Welcome to the Authority Hacker podcast, the place to learn incredible actionable marketing tips to dominate your niche. Now your host, Gael Breton.
Gael: Hey guys, welcome to the fourth episode of the Authority Hacker podcast. In this episode, I am very honored to receive a Breanden Beneschott who is the co-founder of toptal.com. Toptal is what they call a top sourcing company where you can hire very qualified developers to do any kind of development you want. Breanden, thank you for being here.
Breanden: Thanks Gael, happy to be here.
Gael: Can you just explain what is Toptal, because I went over it in one sentence, but obviously you must have some more things to say, so please go ahead.
Breanden: Sure, so we are about three and a half years old, founded in November 2010. We are essentially a marketplace for top developers and top companies. A lot of companies including a lot of startups come to us when they need short term or even very long term software engineers pretty much on demand, and we have a very wide service where we are working with them, to line them up with some of the best developers all over the world. There is about 900 of us at Toptal, guys from everywhere, from Princeton, to Facebook, Google, Cern, places like that, we have a very rigorous screening process for every engineer before they are admitted to Toptal, and once they get in, they are eligible to work with Toptal clients. So we are essentially the link between really good developers and companies looking for very good developers all over the world.
Gael: Yeah. So it's basically a very vetted and curated marketplace for people to go and not waste their time they just get someone good right away, right?
Breanden: Exactly, so we have a lot of CTOs from very well known companies to get like when Facebook or something post a job they will get thousands of resumes but that takes a tremendous amount of time to actually go through every, and so the idea with Toptal is our acceptance rate is around 3%, we have a very similar screening process to Facebook and Google plus it's a little bit more strict because most of the work that we do is remote and so we really have to look for a lot of discipline. But at the end of the day, we are looking for extremely smart people with tremendous work, ethic and frankly a lot of entrepreneurial drive, and then once they have proven that to us and have done everything from really difficult code challenges to programming projects etc, then they are within what we call the "Toptal network" and our job is to make sure that they are working with great companies and challenging projects.
Gael: Yeah, I guess that makes sense. When you are hiring developers it's always very complicated, but I think a lot of people who have been listening to that and I think their brain probably stopped when you say we have 900 people in 3 years. Which is crazy, crazy growth. And one part of your growth tactic is your blog- you guys actually have a very successful engineering blog where you guys post regularly like cutting edge content and so on. It seems like it's going well for you guys, so can you just describe a little bit what you guys do with that blog, because I guess that is what appeals most to the Authority Hacker crowd.
Breanden: Sure, so about a little over a year ago, we sat down and said that we wanted to start blogging for a bunch of reasons, to sort of prove our technical skills and more than just saying that we are top developers, or very good engineers, and have a lot of examples of this, out on the internet. And then also in many ways give back to the various communities and put really needed content out there, like interesting things, amazing tutorials, really innovative ways of doing stuff. And we have a lot of engineers who are in Toptal all over the world. One of the benefits of working here is that you can work anywhere in the world for the most part as long as you have a strong internet connection, and while that's amazing, it means travel and working from beaches and all of that, a lot of times they are pretty isolated, and so you could be working within amazing team at YouTube or something like that and solving very difficult and problems that tons of engineers all over the internet are facing every day. And, we wanted to make a platform where you could share those solutions easily and have a wide spread reach. Then you would also get recognition for that and so it goes on your profiles, we like to call it being published on the Toptal engineering blog rather than just us having a little team here that is pumping out articles every single day, or something like that.
Gael: Yeah, that makes total sense, especially if you guys are saying you want to be the best, it's good to work the talk I guess.
Breanden: Yes, that was one of the main ideas behind it.
Gael: Cool, so that's kind of the vision, but I'm very interested in how it got started, because obviously that's a very ambitious vision, but you can't have all of that on day 1. So how did you go about kind of like NVP-ing the whole process, and what were the first steps you guys took to achieve that vision?
Gael: Is it possible for you to share some traffic numbers or shares numbers, or whatever you are happy to share in terms of what kind of traction you guys get from some of these posts?
Breanden: Sure. So the average post is getting tens of thousands of uniques at this point, and a lot of that has come from the list that we have essentially built up and so we have subscriptions, subscribe buttons, and so that alone has over 25 000 subscribers, Twitter has over 20 000 people on it, Facebook has 10 000, and then we get hundreds and hundreds of thousands of unique visits each month from organic traffic and so some of these pieces have done so well but they are outranking Wikipedia for basic technical terms, they get a lot of traffic.
Gael: Wow, do you have a term that you guys are outranking Wikipedia for?
Breanden: It depends on where you are but I know some of these like NoJS or AngularJs, some of these ones do very well.
Gael: These are very trendy technologies at the moment as well, so that's good for you guys I guess.
Breanden: Yeah, there is a few pieces that have just been totally outstanding in terms of organic traffic for us.
Gael: Cool, there is one thing I noticed when you tried to describe a little bit of some of your process for the first posts- you say that everyone does what they are good at, and you said that the person that has a technical knowledge wasn't even writing, they were just doing a list of bullet points and then someone else was writing the article. I think that's very interesting and very few people do that, so can you describe a little bit your editorial process and what goes on behind the scenes in drafting an article.
Breanden: Sure, so we usually sit down with an engineer here and brainstorm topics and try to come up with something either very cutting edge or very interesting. Like maybe a very elegant solution to a very common problem, maybe it's not so cutting edge but it's just like brilliant in its simplicity, or on the other hand because that's a very hard thing to do, the cutting edge or something brilliant, something easier but still extremely valuable, is coming up with like very well researched very clear article about a technology and so how to build your first AngularJS app, start to finish with source code and everything like that, or questions about like a deep introduction to hiring IOS developers, or something like that. So once we have actually sat down with the people here and decide what we are going to do the engineer or the writer which is an engineer in all of our cases goes and will come up with the first draft and sometimes this draft is a bunch of bullets as I said before.
More often than not it's what you would expect in the first draft, it's several paragraphs or several pages and then we are going back and forth as many times as needed with the team of editors here who all have deeply technical backgrounds themselves, and then we are going back and forth until we feel like this is either extremely cutting edge and fascinating or interesting, or this is something that the internet is asking for in its own way, so we are looking at keyword research like AngularJs tutorials and we are not seeing anything that is really good out there already, looking at some of the competition for these things and saying, "Is this something that is going to be easy to outrank?" Or, this is totally saturated and not really worth it, we are just going be adding something that's already here.
Gael: Ok, and like so you guys already focus on SEO when you make a decision? You guys do a bunch of keyword research and stuff?
Breanden: We've started to, in the early days it was, we were trying to create something from nothing and so we are working with engineers and they are really coming to us offering what they think is going to be, something that they really want to share with the world, like an amazing project they've been working on or something like that and that has worked and then as we've gotten better and better at this and seen what has like really worked you know with ten minutes of keyword research you can get a good idea for how well a piece is going to do assuming that it is A+.
Gael: So what do you do if an engineer comes to you with a really good idea but there is like no volume- do you still go ahead, how do you go about it?
Gael: Sure, makes sense. How much work goes into a single blog post, because it seems like a very lengthy and tedious process to obviously output something amazing, I am very curious how much time you spend.
Breanden: There are very few posts that take less than 20 hours of time from multiple A players and the editors on this- we have been very lucky to work with some amazing people, guys from Facebook, and Amazon, and US Government, and Princeton, places like that and they put in 20 hours of extremely heads down focused time at least on virtually every one of these posts and so it doesn't scale, you can't just do say you are going to publish five of these every single day, when you are working with a small team or something like that, and we really have to make sure that everything you put out there is going to be amazing and impactful because we are putting so many resources into it.
Gael: Ok. And knowing you, I know you love scaling stuff. So you named your own problem. what are you going to do about it?
Breanden: Well, with SEO at least in my experience here, it sort of builds upon itself. And so, you have articles that do really well and they don't stop doing really well for you, a lot of times if you AB test titles and things like that, they'll do better and better for you over time and then play around with internal linking and all sorts of other things and you can get things that just build upon themselves, and so you'll get published 5 articles a year ago that now give you a 100 000 uniques every month from SEO. And so scaling it, we are not trying to necessarily be a content machine of thousands of posts every single week, but where we are trying to scale it to a certain extent.
So initially, our goal was to do about one post per week, and then over time, we've added resources and improved our own processes and things so we can average several posts per week and we are trying right now to bring it up to about 5 per week. And then, that's probably pretty good for a while, at least given out current launch, we don't want to be sending our engineering lot of subscribers multiple emails every single day, I don't think. But, we'll decide that when we get there, maybe that would work.
Gael: Yeah, I think a lesson to learn here is that a lot of people when they start corporate blogs, they just put a frequency in their head that they kind of force themselves to follow, even though there is no process, nothing, and you end up with not super quality content, you know. And, I think if you are listening to this podcast and you want to start a corporate blog and look at the Toptal model, you know, I guess you guys started really slow, I think between the first and the second blog post there was like a month or something-
Breanden: Definitely, I don't remember exactly, but there was a few weeks or something like that. And if you really want to approach these things initially saying, ok there is a quite right here or a goal at least that we can't miss, and we are going to be posting 5 of these a week, but when you really get down into it, for us, it's been really important to just say that what's way more important to our success or critical to our success is compromising on the deadlines there, rather than the actual post themselves. And so, that's just part of establishing the vision in the very beginning and then actually adhering to it.
Gael: That makes total sense. There is one thing that also drew my attention in one of your previous responses- you are testing headlines and internal linking and so on. I am very interested in what you guys are doing and how you do it.
Breanden: Sure. So, a lot of our posts, if you don't know the basics of SEO- essentially, when you are typing something on Google you get ten results from the first page, and Google is ranking those based on a whole lot of factors including how many people are linking to each of those pages and then, when something gets on the front page how well it actually does there, are people clicking it on the headline and immediately closing the page or going back and clicking on another page or something like that, are they never clicking on your page at all... And so you really have to earn your keep when you are on the first page of results and so for us it's always important to make sure that we are writing good headlines that are accurate but also two headlines that say exactly the same thing but they are just essentially a different ordering of words, or one has a different keyword in it; one might do really well and get people's attention and one might totally fail. And so, constantly testing those and kind of putting your own biases aside is something that I've seen really big wins in SEO.
Gael: Do you have an example of that on a given blog post or something that you guys did?
Breanden: Sure. We have an AngularJs post an introduction in how to build your first app with source code and everything that I mentioned earlier, and it was a really popular post when I first went out, and so that's I think trended on Hacker News and does really well on Redit and gets resyndicated and stuff, and so gets all the links that you would hope for in an A+ piece like that and then when it came to SEO we started looking at, we built the team, more and more people with a lot of more experience than me in SEO are now on the team and making contributions and somebody looked at doing a bunch of research and saying people are actually searching for not for how to build your fist AngularJs app which was the some form of that was the previous title, they are looking for AngularJs tutorials. And there was a massive amount of people using those specific keywords and so you can do something like that and putting tutorial into the title and all of a sudden, within a week you are going ten times as many people on this every day.
Gael: That makes sense. So basically you are just creating something that transfers then optimize it later I guess. So I guess you have that phase where the title is more important for it to be viral than to be search optimized and then later you guys go back and re-optimize articles.
Breanden: Yeah, definitely.
Gael: Ok, that makes sense.
Breanden: I've met a few other people that are not necessarily in this industry but people with really successful blogs and you read on Moz and all these things, I mean, this isn't any secret, most of the successful people are doing exactly that, they are realizing that no matter how good you think something is and it doesn't necessarily mean anything when it comes to SEO and you can make very small changes in this roughly scientifically tested and I get your way to something that is way better than it was doing initially.
Gael: Yeah, I totally agree, actually one thing I am testing on some of my sites right now, that I am playing a lot with, is social media meta tag, so you can actually put your title tag for your SEO that's well rounded SEO, keyword optimized and so on, so your piece ranks really well. But you can also add Facebook tags and Twitter tags and so on, so the title, when it's shared or liked by anyone is actually different and you can have something that's a lot more click bate and like more socially viral and all these things, and you can have both running parallel these days using these social media meta tags.
And that changes your sharing a lot because you basically get your SEO traffic and people share it, it's a different title, for example I have one that's the Amazon partner program guide where I just explain how we use the Amazon program, and it's pretty boring title but it's well rounded for SEO but when people share it on Facebook or when it's tweeted or something like that, it's says how 21:41 make hundreds of dollars on autopilot every month from the Amazon affiliate program. And that gets like ten times more shares and likes and clicks, you know. So I guess you guys are doing the same thing, and you just testing again and again. Do you guys use the webmasters tool information clicks and so on to figure out what works better?
Breanden: Yeah, definitely.
Gael: Ok, makes sense. In terms of content promotion, are you guys still promoting your content other than your internal channel, so other your email list and your Facebook page and whatever way people follow you- do you guys do external promotion in any way?
Breanden: Not really. We do a little bit in terms of promoting them on Facebook and LinkedIn just because there are some interesting targeting options there, and it does bring some traffic and hopefully some conversions that we care about each time as well. We do do a little bit of research of these as well where a small group of people will go out and look for other people that are blogging on very similar content, and then assuming that there is some deep connection or relevance what we may share with them.
Gael: Ok, fair enough. One thing I found that works well as well- if you find other very good pieces of content on the topic, put a quote of these people in the blog post, and then email them, and be like, "Hey, we quoted your article because it was so good."
Breanden: Yeah, I've heard that from several different people and actually I've noticed that now we have a blog with the bunch of traction on it, people do that to us a lot so we get emails every week of people saying, "Oh we referenced you here, etc" so I think it's definitely an effective link building technique, so many people are using it.
Gael: Yeah, I mean even if it sometimes times just a social share from these people, they have a big followership that can make a difference, and if it makes your blog post better, why not, you know. But yeah ok, I wanted to ask now about conversion. So we kind of understand how you guys create content and publish it and get traffic to it. But, what you guys try to do with your blog readers, are you trying to sell them right away, are you trying to convert them in any other way, like what's your goal when someone reads your blog?
Breanden: So there is a few different things going on there that are calls to action and we have a lot of different types of people on our blog and so that's something we do have to take into account, we can't just go for a hard sell every single time, because most of the people on there are not potential clients, they are actually engineers and so on most of our- on all of our posts we have subscribe options, we have sign up as a client so hire a developer option or work as a developer as well, essentially any of those counting as a conversion has is very valuable to us. And so, for us, a question that we have to ask ourselves a lot is what type of piece is this and where is it on out site and do we have other content outside of the engineering blog like hiring guides and interview questions and things like that, we don't have to say who is the most likely type of person on here and how hard of a sell makes sense, is it something where they are ready to buy now and even the buttons should say like "hire developer right now" or is this something I've subscribed to more articles like this or view other interview questions for similar technologies.
So, and a lot of this just starts with our hunch about what we think people are like on this page and then it's just testing all the time of different types of things. I think the hard versus soft sell thing is something we have to test a lot but everything on Toptal more or less points to the bottom like which is getting somebody or somebody that that person knows to come in and hire developers or getting somebody or somebody that that person knows to come in and work as an engineer and become part of the network.
Gael: Sure. And from all these tests and so on of hard sell versus soft sell, and so on, is there any valuable learning that you can share?
Breanden: I think one of the most valuable things that I've learned is that my assumptions about what is going to work are wrong almost half the time. I am very bad at just like, almost everybody here, those are very smart people working at these things at Toptal and it's very difficult for us to be able to accurately predict outcome of an AB test and so what I have learned is you don't let your own ideas of what you think is going to work, cloud your judgment before you test things, let data do the work and make sure your data is clean and that you understand it so that you are not making mistakes.
Gael: Yeah, doing online marketing consulting I would say this is probably the single most frequent point of argument with the client, when people just go with their assumptions. And it's just like, when you've been in business for 20 years and you know your business, you know your market and so on, but it just limits your willingness to test sometimes and you are probably missing out on a lot of stuff if you don't so, what is really good with you guys is you just want to test and see the data. I guess that's the benefit of working with engineers, right?
Gael: And not do too much as well. Because when you run several conversion tactics at once, if there is like a pop up and 3 opt in forms and 2 banners calling with you it just becomes crazy, right.
Breanden: Yeah, we're pretty hungry for growth at all times, I mean when your company is taking off or something you are working on is taking off it's pretty exhilarating and so we are constantly chasing that. And so we've kind of developed an approach where it's not just looking at the Chi-square test results or something like that for statistical significance on an AB test, a lot of times we'll look for something and do a mix of sort of cowboy testing and academic type AB testing where if we see something that is hundreds of percent gain like right out of the get go we might go with that and then iterate more quickly just to minimize the number of variance testing at any time and there is some other machine learning techniques. Again, it helps having a lot of engineers because you can get pretty sophisticated with this stuff.
Gael: Yeah. I can imagine. Now, I just want to go back a little bit, and I want to understand how important is your blog to your business these days? Like, I guess it's hard maybe to track how much revenue generates and stuff, but like in terms of lead generated or something like that and like assisted conversions, like if you can tell us in percentage or something we don't need any number or something, just like it's for example 10% of our leads or something...
Breanden: Sure. So, a year ago it was approximately zero percent of all inbound leads and contributions to our business because it didn't really exist. And then now, it's definitely on track to dwarf every other channel that we have in terms of inbound leads when we have very large budgets on places like LinkedIn, and AdWords etc. And so, this is I'm not sure what percentage it is right now but it's quickly getting- it's definitely going to be the number one thing by the end of the year.
Gael: Is it going to be bigger than your PPC campaigns?
Gael: Wow. So that's pretty big actually. And I'd like to know like how much do you spend on your blog compared to your PPC?
Breanden: We spend a lot more on PPC. Even with the team of very senior people and so much time on every post, we spend a lot more on AdWords.
Gael: Yeah, that makes sense, that's what I thought but I guess you basically lost a lot of money at first on your blog, you have probably lost tens of thousands of dollars when it started.
Breanden: Yeah, during some point we were certainly running a negatives until it caught up in terms of SEO and started paying for itself, we had the benefit of selling very high ticket items and so a big client came through and just paid for it, everything plus a hell of a lot more and that's definitely happened.
Gael: Cool. That makes a lot of sense. Now, let's imagine for a second that you are not working for Toptal anymore and you are starting a new startup company and you want to use all the learnings you've had from starting the Toptal blog, and you are creating a blog for that startup company. How would you go about it?
Breanden: I think the first step like I mentioned before, is just setting your vision and setting it so that you don't have to compromise and so if your vision is it's only A+ content whether it's technical content, or just interesting content or whatever, that's the most important thing. I think you really don't want to be in a position where you are publishing fluff and just trying to make it a numbers game because I don't think that that's going to work very well. And we've done really well over a lot of our competitors because they were publishing fluffy stuff and so we chose to do a completely different approach and I think Google has rewarded us for that, several different updates and they are ranking stuff like penguin or panda or whatever it is. Those things come out and then we benefit. Because Google pushes the best stuff to the top and we are spending hundreds of thousands of dollars creating the best stuff and that's- at least so far has been really paying off for us and so I would follow that approach no matter what I am doing.
The next thing I would do is get a good AB testing framework in place so something like optimize your Google Experiments and Mailchimp or something like that, I would probably use Mailchimp since it's nice and easy and there is some basic AB testing stuff in there as well and that is how I would actually be building this list and then making sure the subscribes and all the stuff are there. I would look at Google webmaster tool, Google analytics, I would come up with an initial launch plan so maybe this is a simple as I am going to put it on Facebook and spend a $100 in a very targeted way, or maybe I have a list or maybe my company already has a list of customers and things like that, and figure out where your first few viewers are coming from and then make it very easy for them to bring other viewers there, so very obvious shares.
And then, from there, it's kind of lean startup methodology of build measure learn, publish measure learn in this case. And that process never really stops and especially because you can change articles when they have been out there, an article can be a total flop and I've done this on Hacker news I know a lot of people have done this where they put something on there and it didn't go anywhere. And then they change the title and then they post it on there again like a week later, and it shots straight to the top and it was the trending for a day and that got them 15 000 very targeted viewers for what they were looking at. And so, I think that the situation is never final, on these, it's always possible to get another last word in there when it comes to content and that's really important here.
Gael: Yeah, that sounds really interesting actually. So you go ahead and you take what didn't work out. Obviously try to fix the article if it wasn't very good but you guys only output really good stuff. And then, change the title and go ahead and go on all these social sites again, right?
Gael: Cool. I don't do that, I should try it actually. That sounds like an easy way to get some traffic if you are maybe struggling to create content at the pace you want yet, you know.
Gael: That sounds like a good strategy to revive some traffic.
Breanden: I think that is a really good point, I think when you have really high standards for your content it's definitely more difficult to scale it and turn it out as quickly as you might like and so being able to maximize the value coming from every one of those is really important and that really comes down to a patient and smart testing and learning cycle.
Gael: Cool, yeah, that makes a lot of sense. I guess that's about it for the interview, but I just wanted to ask you one last question which is- did I forget anything in your process that's like super important and we didn't mention at all?
Breanden: I think when it comes to SEO stuff, doing link building and doing a lot of the sort of clickbait titles and stuff is really- you really need to be careful with that. So it's great that one title might get a much higher click through rate then another, but if it's clickbait type stuff a lot of times buzz feed and things like that, and users aren't- they click on it because they were baited and then they are quickly leaving and stuff, I think you are just wasting your time on spinning your wheels and you are much better to have approach of- we want something great first and then optimize right after rather than just something that's purely optimized for some sort of metric like click through rate. And then, I see a lot of link building because we have been spammed on Toptal and stuff, where people are just dropping in some BS comment and then a link to whatever it is, and I think Google is getting smart enough that that stuff doesn't work for very long, even if it does work for a little bit of time and you are not doing yourselves any favors by trying to go down that route. And so, being as white hat and transparent about all that stuff as possible, is I think really important.
Gael: Yeah. I think you need to just focus on delivering value before you focus on getting traffic anyway. It's just like if you content is average, don't waste your time promoting it, get back to producing again.
Breanden: I think that's a good point.
Gael: And then, you can play with all these things. But the problem with most people is that they want to play with these shiny tools, right, they want to do AB testing, they want to do clickbaiting, they want to do all these things, so they kind of rush into creating the content, and then they just want to play with that they and they get average results or low results so just people just click and that's a BS site. And, that's why they don't really work out, most of the time, you know.
Breanden: Yeah. I don't have as much experience in this as you do but that was my sense when we started going into this and we've been fortunate that the route that is very different from that has really worked for is.
Gael: Yeah, it seems like it. All right, so I guess that's it for the interview, thank you very much for sharing all of that. I think I'll be talking to you afterwards and we'll put something together that people can download to learn a bit more about Toptal and the blog and all that stuff. If you are into engineering, if you are developing stuff and so on I really suggest you get into the Toptal blog, it's toptal.com/blog so you can see everything we talked about and the level of quality these guys are putting into their blog post and so and if you need any development you can also talk to them. So thank you for being here, Breanden.
Breanden: Awesome, thanks so much Gael.
Thank you for listening to the Authority Hacker podcast. If you liked this episode don't forget to rate us on iTunes and we'll give you a shout out in the next episode. If you want more, 100% free niche marketing tutorial and hacks, head over to authorityhacker.com.