How to Build a Successful Software Start-Up With Michael Erickson Facchin of Ad Badger

Guillaume: Hello everyone, Guillaume Le Tual here, host of the Ecommerce Wizards Podcast where I feature leaders in e-commerce and business. Today’s guest is Michael Erickson Facchin, CEO of Ad Badger. He’s an expert in Amazon Pay Per Click advertising and he’s the creator of a software company Ad Badger, which manages all the Pay Per Click Amazon stuff for you. Today’s topic will be how to start a software company and market your new software to success.

Before we get started, we have a sponsorship message. This episode is brought to you by Mage Montreal, if a business wants a powerful e-commerce online store that will increase their sales or to move piled up dormant inventory to free up cash reserves or to automate business processes to gain efficiency and reduce human processing errors, our company Mage Montreal can do that. We’ve been helping e-commerce stores for over a decade. Here’s the catch, we’re specialized and only work on the Adobe Magento e-commerce platform. We do everything Magento, if you know someone who needs design, development, training, support, we got their back, email our support team, [email protected] or go to magemontreal.com.

Alright, Michael, better having you here today.

Michael: Yes, I’m very excited. This is a really cool topic that I can talk a lot about.

Guillaume: Yeah, for sure and this is your personal story as an entrepreneur, which you’ve lived. I’m always interested to hear somebody’s entrepreneurial journey.

Michael: Yeah, I’d like to believe there’s some lessons learned along the way.

Guillaume: I’m sure there are many, no doubt there. So, can you tell me just a little bit about the background? How all this started, this company software project?

Michael: Yeah, about 10 years ago, I was a freelancer and I was doing everything for anyone. I was building WordPress websites, I was doing SEO, I was doing local SEO and global SEO. But pay per click was something that really resonated with me and I think it’s because I grew up playing real time strategy games like Starcraft, where it’s like you versus the competition? How are you going to set things up and how are you going to use different mechanisms that you have available in a way that’s smarter and more efficient than your competition? So, to me, that’s how I describe pay per click. So, it was a really cool way of doing optimizations and seeing results in a relatively quick time cycle and it was really enjoyable.

So, I decided to double down on Pay Per Click for e-commerce, and started my first company, which was searchscientists.com that’s still around today. That was a Google ads, Facebook ads and Amazon ads service company. We were doing things manually for our e-commerce clients. And then in about 2016, 2017, I sort of really saw Pay Per Click advertising as becoming more efficient over time. I see software getting better over time. And I saw really advanced tools for Google ads and Facebook ads, but I really didn’t see what I thought was like the practitioner’s tool, the tool that was really made for day to day optimization, technical optimization on Amazon advertising. But that’s when I started Ad Badger, which is a software company that helps people get more done, get better results in less time on their Amazon ads. That is the high level overview of where we’re at. I’ve been full time at Ad Badger now since 2017 and it’s been really exciting.

Guillaume: It’s been a fun journey, and I’ve never found PPC management to be as fun as Starcraft but I’ve been a big fan of that game as well.

Michael: Well, I have to ask, are you Protoss, Terran or Zerg?

Guillaume: I’m a Protoss.

Michael: Oh, yes, of course, me too. No doubt.

Guillaume: I’ve played them all, but for competition it’s Protoss.

Michael: Nice.

Guillaume: The startup phase for any software company like this one is very challenging because you don’t have any revenue yet for that new venture and you need to build a product that is good enough that people will be willing to pay for. Of course, most people in the marketplace now are aware of books like The Lean Startup, which is great, it helps people, but knowing the theory and putting it in application are two different things. So how did you fund that new software? Was that through your previous company? Did you raise some money or how did you go about it?

Michael: I consider myself a strong bootstrapper. At the outset of thinking about starting a software company I was very hesitant to raise any money. I was like I don’t need it, I will fund it with the profit from the first company search scientist. That’s what we did at first. It’s a little bit of a dangerous game when you do that because I sort of did it open ended. I would just fuel the software company without an end date in mind from the first company. We did that for a couple of months, we did it again for a few more months. Then I sort of saw the inevitability that I could fund this software with some other revenue source for an indefinite amount of time. So I really wanted the software to be able to stand up on its own feet. Software development is very expensive so we did end up taking Ad Badger through a startup accelerator, where we got introduced to a lot of investors and we got some seed money from that as well.

So it’s a little bit of a hybrid approach. That was also important to me because you sometimes hear stories of founders raising too much money and giving out too much power too early to people who are not the founders. Well, that could be good because you can really get plugged into people who have a lot more experience than you in certain things and people who have a lot of guidance and will be able to provide a lot of support. But sometimes I hear horror stories after raising money where the founder loses control over the company. I think it’s rarer than is perceived to be for a lot of people that are hesitant to raise money. Most investors do genuinely want to help and lend their experience to you. So that’s a long way of saying that we sort of view ourselves as a hybrid, where we try to be lean and bootstrapped but at the same time, we did raise some money in the beginning.

Guillaume: When you say a seed round, was that a few 100,000, a million or two?

Michael: We raised less than 500,000, which in the world of CrunchBase, TechCrunch, is a very insignificant amount.

Guillaume: Right, but it is still the correct amount for the seed or pre-seed level. As you’re saying about how much control you give away, we hear of rules like 20% share for each time and by the time you get to the fifth round you have almost nothing left. So you try to avoid such horror stories like that of the Netscape guy who lost all control. But there are other well known ways now like the Google board and its founders where they can fire the board if they want to. So there’s always a way, if you are able to have enough strength and negotiation leverage to protect yourself as the owner. Let’s say you can just raise like 20%, and not lose control and then raise another 20%, but from other investors to diversify the voting power, and put some clause to stay in power. Then you could raise another 20 and you’ll have just 40 percent left by the time you’re at round C. Depending on how great the whole thing is, you could be a unicorn.

Michael: That’s right. One concern I had, and I’m sure you might feel the same, imagine if you were to raise money and then all of a sudden you start getting little nudges. What if you worked on Shopify as well? What if you worked on WooCommerce as well? You know that’s additional revenue there. I haven’t talked to you about your personal preference about Magento or not, but for me, I really wanted to go deep into Amazon advertising with this company, and I really wanted it to be like a lighthouse for Amazon marketers who were struggling with Amazon advertising. So I really wanted to go deep into just one platform and be upfront about that and be able to say I don’t want us to incorporate five other platforms into our software.

Guillaume: It would be a startup mistake anyway because then you have to interface or integrate with the API application programming interface of each of those softwares. And it’s five times more work and you will dilute your work effort, but with a startup you need to get the profit as soon as you can, or just viably breakeven as soon as you can. If you take on a project that’s too ambitious, it’s not going to work for a startup or a software.

Michael: My wife and I had one dog for a very long time and then we got a second dog but it wasn’t double the amount of work. It was probably like three point five times the amount of work because they would like help. They would create emergent problems with each other. It was a one plus one equals three. So I imagined, the more things any piece of software does it doesn’t get a little bit more challenging, it gets logarithmically more challenging.

Guillaume: Yeah, we’ve lived some of that experience building our own in house ERP, and it stayed an in house ERP instead of going to market because it’s too broad of an approach. We would need to raise VC to put that on the market. So it was a mix of personal funding, you had the chance of having another company funding it and you had some VC investment. And did you do further rounds of fundraising after, or the software was already self sufficient?

Michael: I’m proud to say self sufficient.

Guillaume: Which was a rat.

Michael: Yeah, thank you. It was something that was really important for me to be able to get it to stand on its own two feet. And this is a little insight into just, speaking of hearing stories out there. You hear stories about people raising a lot of money and increasing their burn rate a lot, because you’ve run in the negative for a very long time every single month for the sake of growth and development. So you do that for a very long time and then you’ll need to raise money again because you’re going to run out of money. So I didn’t want to be on that loop, not that there’s anything wrong with it. I think it could work for a lot of different people.

Guillaume: I think there’s a lot of things wrong with that loop, but it can work in a spectacular way if you make it work.

Michael: Yeah. I know people with a B2C product who can acquire 5,000-10,000 people a month using a B2C mobile app. So in situations like that it makes sense for them to raise a lot and go super fast, and raise again. It wouldn’t make sense for them to say we should only take in a 1000 B2C customers this month. Whereas in our case, which is sort of a niche B2B, we’re charging a few 100 bucks a month. I do think there’s a strong argument for going slow and being sure you have strong retention, being sure you have a really good product market fit. And going through a natural cycle of, you know, crawling, walking, running, flying, turning into a robot and soaring into outer space. So I think it’s okay to follow that progression for certain kinds of companies, and I think we were one of them.

Guillaume: Great. Let’s talk about that startup phase because it’s so critical. This is where most of those software projects die. How did you get your first customers?

Michael: Through sheer force pushing a boulder up a mountain. I jokingly say, ‘like pushing a boulder up a mountain’, it really can seem like that in the early stages. A lot of times with software in the early days, it really does feel like you have to beg and apologize 24/7. Because there are, especially with Amazon advertising, for anyone who’s used Amazon advertising or the Amazon interface, you’ll know that it’s strangely buggy even without plugging into their API. So there’s some very strange things that it does and we’ve learned a lot of lessons about how to work with Amazon’s API’s. We didn’t know those when we first started.

So in the early days, it really does help if you build something more than just software in the beginning. I mean if you can build a community around what it is that you’re doing, maybe you could build goodwill in the industry, those things really matter a lot. Because I think any new software will probably have their inevitable share of bugs. And if you’re able to, just remind people that you’re human, remind people that you’re here to work for them, remind people that you’re your own toughest critic when it comes to the tool. I think that goes a long way. So some of the things that we did in the early days, was really believing in content marketing and we still do to this day. So being able to talk to people and provide value, even outside of the tool, has been one of the things that I’m most proud of that we’ve done here at Ad Badger.

We have, as far as I can tell, the most popular Amazon advertising dedicated podcast. We were the first podcast launched that just focused on Amazon advertising, and I think that has been really helpful for us. We got so many messages about people learning Amazon advertising through our podcast, that completely warmed my heart. So being able to create a company that will reach more people than will ever sign up for your software, I think is challenging, great for branding, of course, and great for influence and name recognition in the community. And that is the way that we grew initially through content marketing and creating a community.

Guillaume: It was pretty cool. So it wasn’t like Facebook groups, LinkedIn groups, other platforms, Slack private channels, or?

Michael: You know I thought a lot about what is the most optimal channel. So we’ve tried Facebook a lot, we’ve tried a private community, we tried a Slack group at one point, we do have it on our list of tried LinkedIn groups. And I think the ultimate answer is we need a community manager to manage these. Because what will inevitably happen is, in the early days of our Facebook group we would ask, what platform should we grow our community on? And we’d have a little survey, should it be Facebook, or LinkedIn, or slack or private URL? When you ask on Facebook, everyone says Facebook, when you ask on LinkedIn, everyone says LinkedIn, when you ask on a private membership site, everyone says private membership site. So for whatever reason, that’s just the way that the internet works. Whatever platform people naturally prefer is what they want your platform to be on. So I still haven’t solved that. If someone can create some kind of tool that combines all of these so that people can be in a LinkedIn group, and in a Facebook group, but somehow have cross pollination with each other, that would be a phenomenal idea. That way, you can have your communities where they naturally fit, but they’re all communicating with each other. That would be insane and extra special.

Guillaume: So as long as there’s traction with the market and clearly something is happening and people care about that topic as well. That’s a great way of getting your first few customers to engage, give you feedback and be tolerant of the bugs of the new software, while trying to grow something.

Michael: Yeah. And I don’t think that ever goes away. If you stay in sort of startup mode and just stay a constant student, I think that’s a good way to approach software, because we’re always releasing new things. You know for Amazon, advertising doesn’t stand still and for our software, we’re always releasing something new, like every month.

Guillaume: Talking about startup tactics like that for software, did you give temporary discounts like half price for three months, and later become full price because you’re in startup mode?

Michael: Yes, definitely. We did a lot of launch pricing for a very long time, where we basically told people while we’re in beta, our eventual price will be something like $100 a month. But if you sign up while we’re in beta, it’ll be $50 a month. I’m a big fan of grandfathering people, so if someone takes a bet on you early, you reward them with that prize. Increase it for future customers but let those early customers feel like they’ve made it and they’re in the club. So we did that in the early days for sure.

Guillaume: Another challenge other than obviously acquiring your first new customers, which is one of the biggest challenges, another very big challenge is the infrastructure on which this is built and the ability to roll continuous updates. How did you manage that, because this is another pitfall? Let’s say you wanted to outsource not too expensive labor, who does not have proper architectural knowledge for IT. They might not be able to push your changes without erasing your users preference in the database and stuff like

that.

Michael: Yeah. I think it depends. Ad Badger is a software company, we played around with development agencies to build things for us. But actually, we found the most traction in the easiest cycle of development with our in-house team. So actually, hiring developers is a skill that anyone who wants to start a software company should really understand. How do you manage developers? I think they get managed differently than any other employee. In fact, the lessons that you learn through things like agile development, Scrum and things like that could really be applied to everybody else. I learned some really good principles with how to manage tasks, how to manage a project, what a reasonable load looks like, how to reflect and run a retrospective.

I Learned all those things probably too late into Ad Badger’s lifecycle. But once they started to connect, it really increased our velocity and increased our code quality and just increased everything, from the vision in my head, giving it to our CTO, and getting it built. This really improved once I learned how to talk and communicate with developers. Especially with what we do, it’s so technical and constantly changing, and I think it would be difficult for us to outsource any of that. So if someone wants to start a software company, I believe that should be on an in-house scale. Whereas if you are not a software company and you want to have an internal software, maybe something like that could be specked out and given to an agency much easily.

Guillaume: So you did hire actual full time developers in house to build that software? Which is another major expense.

Michael: Time to raise money.

Guillaume: Yeah, exactly. Because each time you try to develop anything, you’re just like, that’s the 20th hour, that’s the 100th hour. And I think it just piles up for each of your features, and the bugging, and so on. So you’re saying while raising money to be able to fund that you also learn a job in Scrum? There’s an awesome book about this. You can go through your scrum master certification on how to do it. You can learn how to communicate better with programmers. Anything else that you’ve discovered on how to motivate programmers and get excellent work from them?

Michael: If someone has never hired or managed developers, get a consultant. Talk to someone who has managed developers. What really changed the game for us was we hired a temporary engineering manager, who basically worked at a larger company, where they managed engineers. They really gave us the codecs to really understand. That is a massive shortcut. You pay somebody for a month’s worth of consultation to get into shape, read their professional product owner, because that’s essentially what you are. A lot of times you think about, I want to build software and it’s just going to be a developers’ built software. Actually no, you want a front end person, you want a UX UI person, you need a back end person, you need a front end person, you need a project manager, you need an engineer. A product owner who’s thinking of features. You need someone to collect user feedback, building software.

There are so many roles that go beyond just a developer. You need to understand that if you just hire a developer that all of those other roles are going to need to fall on you. You should definitely need to be aware of what an org chart looks like when a software gets built. All those things that I mentioned need to happen, because what are developers really good at? They’re really good at building codes and building apps and doing the things, like building the widget. But then when you think about all the other components of how a widget gets built, they need a lot of support to really shine. So I’m always asking what our team members need in order to do their best work. And I feel like that question doesn’t always get asked for developers. So what do developers need to do their best work? This is a question I thought about a lot.

Guillaume: A lot of people believe a developer can sort of do it all. Like, it’s just web development. No, there’s a lot of specific trades in this. It’s like hiring a plumber thinking you can build a whole house. There are a lot of trades and skill sets needed to build the whole house. Yes, there are guys who can build a whole house alone, they have all the knowledge for that. But it is still not what to expect when you hire each person who has a specialty. You need to have the full skill set in your team to have great results. So I agree with you there. So that engineering manager probably also laid out the architecture of your software from a technical point of view. For that thing to be scalable, you will need to roll out updates to avoid having problems because the guy just started coding without thinking in the future.

Michael: Right.

Guillaume: It makes a big difference. Do you have any specific tools or workflow that you’d like to share about that continuous deployment, you know, once you have SaaS software?

Michael: Yes. From my perspective, I think the thing that we follow the most is, from the book that you mentioned, ‘The lean startup,’ feature creep is real and what that means is when I sit down and I think of a new feature, I think of how amazing it could be, I think of just one of them because we have an entire suite of tools. And when I’m thinking of a new tool, I can think of something that will take a year to develop. Then I sort of think, well, what’s better? Waiting a year for everything and then maybe even being off the market, and maybe all these thoughts in my head aren’t actually what users want. Maybe the interface would be weird or something like that and be off the mark. So imagine waiting a year to build something and then being completely off the mark, and then taking six more months to fix it. Whereas if you think of that ultimate tool, you can actually think of it like, what is the core piece of value that we’re actually providing people with this tool? And then how do we get just that out of it as quickly as possible so that it can form the next stage of that tool? So instead of something that will take 12 months, what can we build in one month, that’s going to provide that value in a way that will be impactful and people will like it. That way, we can then bolt on to it next month and the following month, and really improve it that way. As opposed to doing it in private without feedback for 12 months. You can release something smaller, still good, still gets the job done. You get feedback to inform the next month and the next month. That is something that we followed a lot, that’s something that I really do think about when it comes to our development cycle.

Guillaume: That is so true. Go with the phased approach or have sprints, or whatever, take two weeks, three weeks, four weeks, and release functional software each time, and make it better and better all the time. I totally agree with that. Anything else that comes to mind that people trying to start a software company or SaaS software company should think about?

Michael: Don’t be afraid of feedback. It can be really difficult. But getting feedback before people cancel is massive. You know, what are we missing? How could we have improved this experience for you? Trying to get feedback as much as possible is massive. And it’s something that you hear a lot with software. People assume that you should be getting feedback but I feel with software, it’s so pivotal. You actually sitting down with people and watching them through your software can help you learn so much. So don’t be afraid to really get some of that critical feedback. Watch people not know where to click. That really sucks to hear, but if somebody’s leaving you for a competitor’s tool, that really sucks to hear too. But anyone who’s doing this should really want that journey and it would really be nice to hear that. So you really do become your own software’s toughest critic, which is sort of a strange position to be in. But if you’re your own software’s toughest critic, you always have something new to improve, you always have something new to work on. And that energy is such an incredible force at improving your tool.

Guillaume: And when you created this new tool, obviously you were already a paperclip expert, an Amazon expert, you were passionate about the topic, then you decided to create that software. Was there any specific vision or philosophy behind this on how to make all your guiding decisions as to what the software should be?

Michael: Yeah, it was really about what are the things that we do manually and then how do we create something that helps you do it a lot faster? How do we take it from having to log into Amazon, make five clicks, download a spreadsheet, now you’re in excel. You will need to do 10 more clicks to filter the data that you want, and then it’s another 25 clicks in order to paste it in a format that you can upload back to Amazon, or clicking into Amazon opening up five campaigns and digging through them. How do we make it such that all the campaigns are already dug through? We’re presenting you with the most relevant piece of information with some options where you can get something done in a matter of two clicks. Login and approve or disapprove some changes, or login and pause some of your worst keywords or just make some of those decisions for you on the back end. That’s really the guiding philosophy behind the Ad Badger.

How do we make it so that we only make good decisions? So when you’re spending too much time on a keyword, we take care of it. When there’s not enough spend on the keyword we take care of it. How do we make only good decisions? When a keyword gets thirty clicks without an order, automatically pause it. That is how we think about our software development for people.

Guillaume: As a founder, you’ve been in this for at least four years now, what’s your role at this point, even though you’re still very much involved in the day to day work on the software and marketing?

Michael: A little bit of everything. I still love the product development cycle, meaning I love thinking of new things to improve our tool. So being a PPC person, I still have so many ideas for that. Product owner is one of the titles I have. Another position that we are looking for is a chief marketing officer, so we don’t have one right now. So in marketing I love podcasting, I love talking about Amazon advertising. So I do that on a weekly basis as well.

Guillaume: All right, any last author anecdote that you’d like to share from your journey of creating the software?

Michael: I heard some quote before, and it’s said that every company is a software company, but only a certain percentage of companies know this. So even e-commerce companies that are building products should still know and understand software. Even if they’re not building software and selling software, they should still know what tools are available to help them get the most work done and as most efficiently as possible. So working with teams that develop software, working with software itself, or maybe even developing their own software, these are things that I think any company needs to get really good at. Understanding what tools are available and how to get more work done more efficiently. I think those are the keys that every company needs to have.

Guillaume: Awesome. If people want to get in touch with you, Michael, where can they find you?

Michael: Yeah, check us out at AdBadger.com. If you are interested in Amazon advertising and you want more info on that, I podcast weekly at the PPC Den podcast. You can search that and you can hit me up on LinkedIn, Michael Erickson Facchin.

Guillaume: Alright, Michael, thank you for being here today.

Michael: Awesome. It’s been a pleasure.