: That goes back to your initial research on how people speak and stuff. I guess the reason why you didn't want those keywords to be on the page is because you know that people weren't searching for those terms or that it could confuse them as well?
Nick: Correct. The bigger issue was it was these terms were being used enough on the page that it was making the page more about these terms than the terms we wanted to target.
Louis: Right. It's a very, very advanced way to understand whether your content makes sense. As a consequence, making sure that Google understands that this page is for this particular topic, to not think that this page is for this other topic that we don't want to fucking rank for, it's only for this particular theme?
Louis: Right. We went quite deep into this because it's quite interesting. Now let's try to take a step back and talk about the next step in the process. Let's say we have a sitemap, we build our content. If we don't have a lot of money to write all of those pages, you can either take a stab at it yourself, but I would definitely recommend you to listen to one of my episodes with Joanna Wiebe about commercial copywriting and frameworks.
Nick: Funny enough, the guy that I mentioned, Joel Klettke, he works very frequently with
Joanna Wiebe. She's tremendous.
Louis: She's definitely the authority when it comes to commercial copywriting and she has the right tools, the right methods, the right frameworks, for you to start, even today and to write with commercial in mind. You can definitely listen back to this episode where we go through a few frameworks and all of that. Let's say we have that and we've written those pages.
At this stage, we don't have a design per se, we only have text and shitloads of Google docs or whatever else you're using to write all of those pages. How do we turn that into a website that we know is appealing for people?
Nick: This is where we get into the meat and potatoes, to use a hopefully somewhat Irish euphemism in this conversation. It gets really exciting. There's a lot of schools of thought that are sort of broken. One of the first signs of an inexperienced web designer or web development firm is one who will talk to you, they'll meet with you, they'll have you gather a bunch of examples of websites you like, and they'll go out and they'll start delivering designs before the content is done.
Designs should always be informed directly and 100% from the content itself. You should never be doing design without the content because you don't know what in the hell you're designing for. Once you have the content you can begin to look at design patterns that are going to be user friendly and search engine friendly. This is where stuff gets really exciting. It's what we do very specifically. It's one of the things that we do better than most of the companies out there, and it's because there are design patterns that work much better for search engines than others.
Specifically, Google moved officially, they've begun rolling out their mobile first index January of this year, which means that the content that is on your mobile site, your mobile pages, the mobile versions of your pages, is what they're going to use to score and rank those pages in their index for all the different keywords that all those pages might be relevant for.
It used to be the other way around. Last year, the content on the desktop version of your pages was going to derive your mobile rankings on their mobile index. This big shift in what they're calling their mobile first index means that the content that is on the mobile versions of your pages is going to be weighted much more heavily.
The good news for marketers and the good news for SEOs specifically, is that Google is now allowing us to hide content on the mobile version. We can have HTML on the page that needs to be there for the desktop version, that needs to be there for SEO reasons, but we don't have to show it all to mobile users. Mobile users, by the simple fact of the intent of people who are doing these searches, whether they're investigating or they're looking to make a spot purchase on their phone, they don't need the same vast amount of content that a desktop user would be and they've got this much, much smaller screen and a much shorter attention span.
You have to think about the most common use cases of people on their phone, people at the office, in the bathroom, in their cars, standing in line at the grocery store. It's getting to the point, and again, it's using these design patterns that are enabled to encapsulate all the content that you need from an SEO perspective, but being able to hide much of that content from the display. It's still there on the page, it's on the HTML, Google can see it, Google can use it to score, then rank that page, but the user doesn't need to see it.
Louis: Let's talk about some of those design patterns in more details, and perhaps identify one or two. I don't want to fall into best practices that will change. I'm trying to identify, perhaps you can find a particular pattern that we know works because this is how people think and humans act. It's just the way people behave and therefor it's very unlikely to change. Can you identify a particular structure, something that you know is unlikely to change in the next two years, five years, ten years?
Nick: Yes. I think in the next five years, ten years is too far for this industry, but five years I think it's safe to say ... This is going to be an oversimplification but I think you might appreciate it. That is design patterns that are thumb-friendly.
Being able to use horizontal scroll to go through chunks of textual information, using vertically stacked accordions, so having a design pattern where a bunch of text content may shrink down to accordion-driven, because the user may not want to read all of it, they may not want to get all of those details as they scroll down past the page to either take action or move further into the conversion funnel.
Having design patterns where you are able to summarize the topic, summarize the content very quickly in five to seven words where if the user clicks with their thumb on that section of the accordion or scrolls left to right to get to that chunk of text for more information, it's there if they want it but you're not forcing them to experience that content.
Louis: I love that. I'm thinking right now about a company I'm not necessarily admiring, but at least they're doing it well, for now, is Facebook and they're on their way to do it on their mobile. You notice those kinds of things, those expand buttons, those accordions where you can swipe left and right. There's obviously also another example of a company that they use the horizontal accordion very well, or the movement of swiping left or right is Tinder as well.
There is definitely thinking of research and how people behave, which is the core of marketing, understanding how people behave and how they do things. You very well identify one pattern that is at the minute the way we use smartphones, which is the thumb.
Doing real life research when it comes to building your website and just putting those pages on mobile in front of people and just see how they will behave with it, will give you ideas on how to simplify it, summarize it, hide content that doesn't necessarily is relevant for people on mobile. This way you can really build something that people enjoy using right?
Nick: No, I think you're exactly right. You're hitting on the exact right way to come at this. Just to follow onto that, something that's a little bit more fun to think about, and you're a technical person. We're young enough that I think, and I don't want to insult you, I don't know if you're as old as I am, but I didn't grow up with cell phones. I had a pager when I was in 7th grade.
I look at my youngest sibling is 21 and cell phones were always there. She didn't have a pager first. There was never a period in her life when she didn't have a phone. What's amazing and fascinating to me, as somebody in digital, is to watch this evolution of the dexterity of the new power users of mobile phones and these power users of the internet.
I'm starting to see where we have this tendency to use big, thumb-friendly buttons, you're seeing and Facebook's one of the companies leading the way, you're starting to see much smaller click targets to expand things, smaller text links, read more, see more, view more, tiny things that require by relative comparison a very tight click area where you have to have a dexterity of your fingers. You can't just fat finger things anymore.
You're not using your thumbs. You're starting to see people who are using their thumb to scroll through very quickly, but if they want to drill down for more information, they're switching hands and using their pointer finger to hit a very much, much smaller link target. The reason it's smaller, you can scroll past it unless you have specific interests. This is one of those things in terms of the evolution of design patterns to be mobile first that I find fascinating.
Louis: It goes back to, I know I keep repeating myself and you might say, "What the fuck? Why does he keep talking about this?" But it's so important to me is that those principles are not going to change. Observing people in the wild, understand how people use your website or use your product or talk or whatever, it's a principle that you can rely on as a marketer forever.
However, technology moves fast obviously. There is a difference between trying to follow trends and saying, "Okay, I saw this article talking about best practices about mobile, and I really need to use those mobile frameworks that this article just told us to, because if we don't, our competitors are going to win against us."
Instead of thinking this way, just think people first in terms of how are my consumers behaving? How are my people behaving right now? And trying to observe that and seeing if there is any switch. This is how then you can really have leverage because you will know firsthand what you need to focus on next.
It's a minute difference, but I think it's so critical to focus on the right things so that you don't chase this shiny object syndrome or you don't get caught up in this race to the bottom of trying to follow so-called
best practices. I'm super glad you're mentioning this particular example. Is there any other pattern or thing based on how people behave that makes sense to use in your website or at least consider right now?
Developers got really excited about Angular or node.JS five or six years ago, but Google was nowhere near ready being able to crawl so you had to use solutions like Prerender.IO that would allow you to create a cached version of the page, so you had a version of the page for Google and a version of the page for everybody else. Google is moving toward this singularity in a sense, that I don't think those solutions are going to be required for much longer.
Louis: Right. There you have it folks. I think we talked about designing a conversion-focused website that looks good, but yet without having to sacrifice SEO and speed. It's been almost 45 minutes we've been talking about this in depth in quite technical terms. I think we did a pretty good job at going through this step-by-step. Nick, thank you so much for spending the time to do this with me today. I know it's not easy over a podcast to describe the basics as well as the advanced things, and you did a very good job so thank you.
I want to talk briefly about your history because I mentioned in your intro that you started your entrepreneurship journey when you were seven. In a very, how would you all that, in a very American way should I say? Young entrepreneurs, young kids starting a business when they are only seven, six, eight or something. If you have to identify the biggest business or marketing fuck up in your career, what would it be?
Nick: I've built and sold two businesses before this one and I've built and sold countless websites. Some of the biggest fuck ups that I did, there's so many fuck ups it's really hard to pick the biggest one. I've sold domains way too early, I've sold businesses for ... I sold a business for $30 grand. It was just a website that was generating leads for a very specific vertical.
I sold it for $30 grand five or six years ago, and it was just because I sold it to the first person that made me an offer. The person who bought it screwed it up a whole bunch, but they did a much better job of pitching it and finding a bigger buyer. I think they turned around and sold it for several hundreds of thousands of dollars.
There's been opportunities where I've rushed to get out of business because they're distractions. There's valuable lessons to be learned on both sides of that, which is I've got ADHD to the extreme, so I'm constantly at war with my focus and what I'm able to focus on. If I take on too many projects, I usually get to a point where I identify I need to chop some of them off and ideally sell them or find partners to take them over and run them. There have been instances where I've done that a little too quickly and left a lot of money on the table.
Another interesting one is I had an SEO blog that I sold back in 2014. I wrote about it on the I'm From the Future blog, but I had a blog that I made on a whim. It was called SEOnick.net. I hated the brand name, it was a stupid brand. I sold it because I was like, "I think this is a really stupid brand. I don't want this anymore."
I sold it for $100,000 just to get rid of it and the problem is that my name was so attached to it that to this day, people still attach me to that brand name. It's been four years, but people still identify me and attach me to the SEO Nick brand. In retrospect, while the cash was nice at the time, I probably shouldn't have sold it.
Louis: It's true because when I search for your name on Google today to prepare for this interview, one of the first results that came up was exactly this blog, this domain. Were you diagnosed with ADHD or are you just saying that you have ADHD?
Nick: No, I got diagnosed in first grade. It was believed that I had a learning disability. Actually I wrote all about it on another website of mine, ADDhero.com where I had my whole story about how the school that I was at thought that I had a learning disability and I had this major problem and so I went and got tested and they came to find out that it was just I was in the wrong environment. My parents moved me to a different school that had a program for people that were more like me. It helped change the trajectory of my life.
Louis: I guess your parents were quite happy to hear that you didn't have a learning disability,
that you just had this hyperactive personality instead.
Nick: I guess so. I was still a nightmare as a child, so I don't think it really mattered to them. They ended up choosing not to medicate me, which I don't know if they would still say was a good idea or not, 34 years later or 25 years later, however long it's been since then. I was definitely challenging.
Louis: Right. We can have another episode about this as well because that's also an interesting thing. I interviewed a marketer from Phorest, P-H-O-R-E-S-T, dot com, which is a salon software, for beauty salons. He has been diagnosed with ADHD as well, and I think in this industry there's quite a lot of people who suffer from this. It's good to hear your perspective on it. Last few questions I always ask my guests. What do you think marketers should learn today that will help them in the next 10 years, 20 years, or even 50 years?
Nick: Learn how to have conversations with engineers. You don't have to become an engineer. I'm not going to sit here and say everybody should learn how to code, because I think that's bullshit. Being able to have an intelligent conversation with somebody in an engineering department or head of engineering as a marketer, engineering is the future of marketing.
It's a big part of our brand. That's why we went with From the Future as a brand name. It's not going anywhere and the more that you understand how to be able to translate your marketing initiatives and the how's and the why's to the people that are actually going to be responsible for implementing them, the faster you're going to get buy-in across the organization and the faster you're going to get stuff done. Developers and engineers are a funky kind of personality. The smarter they are, the more difficult they can be to work with.
The difference there is developers have to like you. If they're going to do good work for you and they're going to find solutions, because you've got to remember, all developers and engineers are professional problem solvers. They can create problems if they want to, especially with the CTO. It doesn't matter if you're coming in as the outside marketing consultant.
If they don't believe in you and you're not able to have a shared language, a common ground, to translate your requirements to them in a way that makes sense to them and supports their goals and initiatives within their role and to fit their responsibilities, they will go out of their way to make sure they don't do what you need them to do for you to deliver to your client, until such a point that you get fired.
Nick: I did check out a number of episodes from Everyone Hates Marketers and I actually really like the program. That's why I responded to you initially about I'd be thrilled and honored to be on here. I really like what you're doing with this. I think it's a lot of really good conversations. I'm also very biased. I think one of the best resources for people in digital is Traffic Think Tank. Obviously, I'm wildly biased so that almost doesn't even count.
We've got some of the most brilliant minds in SEO and conversion and user experience in there. It's genuinely a community and people share a lot of stuff that they don't share publicly, they can't share publicly. I think that's really, it's become a tremendous place in that sense.
Outside of that, Matthew Barby's blog at matthewbarby.com, he does a fantastic job. Still, I think Brian Dean does a really good job at making marketing more accessible and then Ryan Stewart from Webris. He puts a lot of free information out there. He gives away a lot of free templates for a lot of free digital marketing focuses and I think he's very generous with his time online.
Louis: I still wonder how Matthew Barby manages to create blog posts and be part of this community you just mentioned, and yet he's the VP of acquisition?
Nick: Director of acquisition.
Louis: Director of acquisition for HubSpot, which is not a small company. I just don't know how he fucking does it, so fair play to him. He used to be based in Dublin. I saw him a few times at conferences. Thank you so much, once again Nick, to spend the time to talk to me today. I learned a lot from you. Where can listeners connect with you and learn more from you?
Nick: The best way to connect with me is on Twitter. I'm on there a lot. I'm very active and outspoken. I imagine just because you've dropped the F-bomb a few times in here, most of your listeners probably don't care about swearing.
Louis: No, they love it.
Nick: I was going to say, I swear a lot on Twitter, probably too much, but outside of Twitter I do something, I spend a lot of time in Traffic Think Tank, which again, it's a Slack group. It's not like a forum, it's very much an ongoing conversation so I'm in there every day for a couple of hours.
Louis: For someone with ADHD, Slack is a nightmare for me. I just cannot deal with Slack. I'd much rather have a forum actually, so it's not for everyone. I really, really wish I could join this group, but because it's on Slack, it's like I can't deal with Slack. It's the way it is. Right, Nick. Once again thank you so much.