This episode I’m talking to Dan McGaw, former Head of Marketing at Kissmetrics behavioral analytics solution company. Dan is now the CMO of Effin Amazing, an analytics and growth consultancy company. Dan is an analytics expert and has consulted for companies such as Wistia, Contactually, CrowdRise, and many more.
If you are overwhelmed by the amount of data you need to track and don’t know where to start, then this episode is for you. Listen in as Dan walks us through a step-by-step actionable methodology on how to track the right metrics and KPIs for new and existing businesses.
Listen to this Episode:
Topics Discussed in this Episode:
- Determining the metrics and tools useful to your company
- Key KPIs to rely on
- Building an analytics and integration specification guide
- Script planning and auditing
- Tips and resources for learning analytics
- Entreprenuership and starting a business
- Customer development and validation
- Recommended resources
- Dan McGaw on LinkedIn
- Effin Amazing blog
- [email protected]
- Tools: Kissmetrics, Amplitude, Segment, Google Analytics, Mixpanel, Hotjar, Heap
- Effin Amazing Webinar: How to Integrate and Extract Insights from Analytics
- Udemy, CXL Institute
- The Lean Startup by Eric Ries, SteveBlank.com, You have a business idea, but what’s next by Dan McGaw, Growth Hacker TV
- Effin Amazing Online UTM Tracking Code Builder
- InVision app
Louis: If you check out Dan’s profile that we’ll link in the show notes, you have to click maybe 10 times on show more to see the actual full list of experience. It’s pretty cool. We’re not going to talk about LinkedIn today, we’re going to talk about Analytics and in particular, how to actually implement it the right way in your business, whether you have a big business or small one, we should find a way to implement a step by step methodology that people can use in their business.
I don’t know if it’s me, I don’t know if I’m hanging out with the wrong crowd, but I’ve never been able to work in a business with a business or been in touch with a business that had a good process to measure things, to measure performance online from start to finish. Am I the only one or is it true?
Dan: I definitely think it’s a struggle for most companies, it’s really, really hard to get your mindset into tracking everything and it’s a practice that most people don’t really follow. Marketing is full of creative people that are excited to have ideas and make change in the world through those ideas. The problem is, a lot of marketers aren’t necessarily the most technical people in the world, and they don’t think about the world and basically events or properties. We luckily have worked with companies that are fairly data driven and we’ve taken to a whole new level, and we’ve also worked with companies that had no data at all and then we make them extremely data driven and now they rely on that data heavily to do everything.
You’re not alone. I would say 98% of businesses had no idea what the data says.
Louis: You said this word and I’m not sure that many people really get this word, ‘data driven.’ I’m pretty sure that if you ask companies to say, “Are you a data driven organization?” 98% will say, “Yes, absolutely we are.” Obviously, it’s a rough number that you’re saying here, but it’s true that the vast majority are not really data driven.
Hopefully in this episode we’ll be able to go and maybe help them to get one step closer to this actual data driven organization. What are the biggest problems here? To me, companies and marketers in particular are absolutely overwhelmed by the amount of data out there but that’s a fact, Analytics, HotJar, heat mapping, surveys, there are so many qualitative, quantitative data out there, it’s quite overwhelming. Let’s go through this process together. What is the process to implement analytics so you can focus on the right data and you don’t get shitty data out of it, so you can have data you can trust basically. Let’s go step by step. You go into business, what do you do first and foremost?
Dan: The first thing you have to do as a business is you want to understand what are your primary goals and what are your objectives. Each business’s goals and objectives are entirely different and that’s going to help you decide on which analytics tool you need to use or what architecture you need to set up that tool. For an example, if you’re a mobile gaming company, you probably don’t want to use a tool like Kissmetrics, it just doesn’t fit that data architecture. But if you are a mobile gaming company, you may want to use a tool like Flurry or Mixpanel or Amplitude which are more setup for that type of marketplace. Meanwhile, on the flip side of that, if you are an ecommerce company, Kissmetrics can be extremely helpful but at the same token, Mixpanel is not going to be as effective for an ecommerce company.
You really do need to focus on first like what is my problem? What is my business use case? And then figure out which tool is going to be best for you. Obviously, it’s not easy to basically create that and come up with that, we actually are rolling out an analytics quiz tool which you could take a quick five minute quiz for your business and it will tell you which analytics tool you should use. Because everybody’s going to tell you, “You should use my tool.” Whether it’s good for your business or not because their sales rep just wants to get their deal flow. But the first thing is making sure that you have the right tool for the right job. I would say that’s definitely step one.
Louis: Okay, right. Let’s talk about SaaS in particular and software as a service businesses or businesses selling online that are related to software or digital products, what are the typical tools that you see companies using that fits this particular use case?
Dan: For digital companies or what kind of digital companies? Just to make sure.
Louis: Let’s start with software as a service companies.
Dan: Okay, great question. For a software as a service company, if your objective is to start tracking marketing activities, obviously you need to have Google Analytics. Google Analytics should be installed in everything no matter what. Kissmetrics, from a marketing organization perspective is one of my favorite tools for marketers to use and be able to build out all their tracking. Kissmetrics is really, really flexible in the fact that it was built so a marketer could actually install events, and can install properties without having a bug and developer all the time. Kissmetrics is a great tool for that.
If you’re a SaaS company and you’re looking for product analytics and you’re trying to figure out product stuff, Kissmetrics still does a good job but there’s tools like Amplitude which may provide you another step of granularity and another way of tracking because they have a different way of thinking about the data. It’s very rare that you’re only using one tool, so this is where a tool like Kissmetrics and Amplitude combine together, you’re using both. Probably integrated using Segment which segment.com is a really good data piping tool. If you use Segment, you can integrate both tools with only writing one bit of code, but that would a be a stereotypical setup. But that’s not everybody.
Some people like Mixpanel’s visualization so they choose Mixpanel. One caveat just so everybody understands, the reason why I don’t recommend Mixpanel for most SaaS companies, is due to the fact that they’re not good at multi device. For an example, Kissmetrics and Amplitude can both merge identities across multiple devices to create one profile. Mixpanel can only have one identity at a time, and literally, if you start out on mobile and then you sign up on mobile and then come to a desktop, and then you play around on the desktop then log in, everything you did on that desktop before your first login just gets deleted. Their multidevice attribution isn’t as good as the other ones. Then in SaaS worlds, a lot of your stuff is across multiple devices.
Louis: Do you know what, I’m actually quite surprised that you’re starting this step by step by mentioning tools. In my head I would say, “Maybe this is one of the latest thing you need to do before you actually know what you’re trying to track.” That’s actually an interesting take.
Dan: Just to say, I definitely think you will need to focus on what are your business objectives, which are what are you trying to track and then picking a tool around that. The reason why I say that is because a lot of companies, if you start out with the building your analytics spec, like all the things you need to track, that’s part of planning, and definitely things like that. Each one of these tools track things differently, if you build out a crazy analytics specification guide, the way Kissmetrics tracks compared to the way Amplitude tracks are completely different. You’re going to wind up having to go through that process twice. I recommend to think a lot about what your objective is with this tool. What are you trying to track? What are the KPIs? What are the things you need to measure from a business perspective, because that tool is going to be very, very specific on how you need to set up its tracking.
Louis: Okay, that make sense. Now let’s say I’m not a marketer, I don’t have a big business, I have a project that is small enough or website that is selling ebooks, something smaller. What do you recommend to those people who have side projects that are not necessarily big company yet, or they want to start a startup, what’s the typical process there and does it differ from what we just said around SaaS companies?
Dan: Definitely, for sure. If you’re a smaller company, budget is obviously the biggest thing you have to focus on. You’re going to want to use a tool that doesn’t cost a lot of money, that would immediately basically put Kissmetrics out of the way because Kissmetrics is a minimum of $500 a month, they no longer offer any other plans.
This is where tools like Mixpanel come in, they do have a free trial, Amplitude offers up to 10 million free events. But at the same time, those tools may be too complicated. As a smaller business, you’re probably going to be looking more your Google Analytics because it’s free. You could still get a crap ton of value out of it and then other tools like HotJar which are a little bit more cost effective, they enable you to build a funnel, you don’t need tons of data when you’re doing a simple ebook business, you just need to understand your funnel. HotJar does that for $20 a month and then you also have Google Analytics which can do it for free. You want to make sure that you’ve chosen a cost effective tool in that use case.
Louis: Just to be honest with my listeners here and to make sure that I’m very transparent, you don’t know that, but I do work for HotJar full time as well outside of the podcast, so I know you didn’t know, which is really honest of you to mention HotJar. Hotjar also has a free plan, by the way, not only a paid plan, and you can do quite a lot with it.
I like that, when you’re a smaller business and you’re starting out, what you really care about is understanding your customers and understanding your funnel, you shouldn’t really have to go further than that unless you’re very analytical driven yourself and this is something that you enjoy.
Step one would really understand your KPIs, understand what you want to measure. What are, from your experience, the key numbers, the key KPIs that companies usually rely on, mostly?
Dan: Absolutely. If you’re a stereotypical SaaS product or any kind of product, the first metric that everybody’s looking at is probably the number of people who are signing up to your program. You can definitely look at this from, “Hey, how many leads do we collect? How many emails do we get? Then how many of these people actually sign up for our product or actually purchase our product?” Those are going to be three metrics right there, the number of prospects or leads, the number of people that sign up, and then as well as the number of people who purchase. Those are really simple, simple metrics. I don’t really care as much about traffic or pages per session or session links, if I’m a media company I may, but as a SaaS product, not that big of a deal. As a SaaS product, the next one you really have to be focused on is going to be MRR, which is monthly recurring revenue, and then churn. Those two things are going to be massive for you.
Lifetime value is obviously a derivative of those two things. You understand your monthly recurring revenue, you understand how much of that you churn on a monthly basis, you can build a lifetime metric value out of that. You don’t understand your lifetime value, you have no idea how to build advertising, you don’t know how much you can spend to acquire customer. I’ve been saying MRR, churn, lifetime value, all about equal on the metrics that you have. But between those six metrics, as a small business, that’s really all that you have to focus on in the beginning. Obviously you can look at your total cancellations and stuff like that but you have to do that to get your churn number anyways. I think those were all really, really good points.
Louis: Aside from SaaS and software companies, out of your client base or from your own experience, what would be the other type of companies that would really suffer from the problems you’re solving nowadays with analytics?
Dan: Yeah, absolutely. I’ll give you another use case. We work with a client called forksoverknives.com, they made a really, really famous documentary back in 2011, if you haven’t seen it, jump on Netflix and check it out. It’s been the number one documentary on Netflix for years, really, really good. They’re a media company, they have as subscription service for a meal planning tool, they also have an online cooking school, they also have a mobile app. They have a lot of other problems they have to solve inside of those products.
But as a media company that drives traffic and of course puts that traffic into a product, a big focus of ours is figuring out how long do people spend time on this site, how many articles they read in their first session, how many articles they need to read before they actually give us their email? Email’s a key driver of that business because we have so many different products. For us, when we run an AB test and we’re trying to validate metrics, a lot of that is more focused on how many sessions do they have? How many pages they view in that session? How do those metrics change? Because we can directly correlate the number of pages per session affects our revenue dramatically. If we can get somebody hooked in our content, and coming back on a weekly, daily or even monthly basis, we’re able to get more opportunities to push that person to one of our many products. That’s another place where it’s a little bit different, where you do actually track number of pages visited per session or session links or bounce rate.
Louis: That’s interesting. Step one, you pick your KPI. Step two, you will actually look at the tools available out there and see if they actually can track what you want to track and if they fit your budget.
Dan: I think that’s a great way of looking at it. Instead of earlier, I said step one is obviously picking a tool kind of, but you’re correct. The first step is figuring out what am I trying to tack? What are those KPIs I need to get done? And then the second step would pick a tool that able to do that.
For example, there’s a tool called Heap. Heap is an amazing tool you drop on your website and it just tracks everything. Heap does a great job at just tracking everything. Where it really drops the ball, it doesn’t have a lot of good visualizations. It doesn’t give you as many visualizations or ways to splice that data as maybe Mixpanel would. This is what you really have to understand, can the visualization can give me the answers that I need? Is this the tool that’s easy for me and my team to understand?
Louis: Right. Step three, typically when you start with a new client, or even for smaller companies you don’t necessarily work for, what is the usual step three after this?
Dan: Great question. That’s going to be building our analytics and integration specification guide, it’s called an analytic spec. Every company like Mixpanel, Amplitude, Kissmetrics, all have their own version of what this analytics spec guide has. We even have our own proprietary one internally here that we use across our client.
What you need to do is you actually need to go through your site as if you were a customer and then document all the things that your customer will do. I don’t think tracking button clicks are going to be the thing that you want to do. You’re mainly tracking does somebody make it to a page in the funnel? Do they go to a phrasing page? Which product or which plan do they click on the pricing page? You want to track things like signed up and then obviously you want to track purchase, you want to track those key events by coming up with an analytics specification guide.
This is going to track the events that you send, the properties that are sent along with an event. An event is simply an action and then the property is the adjective of that action. If I jumped, that would be the event, I jumped. Jump would be the action. The properties would then say how high and then high would be 2ft., the 2ft would be a [00:15:26]. It’s a verb with an adjective, and then of course you want to then start figuring out what do we want to know about our customers and adding that along the way. In the signup form if they say what country they’re from, that would also be a property that we want to track not only to the event, but also something we want to store with that user’s identity. Doing the planning of what the integration is your third step.
Louis: Right. You would actually advice for markets in charge of this or whoever else is in charge of this to actually go through the experience herself, right?
Dan: Absolutely. It’s definitely a complicated process and if you haven’t done it before, go to effinamazing.com, go to our resources section and there’s webinar that I have recorded there which teach you how to build a specifications guide and even give you a template to use. It’s not easy to do this, especially as a marketer because marketers don’t naturally think like engineers. What we have to do is the specification guide is simply made so we can hand it to engineering to get integrated. We have to make sure that we go through a process which is good for not only the marketer but also the engineer as well because there are other ones who typically are going to do the integration and if the engineer doesn’t like you or if you don’t make this correctly, the engineer may never do it.
Louis: You want to speak the same language between engineering and marketing, right?
Dan: Absolutely. The webinar that it created, it’s how to create insights and action out of analytics, really breaks down how do you bring this to that engineer, what is the process that you need to do to get it to the right person.
Louis: Okay. Let’s say we have that down, it’s well documented and our engineers are happy with it, what’s the next step?
Dan: You then have to do script planning. You do not want to do all of your integration at one time. It needs to be spread out over 6 or a 12 month period. If you started out doing everything all at once, you wind up with a lot of bad data and a lot of stuff you just can’t trust. We typically break our analytics integrations into a one week sprint per month. For an example, in the first sprint we would choose our 6-10 critical events. Critical events are going to be more like they signed up, they purchased, they got billed, they cancelled, things that are critical to understand your funnel. That first sprint, that’s all the developers build, they build that in a staging environment, we then go in and audit it to make sure everything was correct, because if you even have one event with a capital and another one without a capital, that’s going to be treated as two separate events in analytics tools.
Everything has to be 100% on point then push the production, start building reports, getting value out of that tool. That’s really all you want to do in that first month, is just use that data you have and don’t try to go crazy. If you create 300 events, you’re just creating noise, and it’s not healthy.
Louis: I’m glad you said that. I wasn’t expecting you to say it necessarily but it’s the right thing from my experience that’s definitely what happened many times. I’m curious about one thing. Let’s say you come in into a business or some of our listeners might already have data analytics place and they don’t really like it, it’s too much, they don’t really know how to make sense of it, they don’t really know if they can trust the data, how would you recommend our listeners to deal with this situation where they already have something in place. Do they start from scratch with no report for three months until it’s built? How do you advise people to do this?
Dan: They have data coming in but they don’t trust it and then it’s breaking, is what you’re saying?
Louis: Yeah, and there are too many of it, too much data, they don’t really know where to start. They want to start from something that is cleaner, I would say.
Dan: Absolutely. That’s definitely hard. Try to clean up something that’s already in place. In a lot of these different analytics tools, you can hide events and that’s actually what we’ve done to help companies get started is next thing you know, there’s a client where we start next week, they have some 300 different events which are already tracking. They track them with every single click on their website which is just noise. The first step that we’re going in is doing a list of all the different events that they have and then we’re going to be shutting off all of the events which are not needed right now to be visible in the UI. A lot of these tools will allow you to actually hide events so that we don’t see it in a reporting layer and that’s really what you want to do because you want to get rid of all that noise so you can focus on what those key events are. Those events will still be collected in most cases, you just don’t want to see them in your reporting layer, so that way you can stay focused
Louis: I would actually say that this step probably needs to appear before, maybe in step two or in step three, once you know what measurements you want and once you’ve done an audit and you know what stuff you’re actually measuring that are useless, this is probably the step where you actually shut it off before even moving to anything, right?
Dan: Absolutely, yeah. I would definitely say that that’s part of step three, figuring out what’s in your analytics specification guide, if you have a good analytics spec, you can see that I have all these crappy events that they don’t need to try, you can just turn those things off. One thing I will make sure you understand is that if you do want to have an event removed, it does need to be removed from the code unless you’re using Segment or we’ll get your pricing or whatever tool. Segment is one of the only tools that enables you to disable an event from even being accepted.
Louis: Right. I do know what Segment is, I haven’t implemented it on the business I was involved in but I do know what it is. Let’s explain to the listeners who don’t necessarily know what it is, in simple terms, what does it do?
One example when a user signs up, to make it work across all 15 of those tools, I have to write 15 different lines of code that send that syntax into those individual tools. What happened was it’s about seven years ago, Segment thought this was a really, really bad idea, maybe a little longer than that but they figured that basically what we should be able to do is just write analytics.track, we send that to Segment and then Segment basically becomes the Rosetta Stone or the translation tool that then converts it into the other tool syntax.
Louis: Would you actually use Segment with most of your clients?
Dan: We actually require Segment to be used on more of our clients. It’s very rare that we don’t require segment. The only times that we don’t really require Segment is when it’s a company that’s not running a heavy marketing stack. For an example company who’s just running two or three different tools, it’s not really a big deal for us to use Segment. Meanwhile on a lot of our clients, they’re running a 15, 20, 30 differ tools, what we’re doing on those clients is requiring them to use Segment because it reduces the amount of money they actually have to pay us. What happens is if I have to write 15 different lines of code for your tools, that’s going to cost you a lot of money. If I only have to write one line of code, that’s going to cost you about 15 times less. If I had to write once for Segment, and it works for 15 tools, I just saved you a ton of money even though you may have to pay Segment $2,000 a month, having to integrate 15 tools and maintain it, it’s going to cost $20,000 a month.
That’s really where the cost analysis and cost benefit comes from, is that if you look at it from a developer cost perspective, $150,000 a year and you pay one developer to maintain your stack for a year, it’s $15,000. If you have them setup Segment and it cost you $2,000 a month through Segment, that person only spends maybe two days a month, that person go focus on other things. Now they’re only spending $10,000-$20,000 a year on a developer, now you’re spending maybe another $20,000 on Segment if you’re a big company, that’s a total cost of $40,000 compared to $150,000. That’s nearly a 3X multiple if you want to go the old way.
Louis: Really nice. That’s an interesting side topic. Thank you for explaining it that well, you clearly know your shit. It’s interesting. Now that we have a step four, now we have the engineers who are setting up things and they are doing it using a giant methodology. They do a one week sprint every month to implement metrics as you go, you test them, you create reports, what is the typical next step then?
Dan: Once you started getting all the integration done, auditing could be considered part of step three, of getting that step done. Whenever the engineer has got any of these done, you need to make sure that you audit it in a staging environment and make sure that the data is clean. That is the biggest place that people go wrong, is they just trust engineering to do the integration and then they high five and move on.
The problem is that you have multiple people working on these tools. If one bit of code is written wrong, it could cause havoc in all of your stuff. I just want to stress, you definitely want to start doing the auditing.
To go to the next step, past that auditing, like you had talked about building reports, you’re obviously going to want to create reports for you KPIs, that’s the first place to start. Start very, very simple, do not create reports for everything, make sure that you’re just looking at your phone, you’re creating your initial dashboard to get your KPIs.
That’s really going to be the next step, trying to get value and create insights out of the new tools that you have. That’s probably where some of the biggest struggle happen. Formally being the head of marketing at Kissmetrics, I obviously have an insider’s view of what it’s like to be at one of these companies. A problem that all analytics companies struggle with is getting people to use the reports and to derive insights out of those reports. That’s a huge area that companies need to focus on and it’s really hard because only 2% of all marketers, and this is a fact that was put out by the CMO survey. It’s actually 3%. Only 3% of companies feel that they can actually connect marketing practice back to analytics based upon the talent that they have. That means 97% of companies are unable to connect marketing practice back to analytics due to lack of talent.
As a company, you really have to self educate yourself on that analytics side of the business and connect that to marketing practice, because if you can’t you’re just not data driven, you can lie to yourself and say you are, but you’re not. You have to be data driven.
Louis: What would you recommend those markets to learn and where should they learn?
Dan: That’s a great question. There’s definitely a lot of places online where you can learn. Google Analytics offers free courses. You can go to platforms like Udemy, which has a ton of courses. There’s a Mixpanel course offered by my buddy Rubin, which I think is on Udemy for free. There’s other courses that people about Amplitude and Kissmetrics across Udemy and things like that. I really do think Udemy does a good job at those. But there’s a lot of these class platforms out there that you can check out. I’ll be doing a course for the CXL Institute after the New Year about how do you setup the entire marketing stack, not just the analytics tool, which will walk you through how do you do your analytics spec all the way through how do you integrate your marketing automation tool. But there’s plenty of resources online.
I will say this: where I learned was I sat down and I read Kissmetrics and Mixpanel’s documentation from front to back. This was years ago. We’re talking seven years ago now, that I sat down, knew that I had to do this for my company to get data driven and just read all their documentation and then played with the tool until I figured it out and now I’m one of the best in the industry because I took the time to understand that documentation.
Louis: I think that’s a great next step. Read documentation, create stuff, try to dig into the data yourself, get inside. It’s not going to take one day to get into the level where Dan is for example. It’s going to take months or even years but you have to start, you have to get in yourself. I think you can learn so much from Udemy unless you actually implement it in your business and try it and read the reports, it’s not going to work out, right?
Dan: People learn by doing. It was our tag line at Code School where I was a big proponent of analytics. Definitely test it out on anything that you can, get it setup, spin it up, play with it, break it, see how to break things. You’re going to make mistakes.
We find booby traps all the time still in analytics tools. Documentation is typically wrong, something isn’t conveyed right or explained in a way that you might understand it. Understand that you’re going to make mistakes no matter what and that’s okay, that’s part of it. And if your boss gets in your case, you’re making mistakes, tell him to shut up and learn it themselves.
Louis: That’s a great way to end this step by step methodology together. That’s step number six, tell your boss to shut it.
Just a little bit more about you because as I mentioned at the start, from the LinkedIn, you do have a lot of experience even though you’re adding a lot of consulting jobs into LinkedIn. You’re the CMO of Effin Amazing, which is an Analytics and Growth Consultancy, you’ve been a consultant for many companies like Wistia, Contactually, CrowdRise, before that, you founded a startup that was the Yelp for gas stations. You were the head of marketing for Kissmetrics, the VP of growth for Code School. How many companies did you actually create yourself?
Dan: Oh, I have no idea, no clue, that’s a great question. I started my first company when I was 14. That was one of the first online booking agencies for DJs, I actually represented 160 different DJs around the world. I ran that company for about 5 years, when I was 19 I sold that company off. And then after that I lived low for two years and just had some fun. But since then I’ve probably started at least 10 or 15 companies, some have been good, some have failed, some have been revenue and profitable and I’ve move on to other ventures.
Right now, Effin Amazing is a company I own that owns two other products, we’re starting a clothing line right now. I’m just an entrepreneur, I don’t know anything else better to do with my time other than to start businesses.
I have to say in my lifetime I probably started 20-30 different business ideas. I would probably say 15 of those have made it to the point where they’ve been legally registered. The clothing line that we’re starting here at Effin Amazing, which is called Amazing Corpse, that company is not even a real company, it’s a domain, it’s going to be a Shopify site, we’re going to funnel the money through the consulting company. At that point, it’s tactically not another company, it’s just another brand.
Louis: Holy shit. That’s what comes to me when I hear you. That’s crazy. Let’s dig into this. Why are you so driven? What’s going on? 14 year old and starting a business, where do you think it’s coming from?
Dan: Obviously, I think some of it is DNA. My father is very driven as well and very tenacious. I had to say a lot of it also having the fact that I grew up very unfortunate, the fact that my mom was not very successful, I grew up in the ghetto, my grandparents though on the flip side were very successful. Even though I was a kid that was on food stamps and lived in the ghetto, my grandparents made sure I went to one of the nicest private schools in all of Pittsburgh.
That contrast of being in a poor community where everybody is struggling in the ghetto to then go to school every single day with the richest kids in that city was a very interesting contrast and it very much made it so I understood that there’s rich people and there’s poor people and if I’m going to choose one side to be on, obviously being on the rich side maybe the place that I want to be. That just instilled a little bit of passion in me to want to become better than what I was already in, just provided a great education. But at the same time I think my hustle probably just comes from my DNA more than anything.
Louis: From all of those experiences and those 30 companies you started, the ideas you started, forgetting about this clothing line you are launching, but let’s say you want to establish a new business right now, I’m giving you that as a task, you’re going to start a new business right now with me and we’re going to find an idea together right now. You can’t use your name, by the way, you can’t use your network or your profile, the only thing you can learn is the experience you had and the mistakes you made. How would you start it again? What will be the first steps you would go through?
Dan: In the first step you have to do, we do this with all of the companies that we start, we do customer validation. We follow the lean startup model or the Steve Blank Model where if we have an idea, we go out, we validate it. We do customer development interviews, we ask people probing question which are not steering them to the right answer, we’re trying to understand does this product or does this service actually solve a problem. Sometimes it doesn’t, sometimes it does. Hence why we’ve given up on certain companies. But I would first always start with customer development first to make sure we’re solving a real problem.
Louis: Do you have any resources to recommend regarding customer development?
Dan: Yeah, definitely. One, read the book Lean Startup, Google Steve Blank and read all of his blog posts, really, really good. I think there’s one like Bootstrap Academy. Then if you look it up on YouTube, there’s a video. I did a presentation in front of [00:33:44] about how we do customer development and our process. It’s the same thing that Steve Blank and them do, we just don’t use the canvas. Lean Startup and then there’s another one, I can’t remember the guy’s name, Travis. It has like leanstartupboard.com.
Definitely you want to do customer development, interviewing customers, don’t tell them what problem you’re trying to solve, don’t tell them the solution, don’t even tell them you have a product, just ask them questions around the problem. We created the Effin Amazing UTM Builder when we did our customer development on that. We interviewed marketers and we said, “How do you create UTMs? Why do create UTMs? How many on your people team create UTMs? What’s the problem about how you create UTMs with you team?” We were able to find out people must have a problem syncing pre-sets amongst team members and we created a tool to solve that problem and now we have 7000 users to use it on nearly a daily basis. You have to make sure you’re solving a real problem.
Louis: That’s interesting. You would actually frame your customer development interviews within a certain topic. Here you knew, you had a hunch probably UTM tags around the best solution available to track traffic from one source to another, you frame those questions in this particular topic, right?
Dan: You want to make sure that obviously you’re asking questions around whatever the topic is that you think that you have a solution for. You just don’t want to give them the solution. Iif I came to you and said, “Hey, listen. I got the best ice cream scooper in the world.” You’d be like, “Cool. That’s awesome.” But that doesn’t tell me whether you actually have a problem scooping your ice cream. If you came to me like, “Hey, 3999 for the coolest ice cream scooper in the world.” I’d be like, “I already have a fucking spoon. What do I need your spoon for?” But you can’t understand that unless you ask the question, and then how do you know that your super cool ice cream scooper is even cool? Maybe it needs to be hot.
These are the types of things that you find by doing natural and directed customer development. Most people build the wrong things because they think their problem is the same as everybody else’s and they get so focused in on that they get tunnel vision and then they create something that nobody wants. That’s why so many startups fail.
Louis: I’m so glad you’re saying this because there are a lot of startup entrepreneurs and lot of people in the industry would still say, “Don’t listen to your customers, they don’t know what they want.” I do understand that they don’t know what they want if you ask them for a solution, they are not going to be able to say, “Oh, I wish I could have this UTM builder that enables me to sync with my team.” But I do believe that if you listen to them the right way, asking them questions the right way, as you said, based on problems not in solutions, you will gather insights you want. I find that completely bullshit, this kind of idea that you should listen to them and there’s also this idea that Steve Jobs from Apple never listened to his customers, which is also the worst thing I’ve ever heard because this is not true.
Dan: He totally did. Steve Jobs listened to the amount of revenue that he made, and when there was an issue with revenue, he fixed it. His customers purchase specific products, he may not have ever done focus groups himself but he looked at the revenue and the money he was making. If something failed or succeeded, that told him which direction to go.
Louis: But beyond that, they setup NPS surveys in every single of the Apple Store, they do focus groups not even for years, they track how people react with their product. It’s so funny to hear that they are not listening to their customers and all that kind of stuff, it’s really funny. That’s interesting, I wasn’t actually expecting you to say this, you’re both analytically driven, you’re also driven by empathy, and customer development. What would be the next step once you validated this problem, what do you usually typically do after that?
Dan: Next step for us is always design. We don’t code, coding is the last thing that we do. What we want to do is get markups, get it actually designed in app like InVision. InVision app is really, really good. It’s actually how I raised money for my last company. We didn’t even have a product yet, we went in from customer development phase right into design, had that in InVision, we then went out with an InVision app and raised money. Everybody thought it was a real app because I was demoing it like it was a real app but they couldn’t tell the difference. Obviously we didn’t lie, we told them that it wasn’t built yet.
But we went through and then did customer development and customer validation with those mockups that we had in InVision. Our customers gave us great insights, they told us that this is an understandable, this doesn’t make sense, I don’t know what this tool does. And then we refined that design and made it so that it was something they understood and something they will use, then went to development. I think a lot of companies immediately start developing and they think about design second. That can get you in a lot of trouble because you need to make sure that your user interface has good user experience, and if it doesn’t, in this day and age it’s harder and harder to get traction. Definitely start with design and mockups before you go code.
Louis: Customer developments, once you know that your problem is worth solving, you start designing and mocking up. You will actually show that to potential customer or beta testers just to see how they react with this prototype, right?
Dan: Absolutely. Don’t get me wrong, there are some companies that have to go to code first. I totally get it. We built a data enrichment project two years ago, that product ended up failing and we shut it down. There was no way to show its value without going to code, because there was UI, no UI for it. But in most cases, you’re creating a product which people touch and feel and click, start with design before you go build it.
Louis: You can still design the process and how it works, you can still go through how the product will help you make your life better and all this kind of stuff. Even if it’s interface less type of product. You show those small caps, you improve the design, and then you just code it and you launch and then happy days, right?
Dan: Yeah. We try to always get a couple customers on board before we really build a tool and we have a couple customers that basically help us validate things as we build it, they beta test it. We’re working with three of the largest rap artists in the world with our UTM tool. You would never think that a rap artist would use a UTM tool, their agency does. We’d partner with them, gave them a free license, this is going to give us massive volume of people clinking on the links that we have and processing our system.
Now we can actually test the tool, make sure it works for those people and we build it around their use case. As we go through development sprints, we launch new features, they give great insights on, “Hey, this doesn’t work as well as I thought it was going to. This works as well as I thought it was going to.” We actually give those people free access to the tool. We have other clients that we make pay for it to get beta access, it just depends on your business and what your business objectives are. You’re trying to raise easy money, you probably want cash, but definitely find some beta customers to make sure that they could help you validate that tool as you build it.
Louis: I don’t know if you’ve heard of the Hotjar story but this is how Hotjar started as well. They did the exact same process that you did. It’s interesting to hear. I’m really interested in the answer to my next question, I think you’re going to surprise me, at least try to surprise me. Why do you think marketers have a bad reputation in general?
Dan: Because it’s too easy to become a marketer. Anybody who has an idea can say they’re a marketer and that’s typically the problem. There’s not one marketing in the traditional school world, high school to college, they cannot keep up with how fast how marketing is changing. The concepts that you learn in marketing in college are no longer applicable to the world that you and I live in. That obviously spits out a lot of marketers who have creative ideas which are basically old now. Some of the fundamentals are the same but anybody can say they’re a marketer because they have an idea and they know how to put paint on a white piece of paper and that’s going to be an ad. There’s no barrier to entry, that’s the reason why there are just so many bad marketers out there, but at the same time, that’s the reason why you have to go with people that you know have a lot experience and you can trust.
Louis: More particularly, how would you define a good marketer? How would you test that this person is a good marketer?
Dan: You definitely want to look at their history. When clients come and talk to me and they want a better understand why should they choose me or our consulting company, I tell them to just look at my credentials and look at the companies that I worked with, do reference checks on me, look at our testimonials, look at our case studies, also asking the clients we’ve failed on. I could say loudly. Contractually was not our best project ever. We did everything in our power to make that project successful but there was not necessarily alignment on both sides of the party on what the project was supposed to be at the end of the day. You can ask a marketer, “Hey, what has been successful?” That’s great! They’re always going to tell you what’s successful. But don’t be hesitant to also say, “Hey, what projects have failed?” I had at least three projects and I can say to a client, “This is what happened, this is how it failed, this is how we’ve changed our process to make sure that doesn’t happen.” These are the things that we learn from it and these are the things they learn from it. I can very easily say what my failures are. If somebody tries to hide their failures, you know there’s a big problem.
Louis: What do you think marketers should learn today that will help them in the next 10 years, 20 years, or 50 years?
Next thing is going to be you have to know how to analyze data, you’ve got to learn how to use Excel, you’ve got to learn how to use these analytics tools. Because right now what’s happening is everything is going to personalization, everything is moving to a one to one conversation. If you don’t understand how the code works, and you don’t know how to analyze the data, you’re never going to be able to do personalization. You really have to understand those two things or people like me are going to put you out of business.
Louis: What are the top three resources you would recommend to marketers?
Dan: It’s a tough one. I definitely think growthhackers.tv is a good resource for marketers to learn from other growth hackers, there’s definitely some good interviews on there. That is definitely a really, really good resource. I would definitely recommend, once again, signing up to a lot of blogs, growthhackers.com is another one that’s basically good content, always services to the top, there’s always good, fresh ideas there. Definitely read that. I didn’t get my training in marketing from a college, I barely even graduated high school. I almost dropped out because I ran a successful company. I learned all of my marketing from the internet, I’m a self-taught marketer and very good at it now. Definitely stay on top of things like that.
I would say as the last resource, depending on the vertical you’re in, I would say Udemy, CXL Institute, these types of companies are going to teach you the things that you need to know, not only from a marketing perspective but also from a technical perspective, and in analytics perspective. I would definitely say to check that stuff out.
Louis: Dan, you’ve been absolutely amazing. I’ve learned a lot from you in the last 45 minutes. How can people connect with you, contact you, learn from you?
Dan: Absolutely. Please check out effinamazing.com, we have a lot of resources on our blog. We also have a resources section full of webinars and as well as downloadable PDFs. If you want to get in contact with me, super simple, once again it’s [email protected] If you get bored, do me a favor, just Google Effin UTM and go download our Chrome Extension. It will really help you build UTM, sync them across your teams, make your analytics just that much more effective.
Louis: Dan, that’s a great way to end it. I’m definitely going to try it out. Thank you so much once again.
Dan: Awesome, thank you.
I’m a no-fluff marketer living in Dublin, Ireland (but yeah, I’m French).
I believe you can treat people the way you’d like to be treated and still generate results without using sleazy, aggressive, hack-y marketing. This is why I’ve started Everyone Hates Marketers – a no-fluff, actionable marketing podcast – as a side project in April 2017.
I’m also the Content Lead at Hotjar – a powerful way to analyse people’s behaviour on your website or app and understand how you can improve their experience.