Optimize Episode 011: Alex Birkett on Conversion Rate Optimization the Right Way

Learn More About Alex Birkett

LinkedIn: https://www.linkedin.com/in/iamalexbirkett/ 

Twitter: https://twitter.com/iamalexbirkett 

Website: https://beomniscient.com/ 

What to Listen For:

02:37 Alex’s background 

05:19 What is conversion rate optimization (CRO)?

07:26 CRO: Goal setting and KPIs

10:43 Creating high-performing call-to-actions (CTAs)

16:43 Tactical example: 60 days to improve CRO

24:07 How to: Identifying high-intent keywords

33:56 When to prioritize CRO

36:35 Does AdWords impact organic search or keyword rankings?

38:59 From traffic to prospect: how to pre-qualify your pipeline

40:54 Lightning question round

Episode Transcript

Alex Birkett (Speaking)
00:00
You can also do passive data mining via forums and communities to find topics that people care about quite a bit. For us in the content marketing and SEO space, right, we've got Traffic Think Tank, we've got Superpath. People are asking questions all the time in there. And, like, you see repeated questions asked. And if you can write a piece of content about that, and here's the important thing, triangulate that with a keyword that you can estimate the search volume for in the CPC and all of that, you can start to get a lot of clarity on what the real pain point is and where somebody is in the customer journey. And then the final piece is really on, like, your unique brand POV. So all of the rest is really market data and market facing, but then it's, like, what is your unique position on that? How does your product uniquely solve that? And if you can interweave that and match that with the customer pain point and the keyword research, that's, like, the center of the bullseye. So I'm trying to find those terms. 

‍

Nate Matherson (Speaking)

Hi, and welcome to the Optimize Podcast. My name is Nate Matherson, and I'm your host. On this weekly podcast, we sit down with some of the smartest minds in content marketing and SEO. Today's episode is unique from some of our other episodes. We are diving deep into all things conversion rate optimization. I always like to say that page views don't equal dollars, and CRO is a big part of any growth channel. I'm thrilled to sit down with Alex Birkett. Alex is a co-founder at Omniscient, an agency that helps B2B software companies with content and growth. Alex is one of the best when it comes to CRO and before starting his agency, he worked with a number of incredible companies like HubSpot in growth and user acquisition roles. In this episode today, I'm excited to learn more about his approach to CRO, how we run experiments, measure the performance of our channels, and ensure that we're generating qualified leads and not just traffic. 

‍

Ad Spot

This episode of the Optimize podcast is brought to you by Positional. At Positional, we're building tools for content marketing and SEO teams. We've got a great selection of tools for everything from content optimization to keyword research and technical SEO. And you can visit our website at Positional.com. 

‍

Nate Matherson (Speaking)

Alex, thank you so much for coming on the podcast. 

‍

Alex Birkett (Speaking)

Thank you so much for having me. I'm stoked. I love talking about CRO. I don't get to do it as much anymore, so I'm very excited. 

‍

Nate Matherson (Speaking)

Well, I'm excited to have this conversation. CRO is, like, a big part of any content marketing and SEO channel because once we get the traffic, we need to ultimately convert it because that's the reason we're all building these channels. And I would love to learn, like, a little bit more about your background. I know at your agency now, you kind of do everything from content to SEO and CRO. How did you get into this world? What led you to starting this agency?

‍

Alex Birkett (Speaking)

02:37

Yeah, starting way back in college, I didn't know any of these terms. I didn't know what content marketing was. I didn't know what — I think I read an ebook on SEO when I was in college, but it was really basic. I read bloggers like Tim Ferriss and Ryan Holiday and followed, kind of, their journeys. And Ryan Holiday in particular was interesting because before he did all the stoic stuff, he would write his lessons from his job at American Apparel. He would write his life lessons and political lessons learned. And I just thought that was really cool. So during college, I started my own blog, and I was writing about what I was learning in internships and in classes and from books I was reading, and I kind of built this little portfolio of which, you know, I was kind of doing content marketing for myself at the time. And then when I got out of college, I really wanted to get into a tech startup. I thought entrepreneurship would eventually be the route for me. So I, with that portfolio in mind, I got a job at this really early stage company called LawnStarter. It was pre-seed. I was sitting around the table with the founders trying to find product market fit, and they were heavily SEO influenced. So it's sort of like a marketplace for lawn care — so, like, Uber for lawn care. And on the demand side of the equation, we basically did a bunch of local SEO. So we did programmatic landing pages and then content. And that was largely, like, ego bait list goals, link building. So quite different from the B2B approach, but I learned, I cut my teeth there. Later joined CXL, which is a totally different approach, very premium, email-list-building style content, thought leadership. I rarely looked at SEO. Sometimes we would try to append a keyword to the topic, but really it was about talking about pain points in the industry. Worked at HubSpot, where it's like the machine of all machines when it comes to inbound. So I learned a lot about process, content ops, and content conversion there. I was working on the growth and premium acquisition side, so technically not on the content team, but I would work with the content team. I was always in and around content adjacent to it. And then I took a little detour, went to Workato, where I ran the experimentation team, the experimentation program, optimizing conversions towards the product and on the site, working on paid landing pages and onboarding. And then at some point at HubSpot, I had started the agency on the side with my co-founder, David. We later brought on Ali as well. We all worked at HubSpot together. How we started it was, basically, I was sitting at a bar in San Francisco during a conference, sitting with my old boss, Peep Laja from CXL, and he was complaining about how hard it is to hire a content marketer. He's like, “Yeah, I'm looking into these agencies and, you know, they charge this much, and here's their deliverables, and this and that.” And David knows — we were already doing content consulting just on our own. And we had talked many times about building a business together. We just looked at each other and we're like, “I think we can do that. I'm pretty sure we can do this.” So we just started tapping into our network and kind of built it slowly over time on the side. Eventually, it hit this critical mass where we're like, “All right, we gotta — we gotta go all in or call it a side project, right? This is either a business or we're just doing this for fun. Let's do it as a business.” And we all jumped in full-time probably last year, April, May-ish. So it's been about a year, a little over.

‍

Nate Matherson (Speaking)
05:19
Amazing. And, like, what is CRO or conversion rate optimization in your words? 

‍

Alex Birkett (Speaking)

05:25

Well, I'll start — it's easier to define something by explaining what it is not and the misconceptions around it. So a lot of people, including prominent executives, think CRO is just, like, a list of tactics. Or, you know, if you're a CRO expert, you're going to come in and be like, “All right, sprinkle some social proof here, add an urgency countdown timer here, optimize the CTA with this button color, and magically your conversions increase by 200%.” Definitely not the case. It's also not synonymous with A-B testing. So a lot of people think it's purely experiments, purely randomized controlled trials. In my mind, CRO is more of a framework or an operating system. So it's basically the scientific method applied towards optimizing a goal — in many times conversions, conversion rate, but it could be anything, right? It could be click-through rates, it could be reducing churn rate. Like, I look at it as a very generic kind of general term for an operating system where you take data and feedback, put it through a system and a process that often includes tools and statistical methods. And then the output is better decisions. And the better decision is usually around optimizing towards greater revenue, greater conversions, some sort of a business goal that businesses care about.

‍

Nate Matherson (Speaking) 

6:26

And for me in my career, like, CRO has always been incredibly important. At our first company, we had, like, an entire team dedicated to, like, taking our traffic and figuring out how we could extract more value from it. And at my first company, like, we thought CRO was not — I don't want to say easy, but it was easier. Like, we were marketing consumer financial products. We would throw, like, an ad for, like, a credit card in front of someone and, like, they would click on it. It was, like, a very high-converting type experience for, like, our readers and our customers and for us as a business. But then more recently, I spent some time in, like, B2B SaaS, specifically in, like, the devtools space. And, like, we found that, like, bringing that user through the funnel and ultimately converting them, it had, like, a lot more nuance and there were a lot more decisions to be made, whether it was, like, pushing someone to book a demo or signing up for a free trial. When you're thinking about, like, CRO for, like, your customers or in your roles previously, like, how do you, like, go about defining, like, what that goal or that conversion is that you're optimizing for? 

‍

Alex Birkett (Speaking)

07:26

Well, it really depends, right? Like, it depends on the business model. And I totally agree with you. Some models are easier to apply CRO to than others, specifically consumer spaces where you've got much more traffic, much more conversion volume — so basically your statistical power is much higher and you can see smaller wins. In the B2B space, like, especially for, let's say, like, an enterprise software company, you know, how many people can you potentially sell to? How many leads can you potentially have? So A-B testing is often, like, not even warranted in those cases. But the goal depends really on, like, how the product is set up, right? So a company like, let's say, Jasper, AI content-writing tool, they have a freemium version or free trial. You can optimize towards that. That's a relatively low-friction gateway into the product. And then basically all of the optimization and product management work comes from the onboarding up until the monetization point — versus a company like, I don't know, SAP or something like that, right? You're probably not gonna optimize for demo requests or talk to sales except on core pages where that intent is actually key. And that's another point, too, is, like, you're not always optimizing for the same goal across the site or across the product. It really depends on the context of the specific experience. But then you're really trying to map out a funnel or trying to bring people from maybe problem-unaware, solution-unaware, all the way down to the point where they're comparing your product among others in the space. And then you're optimizing towards that end conversion. But a conversion, a goal — I mean, it's really subjective. A lot of the times in content, we're dealing with content or conversion assets, things like ebooks, trying to get people onto an email newsletter. In our space, in my specific context, right, we have an agency, and we're gonna charge upwards of 10K a month, and that's not something that people tend to make on a whimsical decision. But we're basically looking at our email list as the freemium version of our product, so to speak, right? We're doing a weekly newsletter, we're talking about the things we're learning in the field, client case studies, interesting thought leadership, and you get to know us and get to know our processes. So if we can get people in the door through that email list, eventually on the email list, there's gonna be a certain amount of people who self-select as ready and qualified to work with us. And that's at the point where we can start sales conversations. And, you know, we've got a contact page and we can optimize that obviously towards, like, qualified leads and qualified conversions. But we're looking at approximate conversion much higher up the funnel, just based on our business model. 

‍

Nate Matherson (Speaking)

09:34

Yeah, I think you make a really good point there that, like, your conversion or what you're optimizing for should be different depending on, like, where in the funnel that page or that piece of content falls or, like, that user that you're targeting is. And I think this is, like, a big mistake that a lot of our customers make — like, they'll blanket have, like, the same call to action across every single page on their website when, like, these pages fall into many different stages of the funnel. And, like, one page might benefit from, like, an email collect, whereas, you know, a more down-the-funnel type piece of content where someone actually is looking to buy a tool might benefit from, like a, hey, “book a demo right now” type CTA. So I think that's a really good point. And I definitely would be, like, for all the listeners, thinking about, OK, where in the funnel do these visitors fall, and then trying to craft, like, specific CTAs to that point in their journey. When you think about, like, the different types of CTAs or, like, different types of conversions, like, email collect is, like, an obvious one to me, like, “book a demo right now” or “start a free trial” is, like, an obvious one to me. In your career, like, what other types of CTAs or, like, calls to action have you seen work well in the B2B SaaS space?

‍

Alex Birkett (Speaking)
10:43
All right. I'll get tactical in a second. But I think first off, like, a good heuristic is if you're just starting on this work, you want to try to optimize towards the best value, which is usually going to be more towards the revenue point. So if you don't have any data, if you don't have any content to, like, say, this is the conversion rate to an ebook, to an email list, to a demo, like, I would try to create a content strategy that maps towards that demo or that product signup or that actual purchase on the website and then work your way up. Because oftentimes people start with the assumption that they can't do that. And they'll start with very high-level CTAs around like, you know, “follow us for more insights,” “join our newsletter,” blah, blah, blah. When they actually probably could, you know, in some cases they could actually be driving more actual product users. So that's a good heuristic to start with. On a tactical level, let me think about CTAs. So first off, have CTAs. A lot of the times, we're speaking in the content and SEO standpoint now, a lot of the times content marketers are very shy, they're very sheepish around putting CTAs on the blog posts. They don't want to be too salesy. You would be surprised at how many CTAs you can put on the page and not dilute the user experience. Now, you probably don't want to go Neil Patel level, right? You don't want to, like, have 20 CTAs and pop-ups before you ever get to read the first paragraph. But on the other hand, a lot of the clients that we work with, we look at their existing blog posts and we're like, “Actually, before you even create new content, add a CTA at the bottom of the blog post, right?” If 20%, if 10% of people reach the bottom, my friend Brian Massey calls this a dripping pan CTA. It's just, like, when you reach the bottom, there's gonna be a very small percentage, but they're pretty high intent. They read the whole post, they're pretty excited to sign up. Just, like, it's really low-hanging fruit. So add that on there. And I'm kind of getting to a point here, like, I've got this framework for content conversion where it's like there's three kind of levers. There's the offer itself. And that you can kind of bucket those into, like, content or code. The code would be — and this is in a software context, obviously — but code would be something like demo request, product signup, anything related to the product itself, right? Like, that's going to be a very business proximate signup. And then content could be anything from ebooks to quizzes to webinars to podcasts — like, anything on the content side, on the media side. So you've got to choose based on the intent of the content. And oftentimes, like, if you're dealing with bottom-funnel content, you can go more towards the code, you can go more towards the business offer. But if you're doing content like thought leadership or let's say an article on, like, what is SaaS SEO, probably not gonna be able to sign somebody up for a product based on that. So I would look more on the content side. And then P would be, you know, position. So we've got O, offer, P, position. The position is really the real estate on the page. And heuristically, I try to do three CTAs on a page. You can test from there, but if you have something above the fold, I've seen it — at HubSpot, it was very effective to put the CTA in the author bio. So it basically, it would look like it was coming from a recommendation from the author that wrote the post. It'd be like, “If you like this, read this whole ebook that I wrote on this topic.” But it could be in the introduction, it could be a text CTA, it could be an image CTA above the fold. And then you wanna have a dynamic CTA. So that could be an exit intent pop-up, that could be a scroll depth pop-up, something that is behaviorally based. And then you wanna have that bottom page, that dripping pan that basically, you know, signals for the 10, 20% of users that reach that point, you want to have something at the bottom to capture that. And then the last P would be persuasion. So we've got OPP, like Naughty by Nature, right? And the persuasion is really the messaging, the design, how you actually speak to that offer. So if you're offering an ebook on the state of the AI industry, you could just say, “Grab our ‘state of the AI industry’ report,” or you could change the copy to say, “Afraid of getting left behind? Like, don't be. Here's our report that shows, you know, 20% of executives haven't even implemented this yet.” But there's a million ways that you can speak to the same offer. There's a bunch of different designs you can do, a bunch of different trick triggering mechanisms. So these are all the different ways you can sort of filter in CTAs. Tactically, a couple of CTAs that are pretty interesting to me in the B2B space:, I like to use different industries as inspiration because it breaks you out of the mold of doing the best practices and copying competitors. So in B2B, everyone's doing webinars. Webinars are easy, kind of banner blindness at this point. In the consumer space, it's very common for a D2C brand to trigger a CTA for a quiz, right? So if you're on a coffee website, it's like “Take this 10-step quiz to figure out which type of coffee is best for you.” You can do that in B2B too, and you can make it fun. We did that for instance, for our podcast: like “Get a customized podcast playlist based on your unique attributes.” And I think quizzes, anything interactive, anything fun in the B2B space, it stands out. And this is very ephemeral. I don't know if it's going to work next year, but that's a fun one right now. 

‍

Nate Matherson (Speaking)

15:00

No, I think that's really interesting. I actually haven't seen, like, a B2B SaaS company that we work with use a quiz, although we've got a handful of, like, consumer-facing companies who are very actively using quizzes as part of their CTAs. So that might be something we try here at Positional. And I do think you make a really good point — like, our customers will always ask me, like, “How do I increase conversions on this page?” And the first thing I'll typically tell them is just add more CTAs. Like, I agree with you that people are usually a little too timid. You know, if you're only using, like, a single CTA at the top of your header, you know, as soon as someone scrolls, like, they're no longer seeing that. And so like you kind of described, I think adding CTAs within the text of an article or at the bottom of the article or on the sidebar or table of contents that floats with you is one of the first things I would do. And then I also do agree with you that pop-ups — I know some people hate them and, like, don't wanna add them to their websites, but in my career, like, I've had pop-ups make, like, $10,000 a month in revenue. And so I know that pop-ups are actually really, really useful. Obviously, you don't want them to be intrusive and you should allow, like, your reader to click out of them and not disrupt, like, an experience from an SEO side of things. But pop-ups, yeah, they can be super effective and actually driving conversions. And there's a lot of easy testing that you can do, I've found, with pop-ups as well. Moving on a little bit. So say for example, assuming like my Positional website was getting 100,000 visitors per month from, like, organic search. And, like, we were to hire, like, someone like you to come in and, like, help us, you know, improve conversion rate. Could you maybe speak a little to, like, your process? Like, what would be those first things you would look at in, like, the first 60 days? Or, like, what would you do in the first 60 days to try to improve conversion rate across my website? 

‍

Alex Birkett (Speaking)

16:43

It's a broad question. We can break it down. First off, I would ask you to define what a conversion means across the website and whether that's the same across the website.

‍

Nate Matherson (Speaking)
16:52
OK, in my example, let's assume that a conversion is, like, a new free trial. And I've got, you know, like, a product, let's call it, like, $200 a month. And, you know, it's the more expensive product, but it's, you know, not, like, an enterprise $50,000-a-month product. And like, I'm primarily trying to get people in the funnel from, like, a free trial. Is that some helpful context?

‍

Alex Birkett (Speaking)
17:12

Yeah, for sure. And then the next question I would ask is if you have a data infrastructure that has been measuring the conversion rate across all pages, how much historical data you have, if there's any data anomalies, and if we need to debug anything from a data side. I guess we should probably assume for this exercise that everything's appropriately in place, but that would be something before I would ever start doing tactically anything. I got to make sure that the data, the telemetry, everything's in place so that we can actually track the changes because it's not always.

‍

Nate Matherson (Speaking)

At this point in time, in this example, we've got, like, basic Google Analytics set up; like, we've got some event triggers set on, like, the different CTAs, like, but that's about it. Maybe there's some things that we could do to go and improve the actual toolset that we're using or, or how we're tracking this data. 

‍

Alex Birkett (Speaking)

17:58

Yeah, yeah, totally. So as part of the exercise, I'm always going to go in and debug and audit the analytics setup. And in this case, because you are product led, you're doing a free trial, I would want to connect in some way to a database, your product analytics and your website analytics, just so we can also make sure — and maybe this isn't possible yet given, like, the sample size and the traffic, but we can potentially track downstream whether those free trials convert, and we can actually figure out, like, the revenue per experiment, right? In the beginning, it's probably fine just to optimize towards free trials, especially if you're just, you know, flexing that muscle. But eventually you'd want to connect those two systems. OK, from there, I'm going to basically build out a growth model. I'm going to find the baseline conversions, assuming that everything's set up appropriately in Google Analytics or whatever web analytics tool you're using, figure out baseline conversion on the website as a whole. The average actually — here's, like, a … this is going to be kind of snarky, just forgive me. But average website conversion rate, it's a useless thing. Because what you could do is basically you could cut off all the pages that are converting below average from a search perspective, you could cut off all the paid sources that are converting below average, and you could gate all of your content with an email gate or a free trial gate. And it's probably going to be a business poor decision because your overall conversion volume goes down, but conversion rate goes up. So we have to think in terms of context, right? And what I'm trying to do with that growth model is figure out, all right, so what's the landing page flow that leads to a conversion? Like are we going home page to landing page to pricing page to free trial signup page? Is that flow appropriate? What's the bottleneck at each step? Like, is the homepage, like, very low engagement? We can use a quantitative growth modeling to do this. We can figure out how many visitors are on each page, how many convert to the next step, et cetera. And then we can sort of, like, stress test using the model based on, like, assumptions. Like, if we increase the conversion rate by 20% at this step, what is the down-funnel impact of that? And basically we can start to signal map and figure out what areas of the site are most important. With that, I would also ask you qualitatively, you know, “Have you tried testing different areas?” Because I've worked at companies in the past where, yeah, the conversion rate was lower than it should be, let's say, on a given page, let's say the demo signup form, but they had been testing it for two years and basically had maxed out the — you know, like, there was not much else they could do. So it's, like, knowing that and having that documentation helps me say, all right, maybe we'll look at this in the future, but we're gonna ignore it for now because they've already done a lot of work here. And there's an under-optimized page that hasn't really been touched in a while. And maybe we'll focus on that, but I'm trying to impact map. With that, I triangulate with qualitative data. So let's say the homepage converts quite low. And let's say, you know, we use, like, a heat map tool or a user testing tool or session replays. And we see people don't even make it down the fold. And, like, there's a false bottom of the page, right? It makes it look like, you know, you can't even scroll. So, like, then there's certain things we can do to fix that and get them to scroll down the page, or we can lift the CTA up or, like, optimize the top. Like, there's things that we can do qualitatively based on that information. But I'm really just building a map of the territory to figure out where I should put my focus. Then there's gonna be more basic research to try to build, like, a roadmap of hypotheses. I'm trying to find broken stuff on the website, low-hanging fruit, things that we can fix and with a very high probability are going to increase conversion rates or at least lead to a better user experience. And I'm gonna prioritize those early in the roadmap. We may not even test those, right? If we've got things broken on mobile, we're just gonna go fix those first and that's probably gonna increase the conversion rate. And if it doesn't, it's at least gonna lead to a better user experience. And as you add 100,000 more visitors, it's gonna be a good experience for them. And it's gonna show up in the data. And then you're basically trying to gauge, like, what is the potential impact of a hypothesis? What's the confidence level that it's going to work? And what's the ease at which you can implement this? And there's tons of scoring models you can use. There's the ICE framework. There's the PI framework. I like the PXL framework from CXL, obviously because I worked at CXL and was there when we launched that, but it's based on objective criteria that tries to get you a confidence level based on binary scoring. So it's, like, has this issue shown up in user research? Like, is this page greater than 10,000 visits? You know, like, it's got a bunch of objective things that you can say, rather than say, like, how confident are you on a scale of one to 10 that this is going to work? It's like, I don't know, like 10? Like, I came up with the idea, so obviously I believe in it. And then I'm just going to go down that roadmap, figure out, like, a testing plan and try to start launching tests. And if you've got 100,000 visits a month, you've likely got, for a trial signup, probably, you know, at least 1,000 to 5,000 free trial signups. So you've got adequate conversion volume to run A-B tests, at least on core pages. And then if you don't, right, we're going to make looser decisions. We're not going to run, you know, proper controlled experiments, but maybe we're going to do some message testing or qualitative testing. Just to validate the assumptions before we push those changes live.

‍

Nate Matherson (Speaking)

22:20
Well, thank you for answering my very broad and difficult question. I think it was a good answer. And that we kind of walked through at first, like, the tooling that we kind of need baseline and then kind of walking us through the framework and kind of reviewing what work’s already been done, trying to identify those new opportunities or pages that could have the biggest impact and value. And I do think in that example, like, we've got, you know, quite a bit of traffic already. So I agree with you. There's probably quite a bit of experimentation we could do with, you know, 100,000 people coming to our site each month. 

‍

Alex Birkett (Speaking)

22:49

Can I add one more thing to the previous answer before we move on?

‍

Nate Matherson (Speaking)

22:51
Yeah, totally.

‍

Alex Birkett (Speaking)
22:52
OK, because the other thing that I was thinking about — I was speaking purely from a CRO perspective, but putting back on my SEO and content hat, a traffic-generation hat, I would also start to diagnose, like, the traffic that's coming in, right? Because conversion rate, it’s — there's two sides of the equation, like, the numerator and the denominator: it's a composite metric. So it's, like, the numerator, that's how many people are converting, and that's what I was focusing on, but really the people coming in, right? Like, if you are inviting in a bunch of unqualified traffic, like, you'd start to look in terms of, like, your acquisition strategy. And then it's like, all right, if it's SEO that's bringing in people, like, is it high intent? Are there opportunities to build out high intent pages? Is a lot of this unqualified glossary type traffic where it's like definitional terms, right? I would start to figure out, like, can we inform a better roadmap that brings in higher-quality visitors? Because no amount of CTA testing, no amount of design, is going to change somebody from being unqualified, unable to buy, uninterested in buying to doing so, right? You can't trick somebody into that. So I just wanted to add that point — is, like, the traffic, the people that you're bringing in, is the most important factor in conversion optimization.

‍

Nate Matherson (Speaking)

I think that's a good point. And so in the B2B space, like, from the content SEO perspective, like, how do you identify, like, those keywords or how do you know, like, what pieces of content to create that might be, like, higher intent to kind of help the denominator as you've described. 

‍

Alex Birkett (Speaking)

24:07

Yeah, I like to triangulate based on a couple different data sources. So the most common one is keyword research, and the common format or formula for that is basically search intent or conversion intent. So you can use CPC as sort of an indirect proxy for how valuable the traffic is going to be. And that's what tools like Ahrefs and Semrush use when they're calculating traffic value. It's basically like the estimated traffic per month times the CPC. That's one way. You can also, if you've been in this space for a while, you can start to, like, do SERP analysis and figure out, OK, this keyword is clearly bringing in people who are comparison shopping for a software product. They actually understand the space, right? And there's patterns to those. So it'd be, like, “HubSpot alternatives,” “best CMS software,” “best sales tools,” “HubSpot versus Marketo” — anything where they're, like, mentioning a product or a category is gonna be fairly high intent, but then you can map backwards and there's keywords in the middle ground, the middle of the funnel, often that I find are pretty propitious, quite lucrative as well. How-to terms, if you're invoking the product, all that product-led content, right? Ahrefs does that really well. So a term like “how to do keyword research” could actually lead to a lot of customers if you invoke the product and the content as well and show somebody, well, how to do keyword research with Ahrefs as opposed to just a generic how-to article, which also is helpful for the visitor, too. It's actually a better reader experience to actually show them screenshots and show them walkthroughs using, like, a concrete platform. But yeah, you're basically using search intent. But I also like to triangulate that with customer research. So pain points, figuring out what words people are using. So oftentimes in B2B, these tools, Ahrefs, Semrush, et cetera, they're very good, but they're not always accurate. And especially for low-volume, high-intent terms, you'll find oftentimes — I think it's called zero search terms. People talk about that, where it shows up as zero in the search tools, but then they write the post, and it gets 100,000, et cetera, visits per month. So it's like not all of the terms that people are searching for are represented in the tools. So you're going to have to do some customer research yourself, and I find go on calls, sales calls, customer support transcripts, customer interviews. You can also do passive data mining via forums and communities to find topics that people care about quite a bit. For us in the content marketing and SEO space, right, we've got Traffic Think Tank, we've got Superpath — people are asking questions all the time in there, and you see repeated questions asked. And if you can write a piece of content about that — and here's the important thing: triangulate that with a keyword that you can estimate the search volume for and the CPC and all of that — you can start to get a lot of clarity on what the real pain point is and where somebody is in the customer journey. And then the final piece is really on, like, your unique brand POV. So all of the rest is really market data and market facing. But then it's like, what is your unique position on that? How does your product uniquely solve that? And if you can interweave that and match that with the customer pain point and the keyword research, that's, like, the center of the bullseye. So I'm trying to find those terms.

‍

Nate Matherson (Speaking) 

26:44

Makes sense. Yeah, and I agree with you that, like, the keyword research tools can often be, like, incomplete, especially in B2B and especially in, like, categories like developer tools where, like, the clickstream data providers just don't have, like, a accurate sample of, you know, someone that's looking for, like, API testing tools. So I always say to our customers kind of similar to what you've described, like, to write down, like, the questions you're getting from your current customers and prospective customers. And even if in, like, a keyword research tool, there's not a whole lot of search volume for that keyword, if you have that strong gut feeling about a particular piece of content, I always still say to create it, because at a minimum, it'll give you something to share with your customers when you're doing customer support or share with your prospects as you're trying to convert them through the sales funnel. And I also agree that, like, the alternatives and tool-versus-tool type posts are very high, highly converting. When I was in, like, the B2B space with our large organic search channel, like, the alternatives and tool-versus-tool style posts would often convert the best, as well as, like, those tutorials where we're weaving, like, our product into that tutorial on how to actually solve a problem or question that someone has. Like, just to get tactical on my end real quick, a big mistake that I see our customers make with alternatives posts is that they will display themselves as, like, the only alternative. So it might be, like, an article like “Salesforce Alternatives,” and we wanna market HubSpot, for example, if that's, like, our product, and the single alternative that they show is HubSpot. But, like, a searcher for that keyword really wants to see, like, a list of alternatives. And so I always tell our customers, like, “It might feel uncomfortable, but if you really wanna rank well for “Salesforce alternatives,” you need to display that searcher a list of alternatives. And you can, of course, put yourself at the top, but, like, the search intent there is that someone wants to compare likely a large number of alternatives to that specific product or service. And I think the same goes with, like, tool-versus-tool content. Like, another mistake I'll see our customers make is they'll do, like, their product versus another product. And especially if it's, like, a startup and no one's heard of their product before — like, no one's searching for “HubSpot versus some new CRM that we've never heard of,” but they are searching, like, for “HubSpot versus Salesforce.” And so I would probably go after, like, “Salesforce versus HubSpot” as that primary keyword, and then kind of throw yourself into the mix as, like, a third alternative to consider. So I think it might feel uncomfortable for our listeners to, like, extract your brand, maybe out of an alternatives post or a head-to-head and make it a little more subtle, but I think those pages will ultimately rank better in search. Is that something that you've seen or is that something you generally agree with?

‍

Alex Birkett (Speaking)

29:20

Yeah, yeah. It's like — it’s the flip side. Like, marketers can be so stubborn. Like, the flip side that I was talking about before where they're so shy about adding their own CTAs and their own, you know, product in the content. It’s like this is the other side where they're like, “We can't mention the competitors.” But it's like if you want to match search intent in order to rank, and if you're writing a listicle of the best, you know, CRMs, and you don't mention Salesforce, you're clearly missing something that is industry, like, universal. It's ubiquitous — like, you have to mention Salesforce, right? So yeah, it's … a lot of the times, they don't want to mention any competitors, which, you know, dead on arrival, you're not going to rank for that. It's just not what people are searching for. That's a product page essentially. And if you're … that’s fine — maybe you can rank a product page, but that's typically not what's going to work in those comparison blog posts. Another thing is, like, trying to just find, like, all of the smallest and worst competitors. And it's like … you need to list the ones that people are going to expect. Put yourself at the position one, 100%. Like that was a huge fight that I had to make at HubSpot was, like, “Hey, let's put ourselves in position number one.” Totally fine to do that. Most people don't even notice the logo on the top of the site. I'll say that because we have these listicles on the best content agencies, and I get tons of leads, and I'm like, “Where'd you find us?” And they're like, “Oh yeah, I was Googling, and I found some lists that you were written about in. And I was like, “You know, we wrote those.” They're like, “No, no, no.” So it's always funny, like, how, yeah, people are very shy about putting themselves at number one as well. Your thing on the comparison pages made me laugh because it's like those, a lot of the times you see this on paid landing pages, but it'll be a comparison table, right? With all the features. And it's like your product has all of them, and the competitor has none of them. And it's like, so this competitor is horrible, right? Like, they're absolutely atrocious, but it's like, it's so unrealistic. Like, who trusts that, right? Like, and if you … this is a good argument for good product marketing. Good product marketing is not naive. It's not idealistic. It's realistic. And you're really pointing out your unique selling proposition, your unique features and attributes and who cares about those. And if you have those things dialed in, you can create a realistic, honest, authentic chart and really say, like, “Hey, in our case, this agency is a lot better at infographics and design. We're much better at conversion-focused content strategy.” Like, you'll know, and you'll know who cares about that, and you can start to target it in that way. But you can realistically say, like, what competitors are good at because they're obviously good at things, right? And you're obviously, or hopefully obviously, better at some things. Like, that's core business and product strategy, right?

‍

Nate Matherson (Speaking)
31:38
Right. Yeah. And I would totally bring that copy into the pages. I think it improves the quality of the content for the searcher and will ultimately lead to your pages ranking better in search, too. For example, if someone was searching for, like, “CRM alternatives to HubSpot,” and they were to land on, like, my page, which is, you know, an alternatives type listicle, a “best CRM tools page,” you know, it might make sense to position, like, a CRM that's the best for startups as, like, the best CRM for startups, whereas we could position Salesforce as, like, the best CRM for, like, enterprises or, you know, I'm sure there are many other CRMs out there that have very specific use cases — best CRM for doctors. So I wouldn't be afraid to, like, use honest language when describing your competitors and what they're best for but then you can use that same language, or different language, I mean, to describe, like, what you're best at. So that way, someone who's coming to your page — like, those users that actually do convert are going to be, I think, like, a much better fit or, like, much more qualified than maybe someone who is looking for, for example, like, a CRM for an enterprise, but they're really a startup. So do you agree that, like, you should kind of enrich your pages with, you know, “best product for XYZ” on those listicle type posts?

‍

Alex Birkett (Speaking)

32:50

I think that's a good summary and that's actually a good heuristic for content in general is to provide a TL;DR. You're going to want to back into why that's the case. So why is Salesforce the best for enterprises? I would explain that through features and, like, a deeper review of that. But I do, I like any time that you have, like, a TL;DR, because people tend to skim — especially on these listicles, I doubt they're reading every little bullet point that you add.

‍

Nate Matherson (Speaking) 

33:13
They're not. They're not. They're just reading the headline and then they're clicking the button, at least in my career. 

‍

Alex Birkett (Speaking)
33:20
They’re not. Yeah, for sure. So, I think that is important, and it's a good point too because, yeah, it's like there's no objective universal best, but that's like — people exist, like, you know, like, all across the spectrum. It's like, I like Blink 182; you might like the Rolling Stones. I like both, but you know, it's like everybody has different tastes when it comes to these softwares and, like, especially based on, like, company size or industry, like, there's going to be preferences. I like that. And if you can identify that, it's very helpful from a content perspective.

‍

Nate Matherson (Speaking)
33:43
Totally. And when does it make sense to start focusing on CRO? Like, say I'm an early stage company, I've been in business for five months. Should I be thinking about CRO right now? Or do I need to go out and, like, get more traffic before this actually should be a priority for me? 

‍

Alex Birkett (Speaking)

33:56

Yeah, it really depends on how you define CRO. And if we're attaching it to or synonymizing it, if that's a word, with A-B testing, you don't have to, you shouldn't do it right away. Like — it's a waste of resources. The expected value in economics terms is pretty low. So there's, it's a hard heuristic to set, but let's say you need a couple thousand transactions or conversions per month. And to your earlier point on pop-up testing, you can actually get those to an email list or something like that much faster than, like, a demo request. There's some brands, honestly, like, that may never get enough volume to run A-B tests on their core offer, right? If it's a truly enterprise, like … talk to a company that sells to airlines. How many — like, there's 200, 300 airlines? Like, they could probably never do that, but they can do it on their ads. They can do it on their email list. Like, they can do it up funnel. But I say, like, for statistical validity, you're going to want, you know, 1,000, 2,000 conversions a month. And it just — it gets easier as you scale. Like, when you're Booking.com size, when you're Microsoft, like, it's so much easier to do A-B testing because microchanges are actually going to show up in the data. Otherwise, at small scales, you're basically doing such innovative, complex tests and it's taking up so much time, developer and design time, that, like, it tends to lose steam. Like, people really lose interest in wrapping it up in an A-B test. And they're like, “Hey, let's just do really interesting stuff that’s going to move the needle.” Now, if you define CRO using, you know, customer research, qualitative data, like, validation through message testing, user testing, prototyping, that might just be under, like, a UX umbrella. But, you know, that mindset, I think, can apply even as you're starting a company from scratch, right? If you have no traffic to your site, I still think it's important to think in terms of, like, a search standpoint, like, from an SEO standpoint, what does this keyword mean in terms of intent? Does that match the business offer that I have? And, if not, like, is there a content product that I can create and do lifecycle marketing from that point to get them to sign up? And I think that mental model applies at pretty much any stage. And for ad testing, like, you know, if you're running paid Facebook ads, like, CRO is, it’s like innately intertwined with what you're doing because all of the traffic is so targeted to that landing page. Your conversion rate might be 10, 20%. Effectively, your statistical power, your validity is — it's much easier to see changes on those. So I guess that's a long winding answer to say it depends. 

‍

Nate Matherson (Speaking)

36:11

OK, cool, and I've got a couple quick questions before we jump into the lightning round. So my next question is on ads, because you mentioned it. And, a question that I'm always asked by our customers is “Does bidding on AdWords for, like, a high-value keyword that we also rank on the first page for, like, help us in terms of organic search?” And I'll usually say, like, “Those two things are completely separate.” Like, if you were to ask Google, they would tell you, like, organic search is completely separate from AdWords. But in your career, have you seen, like, any sort of positive or negative impact on, like, the organic search rankings for a page by also bidding AdWords for that particular keyword? 

‍

Alex Birkett (Speaking)

36:45

Yeah, there's synergies. It's indirect, I would say. I think Google has explicitly claimed that there is no direct causation between the two. It's hard to, like, tease out the causality of why this is. Some of our clients have seen increased conversions, increased traffic, because of that. And when they turn off ads, like, it goes down. I have a small sample size — put my confidence level, like, not incredibly high in this answer. I did this strategy at HubSpot called the surround sound strategy, where for terms like “best CRM,” you didn't want to just rank your page, your product page or your listicle, you actually wanted to be included in all the listings because you start to create this surround sound effect where people say, “Oh wow, this page mentions HubSpot, and this one does, too. And this one does too, wow.” Like, and you realize — it starts to accumulate in your brain. It's like this is somebody to take seriously. So I wonder if there's a mere exposure effect — basically seeing something, seeing a brand more than once creates this trust in your brain or brand recognition. That could be one thing. 

‍

Nate Matherson (Speaking)

That's an interesting point on surround sound and for all our listeners, like, if you want to be included in those pages that are ranking well already for, like, best insert-your-name-of-product-industry —  like “best CRMs” you can certainly go and reach out to the websites that have written articles on, like, the best CRM's and ask them if they will include you into their post on, like, the best CRMs, and, like, of course not all of them are going to say yes, but in my career, I've had, like, a fair number of them say, “Yeah, sure, like, we needed to update this post anyways. Like, we’d love to include you into our list of the best CRMs to consider.” So it's a pretty lightweight project. You could probably do this in, like, 30 minutes, if we're being honest. And you never know. Like, if you can get yourself included, you'll get, like, an awesome backlink likely, and also can help with that surround sound effect, like I described.

‍

Alex Birkett (Speaking)
38:23
Referral traffic, too. 

‍

Nate Matherson (Speaking)

38:24

Yeah, referral traffic too. It's really valuable traffic if you're getting included into a page that's already ranking well for a keyword that you want to rank for. And then the next question I want to ask you is on, like, bad form fills or, like, unqualified leads. Like, we've actually talked to companies where they're like, “Yeah, we don't want, like, more organic search traffic because it just, like, clogs up our pipeline. Like, we don't know what to do with it. They're all unqualified.” So how do you think about screening out or, like, cleaning the funnel? Say for example, you're driving a lot of organic search traffic, but it's actually causing, like, issues for your sales reps or for your pipeline. Is there anything that you can do to, like, try to combat that upfront?

‍

Alex Birkett (Speaking)

38:59

Yeah, two sides here. The main one is it's a qualification problem. Like personally, I'll take 20X the form fills and, like, I don't care if they're unqualified; like, I will qualify them. And that's what a good CRM does. That's what HubSpot does, right? You could do lead scoring models. We do it on the form itself. We actually have form fields. Our form is really long. We should honestly probably shorten it a little bit, but we ask budget, we ask the revenue, their employee count: like, a bunch of things. And we can basically say with a good degree of confidence, “Not for us.” We're relatively low volume though. So I guess, like, as you're scaling up and, like, you're dealing with thousands of leads per month, which we're not — we're in, like, you know, tens to like 50 a month. So we can handpick them. But yeah, there's systems, there's tools that are built for this qualification mechanism. But I think it starts with really defining what qualified means to you. And a lot of brands don't have a clean, clear definition of what that is. It's a ridiculous example, but it's like, you walk past a bar. A lot of these bars have these A-frames and it'll say, like, you know, “Free beer and false advertising,” or something like that. So it's like, don't promise free beer on the landing page if you're not actually, you know — if you're going to bait and switch somebody, because, like, that click-through rate, that form fill, it's just paper shuffling, right? So you need to set, like, principles or guardrails for the types of copy that you're writing, the types of messaging. I think within that sphere, like, there's a whole lot of room for creativity. So it's not like, hey, be, like, a very boring, like, you know, business direct, like, blah, blah, blah. Like, you can be creative, but just use good judgment and really ask yourself, like, is this the way a CIO would speak? Is this the language that, like, my target audience would actually resonate with? And if not, you know, just be honest with yourself and only use that type of language or only talk about that … for us, we're not going to write really, really basic content that maybe college students are reading. We're going to try to write content that VPs of growth, VPs of demand, VPs of marketing are likely going to read.

‍

Nate Matherson (Speaking)
40:47
I've really enjoyed doing this episode with you. If it's OK, we'll move on to, like, a quick rapid-fire round. Does that sound good?

‍

Lightning Round

Alex Birkett (Speaking)
40:54
Yeah, absolutely.

‍

Nate Matherson (Speaking)
40:54
All right, so my first question is on tooling. I know we touched on, like, Google Analytics as, like, a tool I've used, but is there a better tool out there? Should we invest in a certain tool as part of our CRO efforts?

‍

Alex Birkett (Speaking)

Probably, I feel like Google Analytics 4 now is, like, a whole conundrum for teams. We've still invested in Google. We've trained our team and have resources to implement Google Analytics 4, although it is much more complex than Google Analytics for non-analysts. I don't know what the alternative is for web analytics at this time. I feel like there's a couple B2B attribution tools that I've looked into, like Dreamdata. HubSpot's decent. I think HubSpot's really good on, like, the marketing-to-sales connect, and you can see a lot of interesting data on the account level and the contact level. It doesn't, to me, replace a good web analytics tool, but I don't know, I guess for now, try to use Google Analytics 4, and if it doesn't work, I'm sure there's an alternative that's better. 

‍

Nate Matherson (Speaking)

What's the biggest mistake you see companies make with their CTAs? Like, just one, very pointed. 

‍

Alex Birkett (Speaking)

Not using them, you know. Have a CTA; have a couple. 

‍

Nate Matherson (Speaking)

And as far as, like, content creation goes, like, how much does, like, an awesome piece of content cost here in 2023? 

‍

Alex Birkett (Speaking)

I'm having trouble with a quick answer in this because it really depends on the space and the subject matter expertise. You know, for … you were talking about developer tools. You get an expert writing a long-form piece of content on that, you could pay 2K to 5K, but a consumer-focused piece of content that ChatGPT could do — is that even, like, that valuable anymore? Like, one to 500? It depends so much. 

‍

Nate Matherson (Speaking)

OK, and then as far as, like, you know, transitioning a little bit to, like, SEO world, away from CRO world. Are backlinks, like, still important? Is that something we should be focusing on in 2023?

‍

Alex Birkett (Speaking)

Probably. At this point, we see it's effective in a targeted way, especially for earlier stage companies. Once you reach a critical mass, a lot of the stuff should be passively done through content creation and passive link assets. I don't know where this is going to go as AI gets more sophisticated, as ranking signals change. But right now, we still see the needle moving with targeted backlink building. 

‍

Nate Matherson (Speaking)

And as far as working with you, like, how can our listeners get in touch with you if they wanna learn more about your agency and the services you offer?

‍

Alex Birkett (Speaking)
43:01
Well, the first challenge is to be able to spell the word “omniscient.” So it's beomniscient.com for the agency website. You can listen to our podcast, which is called The Long Game, available everywhere podcasts are. And if you want to read my crappy affiliate listicles, it's alexbirkett.com. But I'm testing a bunch of SEO stuff on the website. That's the point of the website.

‍

Nate Matherson (Speaking)

43:21
Well, I will certainly read your crappy listicles after this podcast recording, but I'm glad that they're ranking well. And I've really enjoyed doing this podcast with you. Thank you so much for coming on. 

‍

Alex Birkett (Speaking)
43:30
Awesome, thank you so much.

‍

Ad Spot

43:33
This episode of the Optimize Podcast is brought to you by a special sponsor. If you're anything like me, you've probably got a lot of content that's not very well optimized, and it can be a total pain in your butt to optimize it and ultimately get it to rank better in search. And that's what Positional does. Positional has an incredible toolset for everything from content optimization to technical SEO and planning your editorial calendar. And if you don't know by now, I'm one of the co-founders of Positional, and I'd love for you to check it out.

‍

‍