Episode #011

How Ryan Bednar Went from SEO Consulting to Building an SEO Product That Makes $80,000/mo

Turning your skillset into a product that solves customers' problems is one of the best ways to start a business. Learn how Ryan Bedndar came up with an idea, iterated on his product, and found his first customers.

Review

Enjoy the episode? Review us on iTunes! 😇

Transcript

Courtland Allen 0h 0m 8s

What's up everybody? This is Courtland from IndieHackers.com. And today I'm gonna have my friend Ryan Bednar on the show. Ryan and I did Y Combinator together way back in January of 2011. And today he's working on a startup called RankScience, which he's very quickly ramped up to $80,000 a month in revenue, by helping businesses with their SEO. For those who don't know, SEO stands for search engine optimization, and it's all about getting your website to rank highly in Google so that you can take advantage of all that traffic. Specifically with RankScience, he's helping businesses A/B test their search engine optimization. And he brings a ton of experience as an SEO consultant himself. In is podcast, you'll learn a lot about the fundamentals of SEO, about how to use your domain knowledge like Ryan did to grow a successful business and about how to find your first customers and learn from their feedback. So, here we go. Ryan Bednar, how's it going?

Ryan Bednar 0h 0m 57s

It's going well. Thanks so much for having me on here.

Courtland Allen 0h 1m 1s

Yeah, thanks for joining me. So to jump right in, let's start with the simple questions. What is RankScience and how does it work?

Ryan Bednar 0h 1m 8s

RankScience automates SEO for businesses. The way that it works is we built a CDN, companies route their web traffic through our CDN, and we automatically grow their organic search traffic.

Courtland Allen 0h 1m 25s

And that's an idea that I've never heard anyone else doing. How did you come up with that idea?

Ryan Bednar 0h 1m 29s

Yeah, it's a good question. It was sort of a longer evolution. So I'm a software engineer, who became a SEO expert sort of by accident. And so at previous startups I helped start, or was an early employee at, SEO was this tremendous revenue generator. And a lot of times it falls on engineering to manage and run SEO, and kind of by default, if no one else is doing SEO. And so I sort of backed into it. And after my last startup shut down, I became an SEO consultant for a lot of startups, a lot of YC startups. Folks through my network. And then even some larger companies, like some Fortune 500 companies. And my biggest advantage as an SEO consultant, was that I was a programmer. And so companies could add me to GitHub, and I could actually execute all of the HTML changes and SEO tweaks that I was trying to get them to do. And other SEO consultants just typically send over a list of recommendations, a PDF, like here's a bunch of changes, please have your engineering team do this. And because of that, iteration time with SEO can be really long. And so what I ended up building for some companies in-house, as an SEO consultant, was A/B testing software. It was just insane to me that SEO is so important to all these companies' bottom line, but no one could really measure their changes effectively. Like there's a lot of best practices, there's a lot of so called waving dead chickens around. Things that are just like, hey everyone makes these sort of SEO changes, you have to do this. I don't know why, but you have to follow these steps. And so I was really interested in doing data driven SEO, and started building in-house A/B testing tools. So companies could measure things like title tags, on-page changes, to see how that affected click through rates, see how that affected rankings, and from there, RankScience is sort of an attempt to productize what I was doing as a consultant. So the first version we built at RankScience about a year ago was an API. And the sort of the hope was that, companies would integrate our API into their front-end, and they could test front-end HTML changes. What we found was that companies didn't want to do the work. That it was too much work for them to integrate an API and that's how we kind of evolved into, hey if we build the CDN, we can actually do the HTML modification for companies, and that's how that idea evolved.

Courtland Allen 0h 4m 6s

And so, is your revenue public today? Do you guys share numbers?

Ryan Bednar 0h 4m 10s

We don't share numbers, but this did kind of make the TechCrunch announcement from our YC launch. And so we are above $80,000 in monthly recurring revenue right now.

Courtland Allen 0h 4m 24s

That's awesome! Congrats!

Ryan Bednar 0h 4m 26s

Yeah thanks! Yeah so we just started the company About a year ago, but really the last four or five, six months, things have kind of really started to gel. But obviously we still have a ton to figure out, and a ton of work to do.

Courtland Allen 0h 4m 42s

How big is your company? How many employees you guys have?

Ryan Bednar 0h 4m 45s

There are five of us right now. My co-founder Dillon Forrest, is our CTO. We have a lead engineer. We have a director of SCS, someone who helps work on customer happiness and strategy. We have a data scientist and then we have an account manager.

Courtland Allen 0h 5m 4s

That's a lot of growth in one year. Cause I see it just started with you and your co-founder, and you picked up everybody else along the way, right?

Ryan Bednar 0h 5m 12s

Yeah that's right. We just started hiring. It was just the two of us for six months, and then we started hiring after that.

Courtland Allen 0h 5m 18s

So you mentioned that you did YC, and then you guys were in TechCrunch for YC Demo Day recently. We also did YC together back in 2011, and so between the two of us, we've seen hundreds of companies working on various things. What are your thoughts on picking ideas to begin with? Because for RankScience, it came out of the prototypes that you were building for clients as a consultant. Do you think that's a good way for people to come up with ideas in general? By solving their own problems or by doing things on the job?

Ryan Bednar 0h 5m 48s

I think so. I mean for me, and I forget who I was talking aboutut this with, like for me, RankScience, was like the perfect thing to work on. Because I'm a software engineer who finally and sort of reluctantly embraced being an SEO. And embraced SEO being my lever for helping companies grow. And so yeah, it definitely came out of a personal interest and a personal pain point. And also something that I knew companies would value a lot and could have a big impact. And so I definitely encourage folks to work on things that are closely aligned with their experiences and I think it can be hard if you're doing like a consumer business sometimes, but for RankScience, that's sort of just how it evolved.

Courtland Allen 0h 6m 32s

Six months ago you guys started hiring. I assume before then you started taking off. How did you find your first paying customers? Were these people that had been clients of yours earlier on?

Ryan Bednar 0h 6m 42s

The first people that we convinced, so this was kind of tricky, but so we built a CDN, right? And it's hard to convince an established business that hey, you built this new technology, we have no customers, please be the first customer to route all your website traffic through our CDN. So that was a challenge, but yeah luckily through as you said before, we did YC together back in 2011, and so I've been part of the YC network for a while now. I did have a number of people who trusted me, who I'd helped grow their sites before. And they were willing to give RankScience a try, and kind of take that leap of faith. So network definitely helped a lot. Our first couple of customers were from our network and then from there, other people started hearing aboutut what we're doing, either through word of mouth, just the idea or concept of A/B testing for SEOs is sort of a newish thing from us companies. And so that did kind of garner us some referrals. But then from there, my co-founder Dillon published a case study for this company called Coderwall, and we posted that on Hacker News, and it was like how RankScience grew Coderwall's traffic by 57% with a single SEO A/B test. And that got a lot of attention, and we started getting customers from there, through Hacker News.

Courtland Allen 0h 8m 2s

Did you charge your first customers money or did you kind of let them prototype it for free?

Ryan Bednar 0h 8m 7s

The very first people who plugged in, we let them do it for free, because we just needed to. Again, one of the challenges of CDN is like you can build the technology as like a proof of concept, but you don't really know how it works until you start piping traffic through it. And so, we needed folks to kind of volunteer just to see, just to ensure that this thing worked. So I think that was one way to validate what we were doing. And it actually made a lot of sense because we were making a big assumption that companies would plug into our CDN and route their traffic through us. And there's two steps there. So one is like, we need to get them to be willing to plug in. We need their engineering team to okay it. And then two, would they actually pay for it? And so we kind of checked off that first box like yes. If we promise that we can grow your search traffic through continuous A/B testing, they will plug in. And then two, that they would eventually pay for it.

Courtland Allen 0h 9m 7s

Earlier you mentioned that you started off with an API of sorts, and it wasn't that popular. And then after that, you moved to the CDN. When did that happen? And why didn't people like the API?

Ryan Bednar 0h 9m 19s

I tried to do a lot of customer validation before I actually built the API. And so I went to friends of mine who ran companies where SEO was important, or say 30, 40, 50% of traffic. And I said, wouldn't you love to do A/B testing for SEO? I know how to do it. I've done this for other companies. If I build an API, would you use it? And they all said sure, that sounds great. But then when I actually built the API and went back to them, nobody actually wanted to use it. It was just tricky, because it would require in-house engineers to integrate this API into their logic and into their front-end code. And that turned out to be too much of an ask. But it was a great experience for us. And also like a learning for me, right? So people will tell you that they want something, but then when you actually deliver it to them, getting them to actually use it is the real test.

Courtland Allen 0h 10m 13s

At what point did you decide to pivot and try something else? Were they giving you feedback like, hey this is too hard for us to use? Or were you just analyzing your API usage metrics and you decided this is not getting nearly enough use so we should do something else.

Ryan Bednar 0h 10m 26s

It was both. So some of them are just like, okay cool we'll get to this, but then they never did it, right? And I got the feeling that it was just a really low priority. And then the folks that I did get to actually use the API, they would do like one experiment, like ever. That was sort of another thing that we learned out of this, is that coming up with A/B testing ideas for SEO is tricky unless you have a lot of SEO expertise. And actually, unless you've seen what good SEO A/B tests look like, they're not completely obvious. And so A/B testing for CRO, or conversion rate optimization, is a bit easier conceptually to understand because everybody can look at a button on a page and say, why don't we test making that button green, or making that button red, or why don't we make the sign up form bigger? Or let's really highlight the call to action. So those sorts of design changes are easy for people to generate, but good A/B testing experiments are a bit trickier, because you're talking about small tweaks to title tags, or on-page HTML changes. And that wasn't completely obvious too. So that was sort of the second thing that we learned from building this API, was that generating good SEO A/B testing ideas was hard. And that's another reason why we thought the CDN was awesome, because we could execute the SEO A/B tests, and also we could come up with the SEO A/B tests on our own. And we could just fully automate A/B testing for SEO.

Courtland Allen 0h 12m 4s

I'm curious how you came up with the idea for a CDN because after your API failed, it probably wasn't immediately obvious that there's a way to make this work. If only you'd created a CDN. I also would like to know more about how you interact and talk with your customers. Because I've heard from a variety of founders that they get a lot better feedback talking to people in person or over the phone than they do over email or over chat. And it seems like you had a lot of rapport with your customers and you learned a lot about their experiences, so I'm curious what method you used to learn and iterate on your idea. And also, I'd love to know more about how it actually works. Because somebody who doesn't know very much about SEO at all, it seems very difficult to even think about how do you A/B test on Google. Do you run one test before another one? Do you test different pages that are similar?

Ryan Bednar 0h 12m 49s

The way A/B testing for SEO works, is we run tests across groups of pages. And sort of the easiest way to think about that, we have a case study on our site that we did for Coderwall. But the easiest way to think about that is to take an eCommerce site for example. A typical eCommerce site, they all follow sort of like the same site structure. There's category pages, there's sub-category pages, and there's product pages. And each type of those pages are on their own template, and that template is pretty much the same, but there's different content. And so for an eCommerce site, if we can increase the rankings of their products pages by a couple percent, That could have a really big impact on revenue. And so what we'll do is take, for example, a thousand product pages that are on a similar template, split them in half, 500 would be in this sort of control group, and 500 would be into a variant group. And we would make some small HTML change to the variant group. And to start, one of the biggest tools in our tool belt is increasing clicks. And increasing click-through rate from the Google search engine result pages. If we can basically help companies make more persuasive titles that humans wanna click on, we can actually extract more clicks out of their existing rankings. An example of that might be to add some call to action to the title. So if we're talking about, say our product pages are like video games, and that the page that we're tweaking is like Zelda for Nintendo Switch, and the standard title is just Zelda for Nintendo Switch-brand name. So we'll say it's like Best Buy. We would run lots and lots of experiments on that title to see if we could get people to click on it more by adding some sort of call to action to it. So maybe that title tag is then Zelda for Nintendo Switch, in parentheses buy online. Or Zelda for Nintendo Switch, read reviews and buy. Those are just two really simple examples, but we'll run lots and lots of tests across groups of pages to see if we can increase clicks. And we basically monitor how Google responds over time. So an average SEO experiment takes about 12 to 21 days. Google has to index these changes, react to them, and then we actually look at the data. And so we basically sum up those two groups in aggregate, and kind of see how things fall out. To see if there is a statistically significant result.

Courtland Allen 0h 15m 30s

So a lot of what you're doing is kind of manually looking at their websites and deciding what options should you choose between, in terms of the different tests you're gonna run. Is part of your roadmap making that something that users themselves are gonna have to do? Or are you going to build out a giant team of people to make these changes?

Ryan Bednar 0h 15m 50s

So this is where it gets kind of fun. And so the example I gave you was pretty simple and required some human input. But on our end right now, we are pretty high touch with customers in sort of the first few weeks of them onboarding. And so we'll work closely with them on coming up with a game plan for SEO. And what keywords they really care about. What they wanna rank for, what the goals are. From there, we can just kind of load in SEO A/B testing experiments and our software can kind of do the rest. And then the other cool thing is that we have human inputs for SEO A/B testing experiments for our customers, but we also have software inputs. So if you're an eCommerce site and you come on to RankScience, we have all this data from other eCommerce sites, and so we know what good SEO experiments look like, and what most eCommerce sites should be running, based on our previous wins. And so for customers, it is completely automated. And we are actually automating some of the SEO experimentation process already. But I think over time, we'll move further and further into automation, because who's better at generating small HTML tweaks and changes and monitoring results. Is it humans or is it software? We're betting that it's software in the long run. We have a ways to go to fully get there, but right now it kind of works both ways.

Courtland Allen 0h 17m 25s

So let's go back to how you've grown RankScience into the company that it is today, in less than a year. You mentioned that your customers come from word of mouth and that they also come from your Hacker News posts. How did you get the idea to create that Hacker News post?

Ryan Bednar 0h 17m 42s

So I knew that there was this post in the SEO community like two years back, that was really popular. And the Pinterest growth team put it together. And it was called demystifying SEO with experiments. And that was, as far as I know, was the first published post on SEO A/B testing. And it got really popular and it's still a really popular post. And it's worth checking out if you're interested. And so I was interested in sort of the stir that that caused. And I was excited that you could do A/B testing for SEO. And the more folks I talked to, the more I found out that a lot of Silicon Valley companies, that SEO is really important to them, have these internal A/B testing tools. And so that's Yelp, Trip Advisor, Airbnb, but they all kind of kept it really close to their chests. And because it was a competitive advantage that they didn't wanna talk about it. But this is something that could benefit smaller companies as well mid-sized companies. And also just like any company that doesn't have a whole engineering plus data science team to devote to SEO A/B testing. And so that post was really sort of inspirational for us in starting RankScience. But also kind of gave us a roadmap where we thought, hey if we can just start publishing interesting SEO A/B testing experiments and changes, and that people would be interested in it. People would wanna share it. And the Coderwall one that we posted on Hacker News is particularly exciting because if you look at the change that we made for them, it was like we added 12 characters to the title tag and click-through rates went through the roof. And it really grew their entire site search traffic, with just like a really simple test. And so we wanna continue publishing content that's sort of unusual HTML changes that lead to huge SEO wins.

Courtland Allen 0h 19m 36s

The fact that you knew about this other post, did you say it was Pinterest who had the post?

Ryan Bednar 0h 19m 39s

Pinterest, yeah.

Courtland Allen 0h 19m 42s

The fact that you knew about Pinterest posts talking about experimenting with SEO and how successful it was really is something that most people probably wouldn't known about or noticed if they weren't SEO consultants themselves. Do you think there are other big advantages that you've had being an SEO consultant for so long? That has helped you make RankScience successful as it has been that you wouldn't have been able to do otherwise?

Ryan Bednar 0h 20m 6s

Yeah, I think the biggest thing is just coming from a software engineering background, Dillon and I both went to MozCon in September and we're a fan of Moz. Moz is this amazing community of SEOs. And they have really good reporting software, everyone uses it. They're great folks and they put on a great conference. And we talked to like hundreds of people there, kind of trying to validate RankScience and what we were doing. Most of those folks are consultants or they run agencies, or they do in-house SEO for like big companies like Western Union or something like that. But we were just shocked that we were the only software engineers at this entire conference. And SEO is like pretty technical. And that is weird to me. And so I think there's sort of this like stigma attached to doing SEO because of its checkered history in the past where it used to be everyone kind of did link building. Everyone did so-called black hat tactics. And SEO kind of garnered a bad reputation. And so I think like a lot of programmers kind of steer clear of it or don't necessarily want to embrace becoming SEOs. And so that's something I see as an opportunity and an advantage for us.

Courtland Allen 0h 21m 23s

Do you use your own products to optimize your content?

Ryan Bednar 0h 21m 30s

Believe it or not, we don't really do SEO on our website yet. We only have four pages and we have big enough sales pipeline where doing SEO for RankScience hasn't become a major priority. This is something that people tease us about, but it's definitely something I really believe in inbounds. I really believe in organic search. We're gonna focus more on SEO this coming year. But at the moment, oddly enough we don't.

Courtland Allen 0h 21m 58s

What do you think of criteria that differentiate the company that can really benefit from SEO from a company that might wanna put it off until later?

Ryan Bednar 0h 22m 7s

It really depends. So one, if your company is growing in all these other ways, you're getting word of mouth, you're getting referrals, that's fantastic. So Airbnb didn't focus on SEO until later because they had this tremendous viral growth. Airbnb is a naturally viral product. And I remember seeing them at YC and telling them, that, oh man you guys just need to do some SEO. It would be this huge channel for you guys. And they didn't see it as that important because they were growing and they were totally right. It didn't make sense for them to focus on it then. And now it's a really big growth channel for them. But they didn't really focus on it, I wanna say, for like four or five years. So there's an example where it's not that important. If companies like spending a lot of money on AdWords and isn't investing in SEO, that's typically a mistake because the ROI for SEOs is so high and AdWords is so expensive. If your product is something that people are searching for it directly, or people are searching for topically related content, that to me looks like a really big opportunity. If you're creating some kind of app, or some kind of service that's totally brand new that people aren't exactly searching for it, that's where sometimes SEO or just search in general isn't that great or just doesn't make sense. And that actually kind of applies to us in a weird way, right? Because nobody is searching for automatic SEO CDN and because what we're doing is new and different. So that's kind of my thoughts there.

Courtland Allen 0h 23m 42s

What do you think people should do, where do you think people should go to learn about SEO if they don't really know much much about it. Because in the past, I've thought about using SEO to improve Indie Hackers. By the way, I'm gonna ask you what you would do to improve Indie Hackers' SEO later, put you on the spot. But in past, I've thought of, what can I do to make things better? And I've gone online and there's a whole bunch of different sources. There's Moz like you mentioned earlier, a hundred other tools and websites. And it's easy to get overwhelmed and just decide, you know what, I'm just gonna do this later. Where should people start?

Ryan Bednar 0h 24m 16s

I think Moz is like the best online community for learning SEO. I think that's the best starting point. And I believe all of their, or most of their educational content is free. They do have online courses too. So I would definitely point people to them. But then it never hurts, depending on where your business is at, to either talk to a consultant for some sort of guidance, or anyone who's done SEO or done growth before, will have some opinions. Those opinions might be completely colored by just like what happened with SEO at their most recent business or their current business. But Moz is a really great resource, and then just talking to people. There's this super active SEO community that publishes and republishes content nonstop on Twitter and all over the web. But Moz is probably the first place to go to.

Courtland Allen 0h 25m 9s

Are there any mistakes that you see beginners making very commonly with SEO? Or just quick wins, low hanging fruit that people in the audience might wanna know about?

Ryan Bednar 0h 25m 18s

It's tricky because it's so site specific. I think a lot of people don't know the very basics of SEO and so for example, they don't use Google Search Console. Google Search Console used to be called Google Webmaster Tools. And so I always spend a lot of time in there. I think most SEOs do. And so some companies are surprised when I talk to them, or if I'm just giving them advice or whatever. That they might be ranking on like page two already for some like pretty important or relevant terms to their business, and they didn't even know that. And that's something that Google Search Console would tell you. And then that's also a pretty strong signal that, well hey, maybe you if you actually focused on ranking for this term, or these sets of terms, and iterated on it, you could move from page two to page one. And when you move into the number one through number five spots, you actually get a decent amount of traffic in clicks. And so there's a lot of low hanging fruit just by kind of learning the basics.

Courtland Allen 0h 26m 21s

Are there like super simple tactics that people can use if they're going after a keyword, to move up in the rankings just a little bit? It is mostly like title tag optimizations or inbound links?

Ryan Bednar 0h 26m 32s

At a high level, rankings are still based upon links and content. And so it's roughly like 50% links, 50% on-page HTML. I actually think that on-page HTML segment is growing. And so Google does need to see that other websites are linking to you for you to be a relevant authority. In the past, I think that link factor was maybe a lot bigger than it is now. So SEOs got into a lot of trouble by just focusing on building links. And so there were these things called PBNs. People would create these link farms to link to their customers, and to link to their sites. That was a really easy way to game Google until it wasn't. Until Google learned what these link farms look like and what shady backlinks look like. And then they penalized all those people. And what I see now more commonly is that most business have, unless you're starting from scratch, most businesses have enough of a backlink profile that they're not extracting enough value out of their on-page SEO. So RankScience is completely 100% on-page SEO. We don't do any link building or anything like that. So if you're a brand new business starting out, you do need to get some links. Links from high quality sources are the best, but just links from actual real sources are fine too. So those can be your friend's websites, or those could be small blog posts from local newspapers or anything like that.

Courtland Allen 0h 28m 13s

Now I'm gonna put you on the spot. If you were to change anything about Indie Hackers, to make the SEO better on Indie Hackers, what would you change? I have Indie Hackers open in my browser here. And I guess first what I would do is ask you some questions about like, What is your typical or ideal user look like?

Ryan Bednar 0h 28m 20s

I have Indie Hackers open in my browser here. And I guess first what I would do is ask you some questions about like, What is your typical or ideal user look like?

Courtland Allen 0h 28m 34s

My ideal user is usually a developer who's interested in starting an online business or is already running an online business.

Ryan Bednar 0h 28m 42s

Got it. So it looks like to me, you're targeting learn from profitable businesses and projects. To me that seems like a good marketing headline. But it doesn't seem like something that people are searching for. And so I might leave that as sort of like your H one on the page, but I might experiment with like, trying to target something that people are actually searching for. And so maybe they're searching searching for successful bootstrapped companies. Let me search that and see if Indie Hackers comes up. So successful bootstrap startups.

Courtland Allen 0h 29m 20s

Okay, so the H one of the page should be different than the title?

Ryan Bednar 0h 29m 25s

Yeah. It can be different. I would start experimenting with the title and some of the on-page content to target a term that I think people are actually searching for that's related to what you guys are going after. You can do keyword research through some tools like SEMrush, is a really good tool for keyword research. Or you could just actually use Google auto complete. And start typing in things and see what Google recommends. As far as getting an idea of what are some terms that people are actually searching. So if I copy and paste this learn from profitable business and side projects into Google, Boom, Indie Hackers is number one. You rank number one, two, and three for this term, but it's probably too specific and just not something people are searching for.

Courtland Allen 0h 30m 4s

Got it.

Ryan Bednar 0h 30m 7s

So I think successful side projects is probably something that people are searching for. Side projects for programmers, side projects ideas, side projects to make money. I would come up with a list of keywords that are close to that, pick the one that you like the most that seems the closest to your brand, and start experimenting with that and the title and in the H1 or maybe add some paragraph content to the home page and see how Google responds to it.

Courtland Allen 0h 30m 34s

Perfect. I'll try it. And for anyone listening, this is how you trick an SEO into giving you free consulting advice. Is invite him on your podcast.

Ryan Bednar 0h 30m 44s

Well let me ask you this too, do you ever check out Google Search Console for Indie Hackers?

Courtland Allen 0h 30m 50s

I set it up a while ago. I think I looked at it a little bit for a week and I haven't been there since.

Ryan Bednar 0h 30m 56s

Got it. There might be some interesting stuff in there too. So you could go in there, click on search analytics, search queries, sort by impressions over the last 30 days, and that will tell you what the most high volume keywords are, where you're appearing in search. And you might actually find something interesting. Maybe there's some term you didn't realize that Indie Hackers is ranking for, but maybe not getting a ton of clicks for. And that's something you could kind of hone in on.

Courtland Allen 0h 31m 25s

All right, to go back to the RankScience story, you guys have gotten a lot of traffic from your content marketing, from your case study on Coderwall. What other ways have you guys gotten people to your site? And are you focused really on finding new customers or working with the customers that you have?

Ryan Bednar 0h 31m 40s

I told you first we did this post on Hacker News before anybody knew about us. That started getting us some leads. So that was really great, and really important. I actually experimented with promoting that tweet, so I had just tweeted out that link about the case study for Coderwall, and I had always wanted to try Twitter advertising. I'm addicted to Twitter. I probably spend way to much time on there. And I do follow like the S...

Courtland Allen 0h 32m 4s

Me too.

Ryan Bednar 0h 32m 6s

You too, yeah. And I actually had a bunch of success from just promoting that tweet. I think part of that is it was interesting content, it's a new angle on SEO. It's kind of a refreshing angle on SEO, right? Like this company is actually data driven and would they do A/B testing to validate their SEO changes? So I don't know, I probably spend just like a couple hundred bucks on Twitter promoting that tweet after it was on HN. And it got a lot of distribution that way. And it seems like a pretty profitable channel for us. I think that one single piece of content did kind of run its course though. Like over time, that kind of faded out. But that's something that we'll probably try again. As part of getting into YC, we are lucky enough that we had a TechCrunch post about us. And when the TechCrunch post came out, I was initially a little embarrassed, because the angle on it was like, RankScience is coming to replace your job. RankScience is coming for your jobs, like SEO consultants are in trouble. And that wasn't really something we were going for, like the messaging we wanted. And so when I first read it, I was like, oh man I'm not sure if I like this. But the writer actually did us a huge favor because the angle she took on what we're doing, and yes we are looking to automate SEO software, that's actually true. We're not trying to put people out of work, but we are trying to automate SEO software. But the way she kind of spun it was sensational enough so we had a bunch of leads from TechCrunch and then it really got shared on Twitter. So if you search for RankScience on Twitter, you'll probably see that people are still sharing that article. And so we got really lucky with that. And then Search Engine Journal picked us up sort of organically and wrote more about us. And we got a bunch of leads from that as well. So that was great. And so we kind of have this sales pipeline through this press that we've gotten. That's not totally sustainable probably for us like long term, but right now we had something like 1500 sign ups from that and are kind of still working through those leads.

Courtland Allen 0h 34m 10s

How does your business model work? How much do you charge people to use your services?

Ryan Bednar 0h 34m 14s

It depends on organic search traffic. So it depends how much traffic is actually going through our CDN. So we work with some sites that are really small and are just starting out. And then we work with some sites that are pumping through tens of millions of visits per month. And usually all that bandwidth goes through RankScience. And so our costs go up the larger a site is. Right now our pricing is pretty simple. It's basically anywhere between $2,000 a month, all to up to about $8,000 a month.

Courtland Allen 0h 34m 50s

Yeah that's interesting because the fact that you're a CDN with kind of like this value add of having A/B testing, people are already used to paying for a CDN, so it's not like they're gonna blink when you charge them money to use your CDN, cause they're just gonna switch from whatever they're using already. And then you can just charge extra based on the extra value that you provide with your A/B testing.

Ryan Bednar 0h 35m 10s

That's totally right. So companies are used to paying for CDNs. And then the other thing that they compare us to actually, is what it costs to hire an SEO consultant or what it costs to hire someone to do this full time. And so when I was an SEO consultant, I would charge anywhere from $3,000 to $6,000 a month and I could do a handful of clients at a time. Say the range for of someone at the high end, someone who's like really good, might charge like eight to 10K a month. Or an in-agency might cost that much. So companies look at us, and we're priced a little bit below what it costs to work with an SEO agency. That being said, I think we have a ton to figure out with pricing. Our pricing model is kind of simple right now. It kind of excludes us from working with sites that I think we could help a lot, who maybe haven't done SEO before and think we're pretty expensive. And so I think there's a lot of room for opportunity there. But that's kind of how things are working right now.

Courtland Allen 0h 36m 14s

Is there anything that you guys have done that's significantly grown your revenue? Like any pricing channels or any pricing changes or searching for different types of customers that might be more willing to pay you.

Ryan Bednar 0h 36m 27s

We've had tons and tons of people, especially from the SEO community who really want our software, but they think we're just really expensive. And there's always going to be people who want you to charge 20 bucks a month or a hundred bucks a month for your software. And those aren't ideal customers. And there's gonna be people who think your software should be free. Those definitely aren't ideal customers. So we've tried to ignore that as much as we can, and focused on companies who we can have a really big impact in helping grow their sites. And also companies that actually really value our service where SEO is important to them, but they don't have time for it. Their engineers are busy working on product. They wanna do A/B testing. They wanna be data driven. And we can really have a big impact. And so we're focused on sort of like good fit customers right now. I think you have to be careful in listening to what the market tells you when you start doing pricing. I also think that because we're not like cheap, sometimes you're gonna be too expensive for some folks. If there isn't anybody who thinks you're too expensive, then you're probably not charging enough.

Courtland Allen 0h 37m 39s

Something I've seen, especially with a lot of developers who start companies, is they tend to under price and part of it is that it just feels bad hearing people say that your product is too expensive and they don't buy it because of it. But another part of it, I think for developers at least, is when you see how something is made, and you're able to make it yourself, it's easy to discount the value that people who can't make it, hold it in. So if you're whipping up your own CDN, and you're like, okay I can do this. And you might think, hey it's not worth that much. Anyone could just build it. I'm not gonna charge too much. But in reality, it's almost always better to just charge more. It's better to err on the side of charging more than to charge too little.

Ryan Bednar 0h 38m 18s

I think that's right, because you can always kind of bring the price down later if that's actually true. If you're way too expensive, you can offer discounts, you can offer promotional pricing. But raising your pricing later is a hard thing to do. I feel like as developers, we naturally, I don't know why this is, but we scoff at the prices of software. Because I think we're like, well I could build that. And that would just take me a weekend. Or that would take me a day. Or that would take me like a month, why would I pay all this money for that? But as a programmer your time is so valuable. And I don't know why developers specifically don't wanna pay for quality software.

Courtland Allen 0h 38m 57s

So what are your biggest goals that you're working on right now at RankScience?

Ryan Bednar 0h 39m 2s

So we just finished YC. We're now kind of focused on growing the team. Our network is ever expanding, and so we need hire site reliability engineers, we need to make sure our CDN is fast and secure and really stable. And we're hiring front-end engineers as well as we build out more and more tools for our customers and for ourselves. So we have a ton of work to do, like really improving our product. I think we've kind of validated this idea or this space. We really believe that the opportunity here is big, then we'd really have to execute. So we really think that we're just getting started here. And we wanna scale over the next year by two or three x. Those are sort of our goals. That being said, YC just kind of came to a conclusion and we're kind of just now regrouping from this three month pressure cooker forcing function for us. And trying to figure out how to grow sustainably. And that's kind of where we're at.

Courtland Allen 0h 40m 14s

How did being in YC affect your business?

Ryan Bednar 0h 40m 17s

It was tremendous for us. So we entered YC at about 28K in MRR, and they said, we want you guys to hit 80. That was like the first week. And we were like, oh man I don't know if that's possible. And so they set these pretty lofty goals for a three month period. And then they kind of told us how they thought we could get there. And then they helped us come up with a plan. And then helped us do it. I think it was really good because as you're forced to grow, you learn about what parts of your products need help, what parts are held together with duct tape, you learn about some of your key assumptions. And then I talked to so many customers and so many potential customers during YC, that I think I have a really good idea of what sorts of features we need to build next if we wanna move up to higher tier customers, even larger websites, what sorts of things we'll need to do to get there. What things those companies care about. So I think it just forces you to grow and learn so much in a three month period that's hard to replicate if you're not in the program.

Courtland Allen 0h 41m 33s

Is there something that you would say special that doing YC, some sort of special fire that lights under you that you can't have on your own or that's harder to have on your own?

Ryan Bednar 0h 41m 42s

I think it's totally true. And so I don't think we definitely wouldn't have hit these growth targets without YC as a forcing function. But then also just the community, the network, being around these other startups who are learning, who are making mistakes, but are also like supporting one another. I think that's collective. It's so important. And that insight into, hey other startups have problems too, is really valuable. They do these things called group office hours where there's a handful of YC partners, and you're in there with say six or eight other companies, and you kind of go through how you've grown over the last week, what your biggest problems are, what you're concerned about, what you need help with, and going through that with a whole group of other startups is just like, I guess it's almost like group therapy, but it was really valuable for us. And that's something that changed from the last time I did YC with you in 2011; the group office hours concept. But we got so much out of YC. The YC partners were tremendous and I learned so much about sales from them. And just being a part of the network, for a lot of our early customers as well was super valuable.

Courtland Allen 0h 43m 4s

I really wanted to ask about sales. So what role has sales played in growing RankScience, versus marketing?

Ryan Bednar 0h 43m 12s

So I'm a software engineer who never did sales ever in my life. And I kind of was learning in starting RankScience, I think I was kind of afraid to get started. But as the founder of a company, you're kind of forced in to it. There's no way to sign up for RankScience at the moment without talking with me on the phone. Which is kind of insane, and we're working on a more a self service automated sign up right now, but learning how to do sales was extremely important for us. I think you can get by without much of a product if you're good at sales in the early stages. But it's also really important for validating what you're doing, getting feedback on what you're doing, figuring out pricing. There's so much that we've learned and so I think I went from a total noob at sales to someone who is like still pretty green, but I've at least leveled up in some areas. And there's a bunch of YC partners now who have grown big enterprise software companies. And getting advice from folks like that was tremendous.

Courtland Allen 0h 44m 19s

What's some of the best advice that they've given you?

Ryan Bednar 0h 44m 22s

That's a good question. I learned a lot about the proposal process, sales process, but at one point, I think we were putting together proposals for customers, I would just kind of send them out. This is probably something that's completely obvious to everyone except myself. But I would just send them out, and have a price attached to it that had to do with the organic search traffic, and some kind of like notes and opportunities that we saw where we think our software could get them. And I would just send it out to them over email like, okay, if you wanna work with us, that'd be great. But if not, no big deal. And I realized, or one of the partners kind of pushed me towards like, hey why would you send out a proposal without scheduling a follow-up call to review the proposal? And I was like, oh yeah, that's completely obvious. And once I started doing that, our close rate went up by like 30% or something. So there's really simple tactics that I was just unaware of cause I've never done sales before and still have a lot to learn. But there's certain ways like that where you can just start to level up.

Courtland Allen 0h 45m 28s

Yeah that's one of the interesting things about doing a startup, is that you end up having to wear all sorts of hats that you've never worn before. And so you're a beginner at a lot of different things. You just make really rookie mistakes.

Ryan Bednar 0h 45m 41s

Absolutely. For sure. One thing for us that is really important is we're a CDN, and so the most important thing is that we need to keep our customers sites fast, that they need to be stable. If anyone's site goes down, they don't care about their SEO and so we spent the first couple of months really just honing in on building a really secure and fast, solid CDN on top of AWS. I think I probably could've been doing more sales and more validations during that period, but it was worthwhile for us because the CDN piece is so important. I think in the first couple months of a startup, first six months, first year, you're gonna make mistakes and it's really about how you respond to them. How you bounce back. Having a co-founder is really huge as well I think. I initially kind of started working on this a little bit on my own, before Dillon joined, or before I was able to convince Dillon to join full time. But I really was kind of just spinning my wheels in a lot of ways. And when Dillon joined on, is when things really started to accelerate. So having a co-founder to lean on to get through some of those bumps in the road, some of those difficult times, is really crucial as well.

Courtland Allen 0h 47m 3s

How did you meet Dillon and convince him to join?

Ryan Bednar 0h 47m 7s

Okay, the last startup I went through, was TutorSpree and I was the CTO and was responsible for hiring and growing the engineering team. And Dillon went through this program called Hacker School in New York. It's now called Recurse Center. And I was trying to hire him to be one of our engineers at Tutorspree. It didn't work out. I think he ended up taking a job somewhere else, which was a great move on his part probably, but we stayed in touch and became friends and ended up both moving out to San Francisco. And the timing just worked out too. And we get along super well. And Dillon's fantastic and just kind of really got lucky there.

Courtland Allen 0h 47m 51s

So I wanna switch gears and talk about a topic that I like bringing up in these interviews, just because it doesn't get talked about that often. And that's the psychology behind starting a business. Do you have any personal habits or things that you do to stay productive and motivated as a founder?

Ryan Bednar 0h 48m 8s

I don't know that I do, to be honest. I talk to lots of founders who have these life hacks, where it's like they get up at 6 a.m., and they have this ritual, I don't know, they go for a run, and they make their coffee before they open their inbox, or whatever. I don't have anything like that. I don't stick to any really strict regimen. I try to give myself some time where I'm not focused on RankScience throughout the week. So I basically try to take at least one day off a week. But the first year of a startup, you are just kind of working all the time. And not to glamorize that, I think I really value working efficiently and working smartly, over just quantity of hours. But that's really it.

Courtland Allen 0h 48m 54s

Oh to ask you a different way, are there any times where you feel particularly demotivated about RankScience? And if not, why do you think that is?

Ryan Bednar 0h 49m 4s

So I have experienced this with other startups. I haven't experienced this with RankScience. And I think one, we've gotten lucky in that we've kind of hit on something that a lot of people are excited about or that there's a lot of inbound interest. And so we're really busy with trying to make our existing customers happy and then also trying to add new customers or onboard new customers. So yeah, I haven't experienced anything like demotivating. I think in the past, with previous startups, where like growth sort of goes flat, that can be kind of demoralizing and that can be hard to work through. So yeah, that's kind of my answer.

Courtland Allen 0h 49m 42s

Yeah, it's like traction is the cure.

Ryan Bednar 0h 49m 44s

Yeah, that's totally right. And I think when things are going well in a startup, you're sort of naturally energized, right if the company is growing. Even if you know that the products could be so much better, or there needs to be better human processes for doing manual work. There's inevitably lots of things going wrong in a startup, but if the company is growing, if revenue is growing, whatever your key metric is, that kind of satisfies all desires. And if the company is not growing, that's where co-founders or employees are pointing fingers at one another. Or problems are really amplified. And that can be really tough to work through. But really I guess, growth is the answer if you're trying to build a big company.

Courtland Allen 0h 50m 39s

Yeah, I think I've felt similar things in the past. Even not building a big company, just working on side projects and things. Indie Hackers for example, I've spent very little time demoralized about our spinning the wheels because it's been pretty popular since I started it. Versus older things like Taskforce, which I did YC with in 2011. Which never really took off, it was always kind of hard to work on, it was always kind of, I hope this next thing works. And so it ultimately seems kind of frustrating because the answer is like, okay if you wanna be motivated about your company, just build something that's successful. Which is not much of answer. But to finish up, what is one piece of advice you would give to people who are just starting out, maybe they haven't started their company yet, but they wanna get started.

Ryan Bednar 0h 51m 27s

I think for me, and this is probably too premature for me to have any major takeaways from RankScience, since it's only been about a year I've been doing this. One thing that I like about what we're doing, one thing that I think has been helpful for us, is we focused on an area that's kind of unsexy at SEO. And there's a lot of companies working on SEO, but a large part of what we're doing is kind of new or different. And I don't know many SEO companies in Silicon Valley that are working on this. And we don't think we really have any competitors yet. And so, I think choosing an unsexy area, or something that you think is really important, but other people are ignoring can be a really big advantage.

Courtland Allen 0h 52m 13s

Cool thanks for the tip. Well that is all the questions I have for you today. Thanks so much for joining.

Ryan Bednar 0h 52m 19s

Absolutely yeah. Thank you so much for having me. It was great sharing with you.

Courtland Allen 0h 52m 24s

All right, later Ryan.

Ryan Bednar 0h 52m 24s

Take care. Bye.

Courtland Allen 0h 52m 28s

If you enjoyed listening to this conversation, you should join me and a whole bunch of other Indie Hackers and entrepreneurs on the IndieHackers.com forum. Where we talk about things like how to come up with a good idea, and how to find your first paying customers. Also if you're working on a business or product of your own, it's a great place to come and get feedback from the community on what you're working on. Again that's www.indiehackers.com/forum.

Loading comments...