Ilya currently is working on MixRank, a Y Combinator backed company. He previously built a performance marketing company, and the one thing he knows best is how to get traffic fast.

TOPIC ILYA LICHENTENSTEIN COVERS

→ His a 500 startup mentor

→ Find out what the Mix Rank does

→ His thoughts on API

→ The tools and processes he uses for keyword research

→ His internal linking process

→ His strategy for a lot advertisers and how it evolved

→ His strategy to run a placement-targeted campaign

→ How to get traffic fast

→ And a whole lot more

LINKS & RESOURCES

→ Influencehacks

→ MixRank

WATCH THE INTERVIEW

READ THE TRANSCRIPTION

Bronson: Welcome to another episode of Growth Hacker TV. I’m Bronson Taylor and today I have Ilya Liechtenstein with us. Ilya, thanks for coming on the program.

Ilya: Thanks a lot. Thanks for having me.

Bronson: Absolutely. Now, Ilya, you’re a Y Combinator alumni. You’re a 500 startups mentor. And you founded a company called Mix Rank that even Mark Cuban himself invested in. So I’m sure we’re going to have plenty to talk about today. We can probably jump in from a lot of different directions. But let’s start with mixed rank. The company you founded that you’re currently at, what is it and who’s it for?

Ilya: Yeah, so MC’s rank is for online advertisers and publishers. Essentially, it’s a huge database of online ads and traffic sources, so we are a tool people use to find traffic sources. So we visit about 100,000 websites every day, pick up all the ads on them, organize them by advertiser groups, and try to identify which ads are actually successful. This is all digital ads. And so if you’re an advertiser, you can go to mix rank. You could search completely for free, actually, and find you could search by keyword or search for a competitor or see where they’re advertising. You can find by category traffic sources in a particular category and so on. So the main audience would be anyone who’s looking for traffic. So if you are buying ads, you can use an mixed rank to build out in your media plan. Find more channels for media buys. If you’re doing SEO, you could use this database to find publishers to link to you and so on.

Bronson: Yeah. No, that’s awesome. It sounds like there’s really a couple of different use cases for it. If you’re looking to advertise in new places, you can see where your competitors are advertising with their money, which you can kind of assume is giving them an aura. And so without having all that data yourself, you can see that it’s working. Like you said, you can use it for SEO, figuring out what keywords are using those kinds of things. So there’s a lot of use cases if you’re going to be going out and buying ads. Why did you create this product? Was it something that you just personally needed in a previous life, or was it some of the reasons you just see a market opportunity? What was kind of the impetus for beginning this rank?

Ilya: Yeah, exactly. So when I was in college, I started an Internet marketing business. I was doing performance marketing. So basically that was arbitrage where I would get paid for a conversion. Someone filling out a form on a page and then I’m sending traffic to that page, paying for that traffic, and hopefully spending less on that traffic than they’re paying me. And I was running campaigns in some very, very competitive markets. So financial products, data and that kind of stuff where there were already a lot of advertisers that had already done a lot of testing and optimizing. So when you run a campaign, it always, always, always starts off losing money. Always and eventually, yeah, it’s just what you expected. You’d have to be incredibly lucky to launch a campaign that’s profitable to right away. Usually what happens is there’s most of the traffic sources are creatives or keywords that are ineffective, and there are a few that are really, really effective that make up the lion’s share of your traffic and your profit. So 8020 rule pretty much. And what I saw was that, well, there’s already all these different advertisers. They’ve launched campaigns, they’ve spent money, they’ve already figured out what works and what doesn’t. And I thought, well, I have a much smaller budget than these big guys. Instead of me going through that same process, if I could just watch what they’re doing closely enough and learn from their mistakes and suffer from my own mistakes, then when I start driving traffic, I’m only going to target, or at least the majority of what I’m going to target is going to be the profitable good stuff. And so initially you kind of started putting together some internal tools for myself to do some very, very simple monitoring of other ads ended up coming out here to San Francisco, meeting my co-founder here, going through Y Combinator and all that. But the very, very first step was just thinking, okay, well, what is a process that I have that is very, very tedious or very expensive that I doing manually that I can use technology to accelerate. And so that’s where the initial idea for this came about. And then my other friends in the industry started using this or asking me to use it. And then I kind of knew, well, maybe there is a product and a market opportunity here, especially because if it works that. Our lie is very, very clear. So you could spend a lot of money testing and waste a lot of money on bad campaigns or to learn what works. Or you can give just some of that money to me and I’ll tell you what’s working based on everyone in the market.

Bronson: Yeah, absolutely. Because when you do a search on mic’s rank, you’re being shown data. That took some company tens of thousands or hundreds of thousands of dollars to actually get. And then you’re giving me that for, you know, a small prize without me having to have a $50,000 ad budget to learn those things that they’ve already learned. So it makes sense. It’s definitely a clear ROI for the person using your product. Now, when you do a search on MCS rank, it seems like you give it a lot of data. When I first went to it and I’m kind of like looking through it, it was almost overwhelming. Like, wow, like there’s so much data. I don’t know where to jump in this dream of again. I don’t know which data is helpful and which data is like the most important thing ever. Walk us through the data a little bit. The power users that really know how to use your product, what data are they really closely looking at when they do a search on your site? And then what do they do with that data? Tell us how your product is meant to be used by someone who really understands it.

Ilya: Mm hmm. Yeah. And that’s a good point that we’re kind of constantly thinking about at the company is the balance between being very, very simple and easy to use versus being comprehensive. So people say, well, you don’t have this thing, why not? So it’s a constant balance. And we try to address that with a lot of documentation and training videos and stuff like that. But the general use case, I would say our ideal customer is right now an SMB advertiser that is spending maybe thousand, a few thousand dollars a month on online ads. So it’s likely that they’re using either Google AdWords or doing direct buys on niche sites. So by direct buys, I mean you might go to a blogger, you might say, well, I’ll pay you 500 bucks if you put my banner in this spot for a month. And that is a very, very, very effective strategy for a lot of advertisers that are smaller to use, because you don’t have to worry about people outbidding you. You don’t have to worry about quality score, like the nuances of all these ad networks, how they optimize all that. You can just say, well, there is a site in my niche that I know is really relevant and valuable. I’m going to go to them and get some of their traffic because I know that the traffic they’re getting is very, very similar to my customers or the traffic converts for me. And so that is kind of the initial use case for mixed rank for a lot of people that come in where they have some kind of workflow, some process for essentially finding relevant traffic sources, finding sites to place their ads on. And what we can do is dramatically accelerate that process where we have a database of 100,000 sites and we can just filter, filter, filter that down to the 5010 whatever sites that are going to be the most valuable for you. And so typically when people come in, either they search for a domain or they browse by category or they search for a keyword. Mm hmm. And where they end up is a list of different types of search results that we have. So ads, landing pages, banner ads, text ads, all that stuff. The first step is that people look for if they zeroed in on a particular advertiser that is highly relevant and they found them by category or by keywords. So in other words, who in this space is the most successful right now? The biggest. So we rank all of our advertisers by estimated spend. So we can say here are the biggest ones in this category, whatever product they’re selling. And then with that advertiser, you can look at the report for that advertiser and we’ll show you all of their traffic sources, all the sites that their ads are appearing on. And the process after that really is just either browsing or exporting that list, figuring out what criteria is important to you that you can sort by. So we show you how much traffic each one of those sites gets. So you can go only for high traffic sites or maybe only for smaller sites that are easier to to get a deal done with a lower test budget. We can see how long that advertiser’s ads have been running on those sites. So they might test a bunch of different traffic sources, 80% of them, they’re going to pause after a couple of days because they’re not converting and then a few of them, they’re going to keep running. Those are the ones that are working. So we can sort by that. And we have about almost two years of historical data now and filter those. So. Once we do that, we have a list of maybe, let’s say a dozen or so publishers that are really, really relevant. And our strategy with them is either going to be to run a placement targeted campaign in AdWords or elsewhere, or to just reach out to regularly for those direct buys. And so now we can actually look at the publisher report, so you can search for that domain of the publisher and mixer, and you can see what are all of the other advertisers that are running on there. In other words, are there five different sites that are selling similar products that are all advertising on there? That’s a pretty good indicator that that’s relevance. Or are there a bunch of very, very generic campaigns that are just picking up cheap traffic that are advertising on there? And the next step then is building your ads, building your creatives so you can see, well, these are all the different banners that are running on the site. So you know what you’re competing against for clicks. And you can see what designs people are using, how the different designs that they’ve tested are performing. So we can show you the ones that have been running the longest and so on. And so you can find creatives that are much more likely to do well and on display advertising, which is where we focus. The creatives are really, really important because it’s it’s all interruption marketing. So you have to grab their attention, they have to click and it’s not like search where they’re searching for something. And so you find a list of relevant publishers based on what’s going on in the market and you figure out what creatives are most effective. And then you design those creatives or you just write some text ads. We also show all of those and then just target those publishers and go, That’s kind of the general workflow and that same process find traffic sources, design creatives, design landing pages. All of that is just the traditional kind of digital marketing process, and we just try to automate and accelerate big parts of that. So that’s our focus is kind of the same workflow that marketers are already using. But now instead of searching on Google for an hour to find those sites, we’ll just give you a list. So that sort of sounds.

Bronson: Yeah. And so it makes sense also. Now why that it’s some bees that are really using your product because if it’s your first day in Google AdWords and you’re like, Oh, I’m going to buy some words, you know, I want to buy some keywords. This, this product probably won’t make sense to you because, you know, if you can spend $50 in Google this month, it’s not really the product for you. This is for somebody who really understands what the data means, wants to look at the data, wants to be informed by the data, and has a big enough budget, like you said, a thousand or a couple thousand dollars a month to actually make it relevant that they have this information. So it makes a lot more sense about how how our use case can actually looks right.

Ilya: And if you’re or if you’re spending only 50 or 100 bucks or whatever, then your biggest gains are not going to come from split testing to get a few more clicks. Right. If you’re spending $100,000 a month, then a15 10% increase in your click through rate is actually meaningful. And that’s something that we’ve learned actually just as a startup where we put out something that was a freemium product so anyone can sign up. It’s really, really easy and we just wanted to watch who’s signing up, how are using ads, what kind of we track, how active they are, how many searches they’ve done, which reports they looked at all that. And so we actually learned that well, the advertisers that this is most valuable to the ones that our best customers, the ones that are maintaining their subscription month after month, are actually the larger advertisers and not just the the kind of mid-market ecommerce retail ones, but big brands, agencies, even Fortune 500 companies, which was really, really eye opening for us as far as to the further direction of the products. And that’s something that I think as a startup you don’t really initially target. You always go for the lower hanging fruit, kind of self-serve, not enterprise customers right away. But what I’ve seen is that and this is kind of a general marketing principle is that it’s very, very hard to educate the market on something completely new. It’s a lot easier to tap into existing demand. And so either you could spend all this time and money to educate your local corner business on, look here, there’s online advertising. Here’s how you do it. Here’s how you build campaigns. Or you could just go to the guys that are spending all this time building campaigns anywhere and say, look, we know that you’re. Doing this. We know this is important to you. Here’s an easier way to do it. And that strategy has been much, much more powerful and effective for us.

Bronson: Yeah, I think about mom and pop shop. Logged in a mixed ranking in a search. They’ll just be confused. You know, they’re confused by, you know, advertising in general. And then I think they would just be more confused by all the data. But like you said, you take somebody that this is what they do and they see it as like, oh, my gosh, like I have access to, you know, almost an unfair advantage now through all this data. And so I think you’re right to really go after the people that get it. You know, I have to educate them on the value of your product. I notice also that you consider a mixed rank to be more of a technology company than just another marketing company. And I see that you’re very interested in the engineering and you know what you guys are building and those kind of things. Is that one of the reasons you guys really put the effort into creating an API like you have?

Ilya: Yeah, yeah, absolutely. And you know, we’re always focused on the technology first. We’re always hiring engineers and building that out and much, much more than the the distribution side necessarily, because ultimately the value has to be the right. Like of all of this data, we have to make sense of it somehow and somehow filter it and bubble up to you. Here is what you should target. Here’s what’s working. Here are the different ads that people are testing successfully versus just. Here’s every ad that ran in the past year that’s not helpful. It’s too much. And so a lot of the things that we’ve done is on collecting as much data as possible, but then also filtering it and really somehow making sense of it all. And with with the API that was actually built just purely due to demand from these bigger customers that say, well, we just want to access the raw data or we want to integrate this somehow in our site, or we want to somehow resell the state or something like that. And so a lot of interest in somehow taking. So we did all the hard work of collecting all this and now people are saying, well, we want to mash it up and use it in different ways. And so there’s companies that do competitor monitoring tools where they ping our API and they pop up a notification every time a competitor at your monitoring launches a new campaign. There is companies that have business intelligence, kind of business information where now there’s another element of that dashboard is the advertising data. There’s companies that do lead gen and lead scoring where if there’s a lead that comes in that they ping us and then we say, this is a big advertiser right now that influences the score. And so there’s all these different things that we’ve doing that we would have never thought of that these people are just coming in organically and asking for. And that’s really, really good to identify new markets and explore different directions and really expand beyond sort of the obvious use case. So for advertisers, so one example that we’ve seen that we would have never thought of is that we have big brands that are coming to us and looking for access to the API or something else where they’re not even looking to monitor the competitors. They’re coming to us and saying, Well, just tell me where I’m advertising. Tell you where my own ads are running, because most of them don’t know. They really don’t know. There is not a lot of transparency. There’s a lot of middlemen. There’s a lot of ad networks where you target a category or even run of network or whatever, and they’ll just say, okay, well, give us X amount of dollars. We’ll just run your campaign on this category and we have this black box where we’ll optimize it for you somehow until you’re out and we update everything daily. And we have very, very granular visibility into that. So that that’s a use case where I would have never thought about that. And that’s a very different world than the one I’m used to. But it’s something that just came up organically because we we kind of open this up.

Bronson: Yeah, absolutely. And those are great examples. So it seems like the API is kind of being used in a couple different ways. One is just an extension of mixed rank. People are using it to get the data they want for their own reasons, just kind of internally. But then there’s also this other group of people that are building new services based on the data. Is that right?

Ilya: Yeah. And so that’s a great kind of growth and distribution strategy where new and this has been done for forever by certain kinds of companies where they put together a widget or something that you embed in your site links back and that that’s actually something that we. Developed on top of the API, where instead of writing code you can just come in and grab a widget and embed it with some kind of summary data or stats about a particular site. And that is something that is a very, very, very scalable strategy for us because we do one of these kind of widget partnerships, whether it’s just a completely free deal or it’s something through our affiliate program where they get an affiliate link. Reseller arrangements or whatever it might be. But the general focus for us as well, there are companies in our space that are more mature, that have spent a lot of resources on building up their audience. What is the quickest way for us to get access to that entire audience? And one of them is, of course, just having them embed something from us with our logo that’s branded that is appearing on every one of their pages. That is something that is a constant reminder of us. That’s a link back. It’s good SEO and all that, and that is kind of a good way of using technology to drive traffic and growth. So I like that.

Bronson: Explain to me a little bit more because I wanna make sure I’m really clear about how the widget works because people are listening and they want to be able to, you know, imitate good ideas. And so what kind of website is this widget on? Is it like a public blog? Is it on some kind of internal network? Where does this widget appear?

Ilya: So it’s typically a product. So it’s an either a business intelligence product or a dashboard or an advertising product. There’s companies paying.

Bronson: To have access to this product and make.

Ilya: Either either paid or either paid or free sometime, or they have a freemium model just like us, right? Or something like that. But the main sort of focus of this is that we provide some kind of incremental value. There is some kind of other information that we can give you. For example, if you have some kind of business intelligence tool where you have some business information about a company, you can add another mixing tab that has the digital advertising information. So yeah, yeah. So like how long typically on the widget it would be something, how long of their campaigns have been running, how many publishers they’re on, who are the top advertising competitors, stuff like that. Yeah. And when they embed this, typically the deal is either we’ll give you a little bit like the summary data for free. And we just ask for a link back for attribution with our logo and you should do better in your pages or something a little more involved where they’re actually saying either we will license this from you or conversely you will say we’ll put up an affiliate link in that widget where you get credit, we’ll do a rev share and the traffic you drive. And then typically they promote that more actively because they’re incentivized financially to do that. And it’s something that is a really, really, really good, scalable, consistent, repeatable partnership versus other partnerships that we do that are really nice. So we’ll do things like webinars with other companies, we’ll do guest blogs, all that stuff, and that’s great and it drives really good traffic and it’s good to build that relationship, but it’s a one off in a lot of ways. It’s not really scalable as something they embed that people see every day, that every time a new user signs up for their site, they see us, they come to us. And so that’s been really successful for us.

Bronson: That’s great. I like the way that you view mixed rank because I heard once I’m ripping off this quote, but that great technologists tend toward platforms and not just products. And so if ultimately you want to be a platform, if you can be I mean, products are good, but platforms are better. And so it seems like you’re becoming the advertising layer of knowledge on the Internet. And so if you can embed yourself in a lot of different ways and other people’s products, you’re the API. They come to you to get this data, you become that layer of the Internet and then a thousand flowers can bloom off of that later. Is that kind of the way you see it?

Ilya: Yeah, exactly. And that’s something that we learned. It’s not something that we just knew brilliantly in the beginning. It’s something that we learned over time and really exactly aligned. What you’re saying is that we are always trying to move from being a tool to a platform, because when you’re a tool, people pick up a tool to do a specific job. When you’re a platform, people build everything on top of that platform. And so what we were seeing really, really early was people using the products that they might log in, at least the casual users might log in once in a while when they’re building a new campaign. Use this tool. That’s great. And then that’s it versus something that is very deeply integrated into their workflow and something that they can keep reusing consistently and is an integral part of what they check basically frequently. So it’s not just something that we look at as a media planning or campaign building tool. There’s also that higher level business intelligence, advertising, intelligence component that people are constantly checking. To see what their competitors are doing to monitor new traffic sources. Coming up, new developments. That kind of stuff.

Bronson: Yeah. And I think it’s easy to see that long term. That’s the direction. You know, I think in the short term, you know, mixed ranks is a great product and you should go use it. But long term, anyone could build mixed rank using the API, but it’s the API they can’t build. It’s having two years already of data being collected. They can’t just go and instantly have that. That seems to be the true value when it comes down to it. Of what mix rank provides on the Internet. So I think it’s great. Let’s talk a little bit about how you’ve actually grown mix rank. You alluded already to partnerships and you also alluded to some of the the webinars you’ll do things like that. Let’s dig into those a little bit more, though. What are your primary channels for growth? If you had to list them out, would partnerships be number one or would something else be number one? And then how would you kind of stack or income if you had to?

Ilya: So it’s kind of hard to to rank them and this is surprising to me, but we actually get a lot of users where we have no idea where they came from so organically. And so that’s that is a part of it that’s probably the biggest just organic kind of word of mouth as far as things that I’ve actually done that have been effective. And we’ve tested a lot of different things. So I could talk about the things that work and the things that didn’t work.

Bronson: I want to hear about both because if you tell us about the things that don’t work, it saves us from having to maybe do things that are dead ends. So, you know, as much you can tell us about both of those would be great.

Ilya: Yeah. And so this is for if you’re thinking kind of standard software as a service, B2B products probably priced between 20 and $100 a month is where this would apply. It’s very different for consumer and enterprise. But the thing that I think has really been effective for us is investing in content and really building a voice and doing things like this, but also writing blog posts, writing guest blogs, participating on forums and Twitter and everything, and really sort of supporting the products with content along with partnerships where that feeds into kind of it’s all the same thing. It’s all about content, right? And always for us, we always want to create content that is valuable on its own. So it’s not a sales pitch or anything. And so we will do webinars with another company where it might be 80% just content. And here’s how to do something here. The trends in the industry, whatever, 20%. Here’s how you can do this a lot faster with next rank towards the end. And so a lot of what we’ve done that has been not just successful in bringing in users, but more importantly bringing in customers and high quality customers because we do a lot with lead scoring and qualifying and everything. And it’s something that as far as bringing in the most valuable users for us, for sure, our content is number one. So what that means is of course writing posts or writing something like a white paper or an e-book that you can download if you give us your email, newsletter, emails that are consistent, all of that stuff and all of that comes together to basically say, Well, this is what you’re doing right now, this is what you should be doing. These are the best practices in your industry, and here’s how our product can help you implement those best practices. So that’s the general framework around which we we structure our content and producing it is actually the hardest part.

Bronson: So driving traffic about next, how do you go about producing it? Because, you know, a white paper doesn’t just up here, someone has to put in a lot of time and effort to really pull it off. So how did these things come to be?

Ilya: So a lot of what we do is we have an email account that’s kind of like a dummy email account where we sign up for pretty much every newsletter or mailing list imaginable and we just see what comes in. This is all for marketing software companies. And so we see what webinars they’re doing. We see what they’re writing about. We follow him on Twitter or whatever. And so that kind of gives us a pool of. Here are some things people are talking about. And also with that. Like any good copywriter, I keep a slight file of good headlines and everything else that they send. And with that, we kind of know. Okay, well, bunch of people in the industry are talking about mobile advertising or a bunch of people in the industry are talking about split testing or whatever. And then we try to be part of that conversation, link to them in a post, reach out to them, whatever. And so that’s kind of like getting traffic to it is easy if it’s part of the topic that people are interested in or if it’s part of the conversation, it’s part of what people want to read about and are reading about. And so we just write kind of pretty efficient and kind of short posts that are about that that will say, well, here are three things. Three bullets about how to do X. Mm hmm. Then we’ll flesh it out. And so maybe each one of those becomes a paragraph in a post, and then maybe later, each one of those becomes a section or a chapter in an e-book. If there’s more to be said about it, or if there’s more interest in.

Bronson: An in-house or. Yes. Doing it in-house.

Ilya: Yeah. All in-house. Yeah. So we, we do that and we should probably do more of it. But it’s something that is really a very well-established way to build credibility and build trust, where when you send someone an email or even a bulk email, they are a lot more likely to open it if they know you and if they’ve already heard from you a few times. It’s not just a random spam email. So we produce that content. We try to reach out to other people in the industry and say, Hey, we’re talking about us to do you want to do something together about this, like a guest blog? Do you want to do a webinar or maybe about this and share the list you want to link to this, whatever. And people are more than happy to, especially when we say, well, we have this many users that we’ll share this with. Yeah. So that the motivation and the incentives are aligned really nicely where everyone benefits. And those kinds of partnerships are something where you build as a start up a lot of credibility to the story. If you have these big guys promoting you and linking to you and whatever, it’s really good. SEO It’s really, really good for building trust and so on. And so that’s a big part of what we’ve done that has worked along with just the general product itself, where we thought a lot about the SEO on the product and every ad that we see is its own page that is all interlinked with others in the hierarchy. So go to the advertiser, report for that advertiser. That’s all the links. And then from there that’s interlinked to publishers that are running it. And so an echo. So our search engine bot can navigate the whole site from any point.

Bronson: So your site’s not just being populated in code on the fly, you’re actually building out these different pages. Yeah.

Ilya: So these are static. Yes, these are static pages for every unique unit of content, whether that is an ad or a landing page or whatever.

Bronson: And so I might not do that sometimes because it seems like it seems like an obvious thing to do, but it’s rare to find a company that actually realizes what it means to build out these separate pages, even though they could just create and code easily. You know what I mean?

Ilya: I have to be careful. Obviously, you don’t want it to be too spammy. You don’t want to look like you’re just creating some kind of doorway site that has these millions of pages pop up overnight. Sure. So the big thing there is that’s kind of our unique thing is that, well, we have all of this really, really relevant, human readable content text ads, right? So it’s not like spam. It’s regular content that someone wrote. So it doesn’t. And also we have enough links from press and everything else where you have. Yeah, yeah, we have a higher trust rank with the search engines. And so yeah, they’re a lot less likely to penalize us. Know someone? Like Amazon or something creates millions of pages or an e-commerce store. They don’t get penalties for that. So it’s something that is not necessarily a strategy where we try to rank number one everywhere. It’s more of a long tail where just all this content and just naturally some of it is going to rank for whatever reason, just because there’s so much to it without us doing anything. And that is something that just comes in and builds up over time organically.

Bronson: Yeah. So you’ve mentioned you mentioned partnerships as a growth strategy. You mentioned content. You mentioned SEO. Anything else that you’ve tried that has worked? Well, we’ll stick with the has worked for a second.

Ilya: Yeah. And the hazard I guess is more on the after someone signs up. So I would consider that still kind of marketing growth hacking there. Yeah. After I sign up, what do you do? And so we send a lot of auto responders and drip email campaigns. We send them documentation. We know when users haven’t used the product in a while and we send them an email as a reminder. Just all of that kind of email marketing and just basically staying as engaged as possible with the users. And as I mentioned, a big part of what we’ve done was lead scoring is really looking at where do they come from, how much are they using the products, are they coming from a corporate domain or is it a free email account, all that stuff. And so as we start building out more enterprise products, which is kind of our strategy going forward, we have this list of prospects that are engaged with us. They know our products, they are using it, and now we can push that into sales and reach out to them. So a big part is just, okay, well, after they’ve signed up, how do we turn them into a customer in a scalable way without having a salesperson talk to them? And that’s what we thought about a lot.

Bronson: An email is a big piece of that for you guys. I also want you to touch on maybe the you mentioned it’s a freemium model. And I think that you guys have put the paywall in exactly the right spot. You know, a lot of times I go to websites that I think, why are they giving me so much for free like you can charge at this point and they keep giving it to me free or the opposite. You know, they charge too early. When I went to your site earlier in the interview, you said 8020 rule, right? 20% of your ads are actually bringing you 80% of your revenue or your success. Well, on your site, I don’t get to see the number one or number two ad that someone’s running. I get to see the ones below that as the free part. And so to see the 20% that really works for them, I have to pay you money. And I think it’s just the perfect spot to put it because you’re you really understand what the value is, what the value is, and where the pain is. When they’re ready to pull out a credit card. It seems like you just kind of. Nelda, does that seem right to you?

Ilya: Yeah, it’s something that, again, kind of out of experience seeing what are the leads that score highly in the ones that are the bigger advertisers? What are they doing? What are they looking at? What’s valuable to them? And so where we cut it off is that you can see a few results, a few ads, a few traffic sources for free, actually, one step back. So you don’t even have to make an account to use the site once you’ve visited. Yeah, once you’ve visited a few pages, then we ask you to make an account to see more. So right now. Or you can view five pages of reports. Then you have to make an account, you make a free account and you could keep view. And now you have unlimited access to all of the reports, but all of them only have basically the first page of results. So there’s some data that is useful to you that you can kind of start with right away. You can look up who, browse who the big advertisers are or find a few ads, whatever. But if you actually have the money to be building campaigns, if you are actually getting value out of knowing what the most successful campaigns are, then you can move through the process to upsell and upgrade it and so on.

Bronson: Now I think it was great kind of how you guys, you know, did all that. What are some of the things that haven’t worked that you’ve tried? Because not saying this won’t work for anybody, but, you know, for whatever reason in your situation, it didn’t work.

Ilya: Well for startups and it’s not just for the US. It kind of showed us over a lot of companies that I talked to. The thing that surprised me the most is probably that paid advertising kind of paid search. AdWords did not work for us nearly as well as we would have thought. And my entire area of expertize for the past five years before this has been driving paid traffic to products that say something that didn’t work. And I think what we kind of saw as a startup, especially as a B2B startup. Is when you’re treating something different or something new. Mm hmm. People are not searching for that product. Most of them probably don’t even know it exists. Or they don’t know what to call it. They don’t know the terms. Things like that. Right. And so what happens is that either you run a very, very targeted paid search campaign that might convert, but it’s very, very little traffic or you target more broadly and you get a lot of irrelevant clicks and you get a lot of clicks. Ads all convert because people aren’t really understanding what a how this all fits in. So they’re mostly on their process. So that’s probably the biggest thing that a lot of startups think they can just go to AdWords, throw some, throw some investor money at it and just go get growth. And unfortunately, it’s not that simple and you have to be a lot more creative and build channels that are a lot more scalable than just your typical pay per click campaign.

Bronson: Yeah, I’m glad the way you nuanced that because we have some people on the show and they say AdWords you know Google ads best thing ever throw a bunch of money grow fast. Other people come on, they think it’s a joke. They kind of snarl at it like, oh, you’re just going to do AdWords. But then you say, Well, it’s because of the kind of company they were. They weren’t searching for this thing because they don’t know this thing exist. I mean, what are they going be searching for? I want to spy on my competitors and understand what ads are running on various networks. Like no one’s doing a search like that. And so I’ll likely nuanced. It is that there’s a time for a company to go that route, or you wouldn’t have users of your products because they’re obviously doing that in their own world. But for in your particular situation, it just wasn’t the best fit to grow and grow as well as one size fits all.

Ilya: Yeah. And if you’re selling a product that people already know about and they’re looking for that, so there’s always someone that’s going to have to spend a lot of money on creating demand. And hopefully it’s a big brand that’s already spent money on creating demand to make a people that really want a widget and you just made a better widget that then you can take that market share. So in that case, absolutely paid search is really, really effective. But if you have something that it’s not entirely obvious that this is a replacement for an existing thing people are searching for, it’s going to be a very challenging path to actually get scale and volume in the way that you need to do to grow just from those paid search or pay per click campaigns.

Bronson: Yeah, makes a lot of sense. At the beginning of the program, I mentioned that you’re also a mentor at 500 startups. So I want to kind of talk about that for just a moment. What are some of the misconceptions that you see among startups in terms of growth when you talk with them? Because they’re coming to you as kind of the expert on growth, the expert on ad buying especially. And so you get to see all of their weird, you know, misguided thinking. What are some of the misconceptions that pop up?

Ilya: The biggest one, I think we we just kind of touched on a little bit is they see buying traffic is kind of the magic bullet and just something that will especially in the early stages that’ll get you that traction or help you raise money, all the stuff. And what what I’ve seen with a lot of these startups is that they’re just not at that stage where they should be paying for traffic.

Bronson: So that’s why they’re not at that stage. Yeah, because some people watch them this are not at that stage either and they don’t know it.

Ilya: Right. And they shouldn’t try to spend money on advertising because it won’t be successful. So the biggest thing is just on the product side and this happens with every startup, so it’s normal, but either they haven’t figured out exactly the messaging on the product or the value prop or the product, or it’s something that the problem they’re solving is not the biggest burning problem that those people have or just the business model doesn’t support it. Right? You can never, ever build a business on paying for traffic that goes. I just want work. And it’s not something that you should be investing on if you are having to buy very, very, very expensive paid search traffic and you think, well, I’m going to monetize it with advertising that is cheaper and just the math doesn’t work. Even if you have a 100% click through rate, you’re making less per click than you’re paying Google for that traffic. That’s why Google makes money. They pay their publishers less than they charge the advertisers. And so the business model has to support that. User acquisition model is something where, you know, the same in reverse. Like for us, we don’t need to build a lot of viral features or kind of sharing like sign in with Facebook, all that stuff, right? Because that is something that is suited towards a certain consumer focused business model. The places where paid advertising make sense is where your pay to acquire that customer is going to be either very low. Right. If you think of like an app paying for installs, something like that or the value of that customer to you is so high that you can pay a higher CPA. So for example, there are startups that are in the financial space where they drive traffic and then they help you kind of like meant something like that, right? They help you find financial products. Well, they’re getting paid an affiliate fee for every single financial product, but they refer you to that might be 30, 40, $50. They can afford to pay five bucks a click to have you come to them. And the most important thing that every startup to just have a really solid understanding of is their customer lifetime value. How much can they afford to spend to actually in a scalable way, not just by looking for bargains or long tail, but in a scalable way to drive a meaningful amount of traffic where there is less than that lifetime value. And I think a lot of startups sort of skew towards the acquisition side of the equation where they’re like, how can I write better ads, get a better click through rate somehow, optimize whatever? And that’s important. But those are in many ways incremental improvements, you know, maybe 20, 30, 50% improvements. What they should be looking at more is on the product conversion business model monetization side, where they might get a ten X improvements because they found another offer that they could push to their users, something else they can sell and so on. And so that’s what they should be focusing on because if you have the monetization side, right, if you can establish a big enough market that you can monetize, then acquisition is not a problem, right? You can get as much traffic as you want. All you have to do is just pay whatever it costs. So if you are getting to a point where, you know, every single visitor to your site is making you a dollar or is making you $2 or whatever, then the world is your oyster. You should go to Facebook and just bid a dollar per click and get as much traffic as you can handle. That’s not the hard part. The hard part is monetizing the traffic, so that’s what they should be working on.

Bronson: Now, that makes a lot of sense. I’m so glad you just said that. And we had somebody in the program a week or two ago that said the same thing. But you took it a step further by saying that by tweaking the funnel further down, closer to the revenue side of things, not only does that make sense, but then you can afford to go back to the top of the funnel and just spend all kinds of money if you want to. But you can’t do it the other way around. You can’t get all this traffic in the top of the funnel, not figured out the revenue side and the business model stuff, and then circle back around and spend more for acquisition. That doesn’t work because the money’s not there. But if you start with amortization, you can go back the other way.

Ilya: Yeah. And the one thing I would add to that is that generally right now in 2013, I think most traffic is fairly priced. So you basically get what you pay for. So maybe ten years ago there were a lot of long tail keywords that people haven’t discovered that were really cheap. But now at this point, with how many sophisticated advertisers that are out there, pretty much if you are paying $0.20 for a click, then there are not a lot of advertisers out there who have figured out a way to make more than $0.20 from that user. The market has kind of stabilized. And so what you really should be doing as a startup when you’re trying to do something completely different is actually figuring out some totally new. Monetization model for cheap traffic. That’s kind of the key. That’s really the key of being successful. That’s the ultimate growth hack is like, how do I take traffic? That is currently very cheap where the market doesn’t realize its value and monetize it to a point that it has not been monetized before. Great, great examples for this is Pinterest. People say, well, how is Pinterest valued so high? Like just a social network? It’s whatever. But the value to them and I think they’re actually underpriced based on their users is that they’ve managed to attract a very, very specific demographic, which is women that are looking for products to buy. Right. That’s why they go on vacation. Yeah, right. They have the women have most of the buying power in many households and they’ve managed to tap into an audience that is very, very, very hard to acquire if they’re not an audience that spend as much time online as the male geeky audience that is plentiful everywhere. And so if you can take that traffic and just say, well, here’s a way to either have them buy products that they have not been purchasing earlier online or have them spend more money, or here’s some way to have them engage more with my product, have higher engagement because I’m paying whatever cents per click for that user. But now they’re my user for life and they check back every week or every day or whatever. Any one of those two things is either more dollars per visit or more visits, roughly, if you could figure that out better than other people in the market have figured out, then they’re cheap. Traffic becomes a lot more valuable to you and that is something that is scalable. And now you could go to investors, you go to VCs and say, well, look, I have this model where I spend a dollar on a user that’s going to be worth $5 to me in three years. Now you can find this can throw money in and just build up that audience. Yeah.

Bronson: Yeah. I think what you’re seeing right there, in case people aren’t following, they need to rewind. Go back and watch that last little segment again. It’s the kind of advice that you just have got to get your head around if you’re going to grow a startup. This idea that, yeah, there’s a marketplace and for this for these ads, for these keywords, right? So maybe it’s $0.15 a click or whatever, but that $0.15 is not worth the same to all the companies that bought it to one company. That one click is worth $3 to one is worth $0.05. And so it’s what you do after the click that really decides what you do with the click. And so, yeah, I think it was great advice you just gave us there.

Ilya: Yeah. And to that, just one more thing to add as we’re seeing that happen right now is with international traffic. So there are a lot probably the majority of companies that are buying traffic online, they’re targeting a US North American, maybe European audience. And if you look at where the fastest growing segments of Internet users are. They’re not there. They’re in China. They’re in South America, particularly places like Brazil. They’re in Russia. All these places where the traffic is just orders of magnitude cheaper. It’s just dirt cheap. You can go to Facebook, look at their self-serve ad tool, and you can just play with different targeting segments and see how much the bids are. And it is just dirt cheap. So if you have some things that you can figure out some way to just and this is why there’s a lot of companies. Right, that just replicate the business models of American companies abroad. But if you can do that in some way where you can have just a different model and even beyond the e-commerce model, like if you can figure out some way to monetize people who don’t have credit cards necessarily, there’s a company in India called Flipkart that’s kind of blowing up right now, and they are an e-commerce company and they’re local to there. And their main twist was that they accept cash on delivery. So it’s a little business model change, right? India, a lot of Internet users, really cheap traffic, not a lot of credit cards. And so the traditional e-commerce companies couldn’t break in there. But they do this where they say, well, you can just pay cash when it’s delivered. And just that business model tweak means that all of the traffic that has previously been disqualified by all the incumbents because they’re saying, well, they can’t buy from us with a credit card, they’re worthless to us. And they see, well, look, all these people. Ah. Clicking. They’re not converting. They’re not. It’s not that they’re not converting because they don’t want the product. They can’t. Yeah, exactly. They can’t. They have no way to convert. And so if you can figure out just some kind of model like that where you can convert the ones that others could not convert, that’s where you can build a really big, meaningful company really quickly. I think.

Bronson: That’s awesome. And then that puts the the owner is back on us to be creative. You know, we’re at the very beginning of the Internet. That’s what people don’t realize. Like it hasn’t even begun yet. It’s a blip on the radar of what it’s going to be in the next ten years, 20 years, 50 years. I mean, like there’s still places in India without credit cards. Like that’s where we’re at with the Internet, you know. And so the creativity is the only thing limiting us as we really think about globally what markets we take advantage of, what can we arbitrage some other place like there’s a million possibilities.

Ilya: Yeah, there’s a lot of things. And especially on something like mobile, I was looking at some data recently that just blew my mind. So something like I think something like 30% of teenagers right now, the primary way they access their Internet is through their mobile device. That’s it. They’re just. So if you are only targeting Web traffic, you are just missing out on that completely. And this is not because they’re not affluent or they don’t have a computer or whatever, but they just grew up with these mobile devices, with smartphones. And that’s their method of getting online. Same thing with places like South America and Africa where they that are it’s amazing. They’re all online. Most people in many countries in Africa are online. They’re on Facebook actually, and they’re using Google and everything else, but it’s all through their phones. So they just skipped the whole desktop PC part entirely. And that is something where there’s going to be completely different creatives, completely different landing pages, everything else, right? Where we’re still kind of stuck in the standard banner ad paradigm from 15 years ago. That is going to obviously shift significantly when people that are just engaging with content and totally different ways. And so I think as a startup you have a lot of freedom to do that, to just try different things and different ways to convert, different ways to monetize. And even brands are doing this now. You’re looking at some ads that McDonald’s was running on mobile. And so normally it would say, you know, enter your email, sign up for our newsletter, whatever their call to action on their landing page, it just says, click here to find the nearest McDonald’s on your maps, on your phone. That’s like that’s their pay. And so something like that where we have all of the stuff like location, everything, we have different ways that people are consuming advertising. So a lot of is in apps, it’s not on the web, things like that. There’s a lot of opportunities to not just say, Well, I’m going to go do the same AdWords campaigns like everyone else, and I’m going to try to think of smarter keywords on them, but just skip that entirely and say, Well, I’m I have a different distribution strategy. I have something that’s naturally viral, I have a different channel. I have a different, if you will, a different creative for an ad that might be a widget that displays some data about a company. And that’s what’s going to get much more clicks rather than me optimizing the same banner ads.

Bronson: Yeah, that’s great advice. And there’s one more area I want you to speak to us about real quick, that you also have some expertize when you teach a class on startup plays about affiliate programs. So I want to ask you, when when is an affiliate program a good idea for a startup?

Ilya: So the we’re going to assume, again, that you have the business model to support that and you can pay for it. Generally, the biggest issue with affiliate programs is just lining up the incentives correctly and lining up the rewards that you give your affiliates correctly there and figure out where they’re driving traffic. Right. So if someone has an email list where it costs them nothing to make that list, that’s very, very different from someone that actually is going to be building campaigns and sending traffic through your affiliate link for you and spending money.

Bronson: Enough money for what they’re saying.

Ilya: Yeah. So yeah, right. Exactly. So that’s makes sense for them. And then the other biggest thing is just quality. So you can structure your CPA that you’re going to pay out on as anything. Right? You can structure it. And in fact, most people that make money with affiliate marketing are not doing a rev share commission kind of Amazon type thing. Nobody makes money with Amazon. It’s all seem well, maybe not nobody, but most people don’t. Most of them. Right. So. The people that make money with Amazon are the ones that put it on their blog ten years ago, and it’s just there. But nowadays, it’s all CPA Legion sending traffic app installs, that kind of stuff where it’s a free action. And you if you’re going to go down that route, you’re going to attract a lot of affiliates that are interested in getting paid for a free sign up to your site. But you have to monitor the quality of the traffic they send very, very closely to make sure those signups are actually backing out for you, converting and monetizing them. So there’s kind of a risk reward balance. Either you could go low risk and say, we’re going to do rev share. When I get paid by this guy, you get half. That’s great. But if you think about it, if you have a product that you charge monthly. Mm hmm. I’m an affiliate. I’m buying traffic. I might not make my money back on that traffic until nine months later when they’ve paid you enough that that comes back to me. And so typically and this has been done, pioneered by, say, web hosting companies, they say, well, we know on average someone sticks around for six months, so we’re going to pay you a high CPA like 60, 80, $100 CPA upfront for that sign up. That’s higher risk for you because they might cancel, might be fraud or whatever. But the affiliate gets incentivized and gets their rewards upfront. And if they’re paying to send traffic to you, they get more funds so they can put back into growing those campaigns for you. So it’s kind of a balance and you really have to just watch and monitor that traffic quality really, really closely because otherwise it almost sounds too good to be true, right? Like you only pay. Yeah, you only pay when you get a customer. That’s awesome. Let’s go. But what you don’t realize is that there are very, very, very huge differences in traffic quality and how they back out and chargebacks and all all this stuff. The big difference that people don’t understand is that there are huge differences in traffic quality and you shouldn’t just be looking at the absolute number of clicks, should be looking at the quality of those, somehow scoring those and somehow identifying, well, this affiliate is sending me really, really good traffic. I’m going to reward him and pay him a higher CPA or give him a bonus or whatever. This guy is sending me low quality traffic. I’m going to ban him or lower his payouts. And big legion companies, big brands have dedicated people that are affiliate managers all day and they just sit there and monitor their quality and dole out rewards and punishments based on that. And it’s something where most of the traffic you get is not necessarily going to be your best customers. What you want to do is be able to identify, these are my best customers. Where can I find more like them? What do they all have in common? Where do they go online? What do they need? Who has access to them currently? Like with a newsletter or something that I could do an affiliate deal with, something like that. And rather than just saying, These are the demographics that I want and let’s just get as many of them as possible. Yeah.

Bronson: Are there any tools that you use or that you’ve used before, you know, platforms for, you know, managing affiliates? Or do you just do a kind of in-house the hard way, you know, managing other.

Ilya: Yeah, there are a few tracking platforms. There’s one called has offers that that’s pretty good. There’s another one called Head Pass that’s a little more expensive. There’s one called Link Trust. But what you want really is just the most important thing with managing is to just not open up the affiliate program to everyone, but rather to approach it as more of a partnership model where you actually talk to them and you know them and trust them. And then you say, Well, we’re going to track this using this third party tracking service, whatever it may be. So, you know, we’re not going to fudge the numbers, but also we know who your audience is. We know what you’re doing to drive that traffic. Right. You always want to know what they’re doing to drive that track.

Bronson: And forget it. Stay on top of the thing.

Ilya: Just like everything like.

Bronson: You know, in the kitchen. Or is it when the cat’s away, the mouse will play? It’s like, if you’re not watching, then all of a sudden you’ve got bad quality traffic and things are just breaking.

Ilya: Yeah. And so one thing that is typically done is that an advertiser will set a cap on how many leads they will pay for. So you might say I’ll pay for 50 leads a day. For now, we’ll monitor the quality if it’s good. A month later, we’ll raise the cap. And so that mitigates your downside somewhat.

Bronson: Yeah, absolutely. Well, Ilya, this has been an incredible interview. I want to ask one more question before we leave. What’s the best advice that you can give to someone who wants to grow their startup? You’ve talked about a lot of the tactics, a lot of this tragedy’s a lot of what you’ve done that has worked and hasn’t worked. But what’s the best advice you can give to us? We want to grow a startup.

Ilya: I think the best advice I can give is just to. Cast a lot and test as quickly as possible. The mindset that you have to have is that starting out, you’re not necessarily paying for traffic or conversions, you’re paying for lessons, you’re paying to learn. The signal doesn’t. And the way to learn quickly is to do a lot of experiments as quickly as possible. So if something is not converting or not working as a campaign, don’t try to resuscitate it and bring it back to life and turn it around. Generally, something starts rolling downhill. It’s going to keep rolling downhill. You just kill it. You know, it’s the same thing as gambling, right? Sometimes you just have to cut your losses and get up and walk away. And so you have to be really emotionally detached from these campaigns because most of them will fail. And we just have to keep trying very, very different things that what you always want to do is just think, okay, well, there’s a channel. How can I test that channel as quickly and as cheaply as possible and then move on to another one? So that’s what you should be thinking. Now you could think about statistical significance, how many clicks or impressions you actually have to get, all that stuff. But just on a high level, you’re smart to not limit yourself to a few channels and test a lot and learn from them, because you’ll always be surprised at what actually takes off and explodes. I mean, I’ve had campaigns go from making $50 a day to $2,000 a day, like overnight. Just because I found that one thing, that one publisher something that one audience that converts really, really, really well. And I would have never found them unless I tested ten other ones that all lost money before I hit that point. It’s kind of like wild catting for oil or panning for gold or something where you have to keep just testing different places and one of them you’ll hit it, but you never know when that’s going to happen, so you just have to keep doing it.

Bronson: Yeah, it’s great advice. Hit on Ilea. Thank you so much for coming on the program today.

Ilya: Yeah, thanks a lot. I enjoyed it.