November 20, 2014
Content Marketing: Analyze Thousands of Blog Posts via Script
In a recent Heavybit talk, Takipi Cofounder Iris Shoor shared her secrets on how she grew her company’s technical blog from zero to 70,000...
Okay, so I'm going to jump right in and get started. This whole talk is on how to create great content regardless of what stage you're at, and it's very focused on startups like all the ones in the room.
One of the biggest things that I've learned and one of the biggest tips I would have about how to do this is that it's all about engaging people. It's about engaging your customers and potential customers even when you don't have them yet because at the end of the day, that's what you need to build a business regardless of any other thing.
It doesn't matter what kind of business you are. You need customers, and those customers happen to be people, and so engaging them is the key to the strategy around content marketing and marketing as a whole.
One of the first things that a business should do but oftentimes most business people don't even do is ask who your customer actually is. Think through, in as much detail as possible, what you know about them, what you think you know about them, and what you think you don't know about them, as well.
In the case of everybody in this room and the target audience here, it's developers. Early on in a business, especially a developer-focused business, what you're trying to do is get that product validation that your product actually makes sense for people to use.
Early on, your customer is actually developers who are working at startups building SaaS products and mobile apps and anybody else that would easily adopt the product. They probably look like him, and they will adopt your product and sometimes get pissed off like he kinda looks, although he might be smiling, and give you, as you're probably going to expect from somebody like him, brutally honest feedback, especially when your software doesn't work for him.
One of the biggest questions that you would ask next is, where do these people hang out?
This is all to determine exactly how you want to do your content marketing, exactly where you want to engage them. They maybe hang out at hackathons, although that's not always the case. They definitely hang out at different developer meetups.
This is just from Meetup.com, and you can just go there and find relevant meetups and also create your own meetups, which has been very effective for many startups. The way I think about it is,
When you have a meetup, everything that's created there, everything that's spoken there, every question that's answered, every presentation that's given is content.
That's all fodder for you to learn and understand more about the audience.
Even more importantly, repeat it. Take that content and do something more with it. Hacker News is great. That's where a lot of developers hang out. Usually if you ask people what type of audience is on there, it's the cranky developers that like to speak up and talk about a lot of things. Especially the ones that are commenting, and that's great, because that's the kind of vocal voice that you really want to capture and understand, especially early on when you're validating your product.
And then obviously, most engineers google around and they end up on Stack Overflow. They probably have accounts there. Some of them are commenting, so it's another great place where developers obviously hang out, and for developer tools and when your customer is a developer, you want to go where they hang out.
Stack Overflow happens to be a place where they end up hanging out a lot, because engineering is hard, creating software is hard, and Stack Overflow is a place that has answers.
Next is, obviously, GitHub. There's very few companies now that are not using GitHub. This is a great example from Keen IO. They're a developer tool, they're an API, and they put out a ton of open source work to help people use their product and make it easier for people to use their product, and it has become a key part of their marketing, especially early on when they were really validating their APIs and what they were trying to do.
Next is something that many folks don't necessarily think about. And this is really key, because it's all about finding places that other companies that you're competing with or other developer tools that are also being built haven't found yet.
This could be things like forums. There's a ton of communities out there. There's also a ton of niche websites, and those websites have the audience that you want and oftentimes you're going to get a lot more feedback from those people when you engage with them, when you create content, and things like that.
There's obviously traditional publications that you can use as well early on just to start testing things. And last but not least,
One of the most powerful things is when developers are actually sharing with each other.
This could be developers that are on the same team sharing your product with each other because they're in different departments or different areas of the company or just the fact that they're sitting next to each other and they're sharing a lot of stuff, especially code.
When you start thinking about it like that, you can go sit there and identify the behaviors that developers are using around your tooling and understand how they're speaking to each other about it. This is very often a missed strategy of thinking about, "how do developers talk about our product to each other?" Think of that as marketing itself.
One of the biggest things on how you actually engage folks in any of these channels is kind of lost on many people because early on you're usually focused on building your product, not necessarily doing marketing, and also seeking validation from customers.
What works extremely well is when you actually share the details of what you're working on, why you're working on it, and are very open and public about it. One of the best examples is when people use the Show HN tag on Hacker News and show off what they have. So even a company like Codecademy that borderline targets developers but also targets people that want to be developers, the founder posted on Hacker News.
He was very specific about how he talked about himself and his engagement with the community. He was discussing things like, "I've been a lurker for a while. Haven't been a developer. I've built this tool, and I would love to hear what everybody thinks about it." It was a very long thread. I'm sure they got a lot of traffic. And they were even soliciting feedback specifically from the Hacker News audience by saying, "If you have anything that we can help you with, send us a email at HN@Codecademy.com"
This really started building empathy with the community and helped launch the site. This was one of the first postings they did when they first launched the company, and it was the founder essentially telling his story about why he did that.
I can't stress this enough, especially in a world where everybody is bombarded with marketing and there's ton of ads in front of you, even all the re-targeting that's going on when you visit a site and then the ad keeps following you all around,
The most powerful thing you can do is do the one thing that's most human, which is when we tell stories to each other.
So when you're early on and you're seeking the product validation, storytelling and telling your own story, telling the story of your early customers and what they're learning, how they're using your product, can be one of the most powerful things in terms of content and as a key tool that you have at your disposal and might not be thinking of.
I really like to share this graph, and this is from Google Trends, and it helps show the trend over the last, I guess, almost 10, 12 years of what's the popularity of content marketing and how it became a topic and became something that many of us are doing. It's really growing in terms of the volume of search traffic that it's getting, which implies that there's a lot of thinking going around on it, there's probably a lot of content that people are creating, and it's emerging as a key strategy for many businesses to use.
I like to say that people tend to ignore ads, but they really love great content.
That's why you're seeing a lot of things like sponsored content out there, because it's content, but they paid for it, so is it an ad or is it content? And so the reason for that is because people do love great content. They'll read it, and in this context just to clarify, I don't mean to say content is only words.
Content can be a presentation like I'm giving. Content can be a YouTube video. Content can be a Q&A that happens in person as long as it's recorded, written down, and you can reuse it. And content can even be any kind of copy inside of a ad I would consider content. So we are producing content all the time. I'm producing content right now as I speak.
If you just look around you, there's a ton of content out there, and it's not necessarily all in the form of advertising. And so it's really important to think of a broad view on content and also realize that the thing people search for, thanks to Google, is basically content. They're looking for ways to solve their problem. When they type in the box in Google, they hopefully will find your site. And that's such a big driver of traffic on the web.
So if you want to learn how to make better content, and one thing I really like to say is that there's a ton of content out there, and if you search for it, you can find it. This is a little search engine I built to find anything about content marketing you're looking for. Just type in the keyword in here, and it actually searches a bunch of the blogs and sites that I've created and helped create, and you can get some really great content about content marketing or whatever topic you want to search for.
Jumping right into marketing as a whole, one of the statements that I've found to be really impactful and useful when I'm walking into a company, any size company, and we're talking about marketing, oftentimes they have what I would say is a list, and it's a long list of strategies or tactics that they're considering doing in marketing.
And these are usually things that they're not doing yet. I just walk in there and say, "Do we know what's working already?" They usually say, "Yeah, this one thing," or, "These two things," or, "These three things are working." And then instantly, I can change the conversation just by saying, "Well, let's do more of that. Let's talk about that. Let's spend most of the time talking about what we're doing today and how we can do more of it."
Oftentimes, the founders and the teams, they look at me kind of funny, and they say, "Well, you're here to help tell us how to do more stuff that's not what we're doing." And my argument for them is,
"If you think everything you're doing today in marketing is being done as well as it possibly can be, then sure, let's go talk about new strategies and new tactics."
The number of times I've heard them say, "Sure, it's working great," is zero.
Every single time, we go into this discussion about what they're actually doing, and we find all these inefficiencies and all these things that they're doing really well but that they can scale without having to think of new strategies and tactics. So this is a very simple rule that I use, and it works for every single company I've been to, especially when they're thinking about, "We need to do new things." Usually, it's, "We need to do more of what we're already doing, and we need to do it a lot better."
This is another fact, especially when we get into the realm of content marketing specifically, or even advertising, is that you can easily come up with new tactics and new strategies just by looking at what's out there and figuring out what's working for other people and then imitating it. And I don't mean you rip it. I don't even mean you copy it and then tweak it.
I literally mean you get inspired by it and you use that, you learn from it, and then you go create your own strategy that fits your company culture as well as your customers' culture.
I'm speaking to engineers and engineering-minded folks, reverse engineering is the key to this strategy. Realize that all the content that's out there, especially in the form of blog posts, SlideShare, YouTube, GitHub if you want to consider code as content, which I do, especially open source code.
It's publicly available and you can understand how many times it's been forked, how many stars it has, how many retweets it has, how many shares it has, and all those kind of things, and that can help you understand, what does success look like and what does failure look like with content? This data is publicly available, and you can use that to inform your own strategy.
So I'm going to give you some tips on how to actually look at this content space no matter what kind of specific category you're in. So this is a really cool trick in Google where you can type in a keyword, and then in the top bar where you have Images and Video and all those tabs, at the last tab, you actually have something called Search Tools.
If you click on Search Tools, there's a dropdown that basically says, "All time." And you can go in there and say, "I want to only see all the results from the last 24 hours, or the last week or the last month." And when you do that, you have turned Google into a search engine for the most recent content that's out there about that keyword, which is a lot different than how we use Google, right? We type in something in Google and we expect the best results.
In this case, when you're trying to understand what's out there and reverse engineer it, you need to know what's recent, not necessarily what's the best because you want to understand, what should you be writing on? And this kind of data and this lens on it can really help.
Next up is a product called BuzzSumo, and this product is great because they've basically indexed and crawled a lot of the Web and found all the blogs out there and basically created a search engine for it. And they don't just show you a listing of all the content in any keyword you type in. They actually let you sort by popularity.
They let you sort by popularity in terms of number of shares it's had on LinkedIn, Facebook, Twitter, Google Plus, and Pinterest. And once you do that, and you can even sort by all of them combined, you basically are able to tell what's popular, what's not, what's worked, what's not, and it's super easy.
Again, you can find popular content. In this example, you could find popular content that a specific site or blog posted. So I put in DigitalOcean.com and then I was able to see all the content that they've posted and how popular it is. This is so important and so easy to do, yet most people don't even do this, these simple searches, to figure out, what's popular content that my competitor has written or a related company has written or that is popular for a certain keyword?
This is another example from PagerDuty. You can do the same thing with many different sites, and so if you're either competing with PagerDuty or want to understand what kind of content really worked for them, you can do the same thing.
Another great one is this site, Algolia, which is a developer-focused search API and platform, and they've created hn.algolia.com.
When you type something in in their box, they let you sort by popularity and a few different factors, and that allows you to see on Hacker News, if you think Hacker News is a great channel for you, especially early on, what are the popular posts? And then you can go read them and try to understand, why are they popular? And then you can essentially imitate it.
This was a really big find when I was looking through it recently, and I was able to find, as I was searching on Algolia, I was able to find a big list of YouTube channels that developers recommend. This is huge. This is like a meta strategy, where I was searching Hacker News for a bunch of content, and then I found another channel or a set of channels because of it.
I figured out that there are very popular YouTube channels that developers actually recommend. And this will allow you to get even more content ideas. So, just use these tools and make sure that you're actually searching and using them appropriately.
One of the reasons that you want to go search for all the content that's out there is because there's a very simple rule of content I like to use, and it can at first shock people, because
My thesis here is that you can't just create great content. You have to also create a lot of great content if you want to really get scale and get eyeballs.
There's a very simple reason for that. If you go out there and look, there's a ton of content in almost every category, and you can say we're drowning in all this content. But yet, people keep producing more and more content. And if your content stands out from the rest, that's great, but if you're putting out one piece of content a month, that's not enough.
You're not going to get enough customers. You're not going to get enough eyeballs. You need to learn how to create that great content and create a ton of it. So obviously I get asked, "How do I do this? Why am I not doing this already?" And a lot of the other questions are along the lines of, "This is so difficult."
One of the key reasons that you should understand that's first and foremost, most important, is that when you create high quality content, people remember it, and they might even come back to your site because they want more of it. So you don't want to go create high quality content and then have low quality content on the same site, because that's just not going to work.
You can't create a lot of great content unless you create enough content to understand what your audience really resonates with. And when you don't have the quantity, you can't do the AB testing and you can't test for conversion and you can't increase the amount of signups you get from that content. Another big, big piece here is that you're not going to be able to get great search results and great rankings unless you create enough content for Google to pick up and rank you.
If you just have one piece of content on your site, and it's a great piece of content, it's targeted really well with Google search, but that's the only piece of content you have, they're likely not to rank you for it, because you need more content, you need more links. You need all that good stuff to be able to actually rank in Google.
Another piece of advice on content and quality is that when you have a lot of content, you are essentially able to build a brand faster, because more people, and more often, people are able to get to your content, read it, see your brand, and hopefully get excited and really come into the brand. If you posted a piece of content every day and you sent it out to whatever email list you had, whatever audience you had, there is a high percentage of people that are actually going to read it every day if it's great content and if it really aligns with what they want.
That's what builds the brand, not by just posting once a month. It's actually by posting a lot more regularly and getting in people's heads. This is another pet peeve that I've seen and something that people repeatedly do, but if you've defined your customer segment and your audience and you're creating these company culture posts.
What I mean by that is posts about how awesome your company culture is or this new cool thing you're doing around product development, but it's not related to your business or your product, then you're essentially wasting your time targeting an audience that you don't care about.
Instead, you should focus on your audience and focus on that engineer I showed earlier instead of writing these company culture posts. But everyone decides to do it, because these are some of the most popular posts on Hacker News.
So jumping into the process and workflow around content, one of my companies, Quick Sprout, we did a ton of research, talked to literally, whether it's online surveys or interviews, we talked to about a thousand plus people, and we came up with this content marketing pipeline, or content production pipeline.
Every piece of content you create, again, doesn't matter if it's a blog post or a video or anything like that, it goes through these seven steps. And when it goes through these seven steps, there's different amounts of time that people spend on the different steps, and one of the most common, easiest things to do is come up with the topics.
If you use some of the tactics I mentioned with BuzzSumo and Google search and searching Hacker News, you should be pretty easily able to come up with some topics that are pretty popular. After that, you would pick one that you want to work on, research it, which means you're googling around, finding all the related content you can so you can link to it or imitate it, depending on what it is.
Then you actually create the real content. From there, you review and publish it, make sure it's great, or as much as you think and your team thinks, and then you promote and distribute it, and then you want to know how well it's doing.
One of the biggest things that people don't spend enough time on is promotion, because if you create great content and you can't promote it, did you really create the content? If nobody sees it, does it matter? No, it doesn't matter. So, what you want to do is you want to come up with tactics that make sure that you have some process where you control the ability to go message people or send the content to people.
So, emailing relevant people that have blogged about what you wrote about, shared what you wrote about before, is a key tactic. People love getting email, believe it or not, especially when it relates to something they care about.
You have to craft a carefully worded email and make sure it makes sense for the people you're sending it to, but it's a really easy way to promote, especially when you don't have much of an audience and you're in your early stages looking for product validation.
Another key strategy is if you start interviewing people or do day in the life projects where you do a day in the life of a customer or day in the life of a sysops engineer if you're targeting sysops people. Another thing that works really well is finding, and this doesn't work well in other categories, it works really well with developers because it hasn't been overused, is finding influencers and figuring out how you can work with them.
Developers influence other developers unlike most other categories inside of a business.
Marketers don't necessarily influence other marketers because they're all trying to compete for the same traffic usually, while developers, they just want to get their job done, get it done more efficiently, get it done better, and they're always trying to keep up with what's going on and the new technologies, and usually they're looking to other developers to help them do that.
So that's still a key strategy in this category, and I don't want to underscore it or spend too much time on it, but it's something that I've seen work really well when you can find the right influencers in the developer communities you're going after and you do things to help them promote your brand.
This is a really, really impactful way to take one thing you've done and turn it into many things. If you create a lot of words and you do an interview with a founder or a developer that's related to your topic and in your target customer set, I would recommend that you actually record it and turn it into audio and make it maybe a podcast or share it with your audience in some way, or actually video record it and put it on YouTube.
We've seen people take words from a blog post and then turn it into presentations on SlideShare and then basically share it out there.
Repurposing your content is totally worth it.
You might be thinking, "Well, if I create a piece of content, I repurpose it, and a bunch of people see the blog post and then the same people see the slide decks, are they going to be pissed off or something?" It's actually not the case, because usually the same people aren't hanging out in those same channels.
When you have different modes like taking text, turning into slides, or using video or using audio, people actually appreciate it. Different people learn in many different ways, and these kind of things help you actually scale your marketing even when you're really small because you can turn one piece of content, one 20 minute conversation, into many, many pieces of content, put it out there, and get a lot more traffic as a result.
I can't have a conversation about content marketing without talking about social media. So here's a graph showing the trend in social media, very similar to the content marketing graph. As you can see, it's up and to the right, and it's growing.
One of the biggest things that social media helps you do, especially early on, and what I mean by that is like Twitter or Facebook or Pinterest or sites like that, especially if they're targeted at your target audience, I would say that even Hacker News is a form of social media, what they help you do is they help you build brand and help you create links.
What you're trying to do is get as much exposure as you can. Social media is one of the fastest ways to get exposure, and once you get that exposure, let's say you put a piece of content out and you get 30 or 40 retweets, it's likely that that's going to turn into one or two links back, or even more, because people see the content on social media and then they decide to link to it on their blog or wherever else.
This is sort of bittersweet for me. I am under user 5,000 on Twitter, and recently at Quick Sprout, we started analyzing a lot of website traffic for about 200,000 websites, their Google Analytics data, and what we learned is that there's very few sites that are high traffic that get most of their traffic from Twitter.
There's also a very low likelihood that you're going to be able to get repetitive traffic from Twitter. And the reason for that is one, their platform is not growing very fast at all. So if they're not growing and you're trying to do marketing using Twitter specifically and trying to get traffic back, it's likely that you're not going to get more and more traffic over time.
They also don't have an algorithmic feed like Facebook does, so if you post something on Twitter, there's a likelihood that only a low volume of your audience is actually going to see it, because it depends on who's on Twitter right now. It's not like they come back to Twitter and they get to see your tweet from like two days ago. They don't get to do that, while on Facebook and a lot of other platforms, that is what happens.
Twitter is not necessarily the best place to go invest your time or energy when you're trying to get traffic. If you want to engage with your audience and start tweeting at them and have conversations, Twitter is absolutely great, but don't expect that to get you traffic. Early on, that can help you with product validation.
You can search Twitter for people that are frustrated with other tools like yours and go engage with them and things like that. A lot of early adopters and influencers are also on Twitter, and so those are all good ways to use Twitter.
If you think you're going to generate repetitive traffic and consistent traffic from Twitter, you have it wrong. Please don't waste your time.
If you feel like you can engage your audience on Twitter, talk to them, tweet at them, and learn from them, that's definitely a key way to use Twitter. It won't scale, but it can really help you early on.
So, let's say you've got validation and you've got your early customer set that happen to be startups, and now you're reaching product-market fit and you're ready to grow revenue and increase the type of teams that you're targeting and you want to get customers that are larger, maybe dev teams that are larger, basically your target customer becomes the CTO, the VP, director of IT, technology, engineering, depending on exactly what size and what type of business you're going after.
And in their case, they look like this guy who is the VP, I believe he was the VP of engineering at Google for a very long time. And in their case, their use case is they want to make their development teams faster. Usually they want more code written faster. And they want it done cheaper. And one of the key differences between them and a little startup that you might've been targeting early on is that they want dashboards and administrative tools and things like that so that they get better visibility into what engineers are doing versus less.
Key ways to attract them are to basically think about them as another type of customer set and do a lot of the similar things that you did with your original customer set when you were doing product validation, but realize that these folks are different. And so one of the things you need no matter what is, unfortunately I'm showing Oracle which I know many of us probably are not into, but they do have what I would call enterprise-grade product.
They do attract enterprise customers. They've been very good at it over the years and continue to be. They have a large enterprise sales team. So if you don't have an enterprise-grade product yet, then go talk to these type of customers or go understand what that looks like. There's a lot of content out there about how to create this type of product now and what you might need to add in your product if you haven't already.
Another thing you're going to need is you're going to need an enterprise sales team. And this is just an example from Optimizely and one of their positions that they're advertising for or were advertising for recently.
If you don't have an enterprise sales team, it's likely that you won't be able to sell to the enterprise.
Early on, it's probably the founders or some early folks on the team that are going out there and talking to these customers, because you're trying to figure out what the enterprise needs. As soon as possible, once you really need to sell to the enterprise, you need some level of enterprise sales or you probably won't be able to scale the business.
This is a pattern that many folks, especially in the Bay Area, don't really like to hear because they feel like they can do it differently than how it's traditionally been done. The truth is the enterprise buyer buys from enterprise sales teams, especially when you're trying to get hundreds of thousands of dollars a year from them. They want that touch. They want that special attention that usually only an enterprise sales rep can provide.
Another thing you're probably going to want to think about is do what PagerDuty did and create white papers, because traditionally, white papers help the enterprise buyer buy from you. And webinars. These are things that you would think that an audience like developers might be a little allergic to, because they have to go spend time on things that are not code, but in the case of the exact buyer that you might be thinking about in the company, they're probably into learning more about how to manage their teams better or how to use your type of tool to do a certain job.
Webinars are definitely not a waste of time for them, especially if they're really trying to solve a key problem for them and want to learn more about the different companies in the space.
And if you don't have a webinar but some other company does, it's likely that they might win the business simply because they had the opportunity to educate the customer and you didn't. Another great thing, this is an example from New Relic, they throw a conference called FutureStack, and in their conference, if you look at the topics and the way they think about it, even the idea of calling it FutureStack, they're really speaking to the enterprise buyer.
They're speaking to the buyer that has the budget to send half their team into the conference so that they can learn all about what the future stack for development looks like, hence the name. It's a great name, and I think they've done great branding around it, and they're a great example of a SaaS business doing their best to move up market.
I like to say is that you might think that content marketing only attracts the startups or the small medium businesses, but every single example I gave I would consider some form of content marketing, from the webinars all the way to anything you can do to go talk to the customer and learn from them. You're creating content. You're learning how to create content. You're learning about the customer. And then you're using that to basically market to more of them.
This is one of my favorite graphs, because it looks like SEO is flattening out, and this is essentially search engine optimization and the idea of basically optimizing your website, optimizing your content, for search. But then when you layer it on and you go look at, okay, content marketing, which is in red, content marketing looks really small.
And then you layer it on and you add social media. Social media looks really large, but when you think about social media, even a consumer, not necessarily a B2B person, not necessarily a tech person, not necessarily someone that's trying to market a business, they're typing in, "Social media," because it's something that's in our vernacular and there's many different social media sites.
A lot of the growth in social media can be discounted because much of the search volume is actually coming from consumers, not necessarily people that are trying to do marketing. And with SEO, it's still tried and true and has always been one of the most popular things out there when it comes to marketing, and there's a very good reason why.
When we did our analysis of these 200,000 sites, there were only a few dozen sites that didn't have enough traffic and didn't have search as their top source of traffic. This was really enlightening to us. It was kind of funny for us, too, because one of our first companies was a consulting company providing SEO services to other businesses, and this was back in 2003.
So it's been about 13 years since then, and still,
Google search, organic search, ends up being the top driver of traffic to every website we basically saw.
Search just boils down to two really basic things that I'm sure many of you are used to and learn about, which is basically content and links. And over time, the barrier to getting links has gone up in terms of, you can't do these link farms where you would create your own websites and link to all of them because Google has figured that out.
The level of content that you need to create is much higher, otherwise people won't link to it because in most markets today, in most categories, the content that's created is actually pretty good and there's a lot of it. So being able to stand out from the crowd is really important.
Even when we first started with SEO back in 2003, these were the two basic building blocks, and even today, these are still the two basic building blocks. Just the quality of what you have to do in these two areas is much higher.
So here are a few caveats on traffic sources. I'm going to go through them pretty quickly. One of them is that, in Google Analytics, you'll see direct traffic. There's always a category for direct traffic, and most of it is unattributable. You cannot go find out where that direct traffic actually came from and know the percentages of traffic, types of reasons people come there, come to your site.
For example, if a developer told another developer verbally about your site, and then the developer that didn't know about it typed it in the browser, they're going to go directly to your website and that's going to be considered direct traffic, but the truth is, they didn't just make it up and come to your site. They actually heard it from somebody else.
That's the kind of data you can't get. That's why early on if you learned how people are speaking to each other about your site, your company, your business, you'll be able to have a much better idea of where this direct traffic comes from, where this unattributable traffic comes from.
Another thing related to direct traffic is that social traffic is very hard to pin down as well. The reason is much of it ends up being direct, because when someone clicks from a social site, sometimes from the Web and oftentimes from a mobile app, it goes straight into from an app straight into Safari, and once that happens, you lose the referrer and you don't know that they came from Twitter, but you might know that they're a mobile user.
And so it's very difficult to track that kind of traffic. This is the caveat on traffic sources and some of the rest of the data I'm going to share. So, when you look at different size sites, we did this study from 2015, all these four sites that we had access to, and we thought about, how do we want to represent this data?
So what we did is we took the top six channels and identified the different percentages of traffic that a few different sites get if they're getting about 10,000 or so visitors a month from all the traffic sources. And if you notice, one of the key learnings here is that everyone has a pretty high amount of direct traffic, hence the nature of it being unattributable and hard to pin down.
That being said, the software company has 44% of their traffic coming from Google search, organic search. And the reason for that is, most software businesses are in a niche compared to consumer businesses or consumer websites like Buzzfeed. And so even early on when you only have like 10,000 visitors compared to a podcast or a brand or a personal site, you're able to get much more search traffic just because there's a lower volume of the search traffic, there's more googling happening at that stage and it's easier to get rankings because it's a niche.
And so one of the biggest things is that another factor is the fact that early on, it does take time to build that search traffic up, and so as a result, you're going to see a tendency for a lot of traffic to actually come from social and social media before it actually comes from search engines for any site. So it can take up to three to six months for a new website for the search traffic to kick in in a meaningful way where it starts delivering double digits amount of traffic.
And so when you start scaling and you get up to, let's say, up to like 50,000 visitors or 20,000 visitors a month, you end up seeing this really key pattern. All of a sudden, 60%, 85% of the traffic, especially for software companies, comes from search. One of the things that's really interesting, and I'm going to go back to it, is that this Blog C, which is actually a consumer blog, gets 46% of their traffic from email.
Software B gets 13% of traffic from email.
Email is a highly underutilized channel in most content marketing efforts.
One of the reasons is, if you put out a blog post every day, there's an argument to be made that you should send a weekly newsletter, not a daily one, with that new content. The problem with that is then you only get one shot every week to get people to click on your content, and then they have to click on all of it or not any of it because it just comes in at one shot.
Oftentimes, it comes in when they're not at work and it doesn't make any sense to them, and so they don't click on it, open rates are low, click rates are low. Later on, though, what happens when you start actually sending the blog posts every day and sending one at a time, people actually start clicking on it and they actually start opening it. And I know it might sound a little counterintuitive because we all get a lot of email, but
The truth is that the more email you send, the more traffic you're going to get, and everybody wants more traffic.
And then this is the blended breakdown. So if you look at all the sites, regardless of what type of site they are, who they're targeting, you end up seeing this very key pattern. One, search ends up being the number one driver and the number one traffic source for the majority of sites, especially when you blend all the traffic into one.
Direct traffic is number two, and number three tends to be social. Then we get into referral, which is other sites bringing you traffic. And then there's email if you have a email strategy, although in our research, we saw that most companies don't have a strong email strategy. And then obviously last but not least is any other traffic sources that are bringing traffic to the site.
Here are a few key insights about SEO and how you should think about it. If you focus on SEO early on, that's not necessarily going to help you except for some of the basics, like having good title tags, good URL structure, and creating actually high quality content that gets shared a lot on social media or on Hacker News or Reddit can be really useful.
The reason for that is that's how you get exposure before Google actually learns how consistent your traffic is, how consistent your content is, and then starts giving you more search traffic. It just takes time. And it consistently leads to higher traffic websites.
Another thing is that once you can get SEO, the metric to look for is once you can get SEO traffic to 50 or 60 percent, which takes some time and it's not necessarily the easiest thing to do unless you're investing heavily in content and link building, then you can almost guarantee that the other channels start going up.
We haven't been able to find the causation of this, but there's a strong correlation when we look at all of our data. 100% of the sites that had 50 to 60 percent of search traffic or more, they had a lot more social traffic and a lot more referral traffic, too.
Another thing is, this is not exactly applicable, but an important data point. Sites like Buzzfeed and news sites tend to actually get much more traffic from social media because they have repetitive content that works really well with social media.
Unfortunately, when you're targeting developers, that's not the case. It's still a big SEO play, or, early on, some of the very niche or Hacker News-type sites that can drive developers to you.
If you are not a news site and not a Buzzfeed-type site, SEO is going to be the key driver.
It's going to take some time, but this is the reason to get really good at creating high quality content, because people are still typing in keywords in Google, especially enterprise customers who are researching a lot before they're willing to spend a ton of money. And that's in most cases. There's tons of caveats to that, tons of examples that are counter to that, but generally speaking, if you focus on search, focus on the keywords that your customers are typing in or potential customers are typing in, you're going to end up winning the search traffic.
Here's a few examples. So the top graph is one of our sites. The bottom graph is also another one of our sites. What you'll notice is that early on, it takes a while and the traffic doesn't really get there, but then as you see, there's a few inflection points where there's just a bump, and all of a sudden there's more traffic.
This is all specific to just search. This is just search traffic to those sites. But as you can see you're talking about six or nine months in before all of a sudden Google wakes up and says, "Hey, we should rank this site. They seem to be creating great content and people keep loving it, and when people find it on Google search they actually click and stay there and they don't come back to Google search."
Google knows its data. They're monitoring all the keywords all the time, and it's all automated, and so this is what's happening now. In the past, this wasn't the case. You could get search traffic much faster. Now, Google is essentially penalizing new websites or new companies that are out there and just starting to do content marketing and they're not hitting their stride for a while, so it can be something where you can be frustrated in the beginning, but you should realize that you have to keep investing in this for Google to end up loving you eventually.
The good news is, early on when you're seeking product validation for the first six months, you should still be creating content if at all possible, and you should start doing that. It doesn't matter as much, because you're really more one-on-one talking to customers, this is like before you have 100 customers, and really trying to figure out how to tweak your product. So search traffic and all that stuff is not important.
But later on, as you start scaling and you get to the more sophisticated buyer or the buyer with more complex problems, they are definitely searching on Google, and if you show up more and more, you're going to get more of that love and you're going to get more leads. So the quick TLDR on that is that search traffic does require a maniacal focus on creating high quality content.
This is the reality today, because every trick in the book has been tried that isn't about high quality content, and Google has squished it down and said, "That's not okay." And they've penalized people that have used it and folks that had really high search traffic that were doing gray hat tactics and all these link farms and stuff, one day they wake up.
Google decides they don't like that tactic or they learn how a whole bunch of people are doing it, and they lose all their search traffic. And when your site is reliant on search traffic and it's 50, 60 percent of it, it really sucks.
The best thing to do to combat that is not do anything that's gray hat or gray area. Instead, focus on creating high quality content for your audience, and you also want to focus on user experience and creativity. And the reason for that is, the amount of time people spend on your site has become a greater and greater factor when people click from Google search.
Google knows if somebody clicks and then they go to your site and then they come back to Google really quickly, that implies to their algorithm that your content was not good enough that someone wanted to stay. So it's really important to focus on the user experience, the creativity of your content, and getting people to actually stay on your website, especially if they come from Google search and you want to keep your rankings.
So, this is a quick framework and a way to analyze data that I came up with. It's been really useful to me, and I wanted to give this tip before I'm done. So, what I like to say is the art of marketing is to create better content. And as I said before, content to me is anything you see, anything you're trying to do to influence people in marketing.
There's actually almost no marketing you can do without creating some amount of content, and there's an art to that. But the science of it is what actually helps you figure out what's working, what's not working, whether it's you imitating a strategy that someone else is using or your own content and looking at the performance of it. That's the science.
I'm going to talk a little bit about the science. One thing you can do, and you could do this right now as long as you have some amount of traffic, is you would go into Google Analytics and go find pages on your site. You can go to the tab called Site Content and Landing Pages and go find pages that have high traffic and a high bounce rate.
That basically means that people are coming, not getting what they want, and they're leaving really fast. And if that's the case, then that's an opportunity for improvement. If you have landing pages, find landing pages, this is from a tool called Unbounce, but you can find landing pages that have high traffic and a low conversion rate compared to your other pages. That's another place where you'll find opportunity.
If you're doing a bunch of email marketing or just looking at a whole bunch of channels, you can go start looking at what areas have high volume or high traffic and we're getting low conversions there. So this is very good when you're trying to measure different marketing channels and figuring out which ones are high traffic but not performing so well.
Because if they're getting a lot of traffic but they're not performing that well, you'll probably likely lose the traffic at some point, especially if it's coming from Google search, or you're going to want to find time to improve it because it's low hanging fruit for you.
So there's a nice little framework I created to help you define this and figure this out. You can do this not just for traffic, but you can do it for many other things. So basically how it looks, starting with the top left quadrant, is if you have high traffic and low conversions, you should consider increasing your conversions by doing AB testing, because you have enough traffic to do AB testing but your conversions are low.
If you're in the bottom left quadrant, you have low conversions and low traffic. You only have one choice, which is you want to get to at least the quadrant right above it so that you have enough traffic to do AB testing.
So those are the types of channels, the types of opportunities where you should figure out if those are places where you can get more traffic. Common channels like that happen to be things like Hacker News or Reddit where you might've done one or two posts that worked really well there, but the traffic isn't converting and there's not that much consistent traffic coming, so then you would basically see if you can get more traffic from those sources by creating more content.
And in the bottom right is another interesting place. That's also a place where you want to get more traffic, and then that place you instantly get more traffic and you're going to be in the sort of do a happy dance place, but here it's like you have low traffic but you have high conversions.
The only answer there is to get more traffic, because you're trying to get in the golden bucket where you can do the happy dance and be stoked because you've got high traffic, high conversions. Those are the kind of channels and opportunities that you want every single channel you have to turn into.
In this case, if you have channels in that category, your goal is to look at it and say, "Okay, I want to make sure we can maintain it. I want to make sure that we don't lose it, and it's more of a monitoring problem, not necessarily a tactical AB testing or get more traffic problem."
So, that's the end of the talk, and I'm very easy to reach. That's my email, and I would love to help you, no matter who you are and what you need.
So I'm going to address your point about, developers aren't good at writing content. Let's start there. They're good at something. They're good at writing code, right? So what if your code was your content.
So in the case of Keen IO, they wrote a lot of open source code and that happened to be their content because GitHub provided them with the distribution. So strategies they would use would be, they're an API for analytics, right? So they'll go, let's say, build an open source add on to Wordpress and then put it on GitHub.
Then everyone that uses Wordpress would have the opportunity to go check it out and create analytics on top of Wordpress for the blog posts and stuff like that.
That's an example of the strategy you can use if you truly think that the people and the team can't write like blogging content, right? Another thing that I would suggest there, and this is very related to the original question, too, about how to do it in the team structure, which I'll talk about, is live coding, or even coding videos of watching people integrate something and all that.
Those are very popular, too, if you start looking on YouTube. So the key there is to go look at the content that's out there in your category using some of the tools I suggested and figuring out if there's any opportunities that aren't just words.
Because words, even developers, they would rather learn how to do a job than go read a bunch of content, so you really need to match that with your audience. So I know I spent a lot of time on blog and things like that, but there's a lot of other channels and a lot of other types of content and places where you can put them.
So that would be my suggestion around don't be afraid to not write a blog post but do other things that work for your audience. One example, and I'll give a big nugget that I haven't seen anybody really do well yet is call it a codinar.
Don't call it a webinar.
And publicly set these up, let people opt in, and let the engineers come and literally watch you integrate your software into their stack or a similar stack to what most customers have, and they'll like it, and it's not a webinar, because you know they don't like webinars, but they like codinars, which hasn't been invented yet, but those are the kind of ideas I'm talking about.
This is where the creativity comes in.
On your question about team, the right answer is usually we find the most success when founders are telling their story and using content to tell their story, especially early on. And the reason for that is, if you're sharing why you started this company, that should be easy for you to write about.
If you don't want to write about it, talk about it, and then transcribe it and then turn it into a story. Stories are one of the most natural ways to write, one of the most natural ways we speak, and so everyone can tell their story. They might not be polished, it might not be brilliant, but there's editors for that. So just write your story.
So if you're ever stuck with, "I don't know who should write this or what we should do," I would just start with, if you haven't already told your story somewhere, whether it's on Medium or your own blog or something like that, start there.
That's a great place. And in fact, back to my repurposing of the content, imagine if you had an early stage founder on video, like a stage like this, on video, I mean, we're in a place where I'm sure they'd let you use their tools here, and basically recorded their story.
Not like it would even get shared or anything, but just the idea of recording it and then turn that into content, whether it's converted to blog, blog posts, or a graphic or whatever it may be. So I think there's a lot of creative ways to do it.
When you're a really small team, it's the founders. It's the early team that's creating the content. And when you're really seeking product validation, it's the product team. Well, who's the product team? The founders. Because what you're trying to do is you're trying to attract your early customers to you, and one of the best ways to do that is by actually sharing your story and talking about why you started this business and what you're doing.
Now, as you start scaling, obviously that doesn't necessarily scale and the founder has other duties and things like that, but still, the most important thing for the company to do is find customers and convert customers, I'm sorry, find people and convert them to paying customers.
Other strategies that do work, and something that I didn't share but I will share now is that you can go find great writers that are out there that are writing about things that are related to your business and go solicit them to either guest blog post on your site, ghostwrite for you with your name on it, or pay them to write in some way, whether it's with their name or whatever, and many, many folks will be willing to do that.
There's a lot of freelance developers out there that are writing great content that you can co-opt and say, "Help us write great content." So those are other kind of ideas. One tip there is, someone needs to be the editor in house and actually edit that content that's coming in if you decide to take a outside content strategy, as I call it.
At Kissmetrics, we've gotten to over a million visitors a month by 99% of our content being written by guest authors. So we're not writing our content. It's guest writers writing our content. We're a analytics tool. We target marketers. And there's a ton of marketing content out there and there's a ton of writers that are willing to write that content.
There's also, once we started building a brand, we stopped having to pay for these guest writers to write on our site because they wanted to reach our audience. And we were happy to do that. It's our blog, and it has a great audience and it's large.
And so we had one editor, and that was the whole team for many years. To get to a million visitors, we had one editor. And that person, his job was literally know our audience extremely well for content and our blog and our customer segment and then just find the best content out there, not us writing it.
We've written probably one or two percent of the content out there on our blog, and that is a contrast to a company like HubSpot that has 60 people on their content team last I checked. And I wouldn't suggest doing it like that. That's investing super heavily in it.
They have a bunch of reasons they do. They create marketing software, and they invented the whole inbound marketing category, but that doesn't mean you have to do it like that, and early on, the early team members should be doing it. That's the most effective strategy, and if you're stuck, storytelling is the key way.
If you take an approach of, again, we're just going to do blogging, it'll be very hard for you to justify ROI for a very long time, unless you're good at the strategy we created where it's not our content. We might pay for it, but eventually we knew we could shift and not have to pay for it because we built a audience. We were very good at that. So I wouldn't expect everyone to be able to do that.
That being said, I think everyone can, you know? So my suggestion there would be, start thinking about it a different way. One, content builds brand, and if you early on, early on you don't have a brand. You're nobody, you're nothing. You're like something new. You're creating something out of nothing.
So just like Codecademy did, if you start telling your story, start really being deliberate about the language you use when you're launching something, then what happens is that turns into your strategy. You keep sharing your story, and people come along the way. And there's a lot of mediums to do that. It doesn't have to be a blog.
Then the ROI is almost, you learn it over time because you've started with a strategy where, "I want to attract customers. The content I write is deliberately designed to do that." And so over time, the ROI is built in, because you started with ROI in mind because you were looking for early customers and doing product validation.
So the classic way that I don't see enough people do is, and I shouldn't call it classic, I guess, then, but is basically write about the problem you're trying to solve, and then at the bottom, even in the middle, collect their email address. Collect the email address of anyone interested in learning more about this and staying up to date on what you're learning about this.
Don't even try to sell them anything, because it's early on. You're trying to learn from them, and if you have any of those, if you get any kind of traffic into those posts, even from Hacker News, there's some percentage of people that put in their email. And then now, you've got an email.
You've got many emails, and that's probably what you're going to use to go get your early customers. So just build an email list and do whatever you can to build that email list, and ROI becomes not even a question.
But if you don't do that early on, I'll give you a really good example of this since I don't think there's too many more questions, but I'm happy to answer them. I was talking to a company this morning, and it's a friend of mine's company and I was doing them a favor and I was even willing to come in and talk to them and things like that, and I already did, and they said, "We want your help with content marketing."
Their at like a five million ARR. They're a SaaS business. They're not developer-focused. They're more like sales and marketing focused. And they raised a bunch of money, and they asked me, the first question I asked them about content marketing when I got on the call is, "what are you expecting from this?"
And they told me in an answer that completely informed the rest of the conversation but they didn't know it yet. They said, "We want leads." I'm like, "Okay, you got to five million ARR, "you haven't built a content strategy yet, and you're telling me you want leads from it, and I'm going to sit here and tell you it's going to take you three to six months to get any ROI even at your scale."
"you're going to have to invest probably 10 to 50 grand a month just to get there." And the reason is, they didn't invest in it early on. It wasn't a key priority for them or a channel, but yet everyone in their space has invested in it, probably early on.
So now what do they do? I convinced them not to go after content marketing. The reason is there's probably other low hanging fruit if they're not willing to invest the time and money.
So I know this sounds weird, but this is one of the strategies where if you try to do it later, it costs you more money, takes you more time, and you probably won't do it. And that doesn't mean it's the best strategy early on or not. I just know that early on, your story is what drives customers to you, and you have a great story.
You're creating something out of nothing and everyone wants in on that. They just do. And you want to learn, though, and you want to pull these people in. So my suggestion if you're early on is start investing in it.
If you want the ROI, just focus on how many emails we collect off of each piece, especially early on, and you'll see it pay off, because you'll learn what gets emails.
What you'll learn is, when you talk about yourself and what you're actually doing and why you're doing it for them, they'll want to be in on it.
Because you're attracting those early adopter-type people that want to try new tools and things like that. The good news is, developers are in a constant try new stuff mode. Whether we want to all admit it or not, how many developers have we talked to that are using React now? Why?
And I'm going to sound really funny right now, but what's wrong with jQuery? It was a joke, there's a lot wrong with it, but that's kind of the mental model, right? And when your audience has that, this strategy works really well.
This doesn't necessarily work with marketing right now, because the marketers don't necessarily have that same mental model, but developers still do.