I had an amazing conversation with Martin Gontovnikas, also affectionately known as "Gonto", during our Saleswhale Masterclass.
Gonto was SVP of Marketing and Growth at Auth0, where he managed a marketing team of over 100 people. Auth0 got acquired in March 2021 for $6.5 billion dollars.
We talked about supporting the sales team, running successful outbound marketing campaigns, and intellectual honesty in running growth and marketing experiments.
A sales director who worked with Gonto at Auth0 actually left an effusive LinkedIn post when we were promoting the webinar:
Which got me thinking.
As a marketer, if I interviewed our sales teams, how many of them would actually rave about us like this?
I know that my previous sales team certainly wouldn't!
This is when I knew that Gonto managed to accomplish something really special at Auth0, and I became really curious to figure out how Gonto did it.
I've tried to summarize some of the interesting learnings below, but do try to listen to the entire recording if you can spare the time.
Tell me something you did right in the early days
One thing that I think that we did right in the beginning was setting up the right KPIs for marketing to measure success.
Instead of thinking about how many leads or MQLs we bring, we decided that our main objective was pipeline and revenue.
If Marketing was not bringing in enough pipeline for Sales to close, then we are not doing good work. Period.
Were there ever moments you couldn't generate enough demand for the BDR and sales teams? How did you overcome that and what's your approach?
Yes, of course.
The first thing that's important is to bring up to the sales team immediately once you realize you are going to miss your target.
"Hey, this is what's going on, and we are going to do some research." Or even better, rope your sales team in to research the problem together.
It's important to be honest with your sales team - "Look, it's not working out, and we think here's why."
The culture of intellectual honesty that we build around marketing is important. And I think one of the biggest shear tensions between marketing and sales, or even marketing and leadership, is sometimes a lack of intellectual honesty, around what's working and what's not working.
Marketers, sometimes we tend to get defensive about our campaigns, and that actually creates alot of friction with sales, right?
Also, the whole idea of reaching across the aisle, and telling Sales, "Hey, we have a problem. Let's sit down together, as marketing and sales, and try to figure this challenge out."
How do you debug your marketing campaigns and tactics?
I'm a big believer in qualitative and quantitative feedback, but a lot of people only focus on the quantitative data. And they only talk about, "Look at the numbers, look at the ratios."
But, talking to people - like talking to prospects, to customers, to BDRs to understand why something is not working. That's a lot richer.
Talking to prospects to see why they didn't reply actually gives you so much more information firsthand, that you don't get by simply looking at numbers.
You have to do the research.
We actually record a lot of our customer conversations. And something that we pay a lot of attention to is the words that they use. Because, you need to talk in their words, their vernacular. Look out for specific words they use when talk about their business or problem or pain points.
Gabriel Lim (02:24):
Cool, okay, I think let's get started first, and then as the rest of them stream in, we can say hello to them. So, hi everyone, thanks for taking the time to join Gonto and myself on today's webinar. We want to keep today's session a bit more loose, so it's going to be more of a conversation. We hope to take questions from the audience as well. So, for those of you who have any questions, feel free to put in the chat, or if you want to come on to the show, I can give you speaking rights, and you can ask your questions in person as well.
Gabriel Lim (02:57):
Cool. Hey, Gonto. Thanks for doing this. Just by quick way of introduction, you used to be SVP of Marketing and Growth at Auth0, which obviously, for those of you who are unaware, they had a huge exit two weeks ago for $6.5 billion to Okta. You, yourself, you set it off as their sixth employee, right, which is kind of like their first, non-engineering hire, even though you yourself, you used to be an engineer before going to the dark side of marketing.
Gabriel Lim (03:36):
So, tell us more about yourself and what you have been up to. Recently, I saw that you just started a new kind of... Is it wrong to characterize it as a VC firm?
Martin Gontovnikas (03:48):
It is similar to that, but yeah, I'm a systems engineer. I actually started coding when I was younger. I started coding Visual Basic 6 and a game, because I liked games. I joined Auth0 as the sixth employee, and I was the first marketing hire, even though I wasn't really a marketer. I was doing developer relations, which meant that because Auth0 is an authentication as a service, and it's a product for developers, I was writing documentation, writing blog posts, speaking at conferences and stuff like that, to bring attention to developers about Auth0.
Martin Gontovnikas (04:24):
As you said, I started as their first marketing hire back then. And then, I always say that we Argentinians have a lot of opinions on everything. We're usually wrong, but we do have a lot of opinions, so I had a lot of opinions on how we could improve our signup experience, self-service and stuff like that. So, we started to try some of those experiences and those experiments, and I was given the opportunity by the previous CEO who was Jon Gelsey, Matias and Eugenio who are the two founders to actually run marketing, even though I have never done that before, so it was quite a fantastic and crazy experience. I always said I'm a weird marketer because I'm not-
Gabriel Lim (05:03):
That's the kind of crazy things that engineering founders, right? Give the role of marketing to someone who has never done marketing before in their lives.
Martin Gontovnikas (05:12):
Exactly, and for me it was like, I actually thought I was experimenting and doing experimentation, only because I knew nothing about marketing, so it was the only way that we could try to grow or something like that. I left Auth0 three months ago, maybe, but by the time that I left, my team was approximately 100 people, so it was a big marketing team, considering Auth0 was 900 or something like that.
Martin Gontovnikas (05:40):
And then, as you said, recently I'm starting something new with Guillaume Cabane, or G as everybody knows him. It's called HyperGrowth Partners, and what we do is we look for companies who have found product-market fits and are looking to drive hyper growth and are looking to scale and get their customer acquisition costs lower but still bringing, like product-led growth and top-down sales.
Martin Gontovnikas (06:06):
What we do is, it's more like a VC, because we try to look for companies that we can help, and we do not charge any cash, we charge equity in return, and we make it like a 12 to 24-month contract, so that's a shameless plug. If you are a B2B company, and you are looking for help on exponential growth, you can contact me at firstname.lastname@example.org.
Gabriel Lim (06:29):
Awesome. Nice. Which makes a ton of sense, as well, given your background and putting your money where your mouth is and taking equity instead of... I love this model, actually, are you guys one of the first in the world to deploy something like this.
Martin Gontovnikas (06:49):
We're not the first, there's a few others. Like I've seen, for example, sweat equity, which is another one, that's something similar. Most do not do it. What I like about it is that we have skin in the game, meaning we get nothing if the business doesn't grow. The other thing that we do is, we give every company that we help a three-month cliff on their stock options. So, the idea is that if they don't get value in the first three months, they can actually tell us to f* off, and we get nothing.
Gabriel Lim (07:16):
Nice. Awesome. Cool. So, let's get to the meat of today's discussion. I think most people are probably on this call because of this really awesome quote by one of your former directors of sales. And, for those of you who are unaware, let me just read out this quote, a testimonial from a previous sales director at Auth0.
Gabriel Lim (07:38):
This guy, he said, "Gonto is the real deal. My team was a direct beneficiary of what he built at Auth0, and there is no way that we would have been able to grow the company anywhere close to the level we did without the marketing machine that he built. After many years of sales leadership, this was the first time that I wasn't part of a team that said that there weren't enough leads or that the leads weren't valuable."
Gabriel Lim (08:09):
Oh my god, so Gonto, tell us more about your secret. I think everyone on this call wants to know how you did it.
Martin Gontovnikas (08:20):
I don't think that there's any particular secret. One thing that I think that we did right in the beginning was setting up the KPI instead of thinking about how to measure success. I see a lot of teams, marketing teams, who think about how many leads they bring or how many MQLs they bring, or how many MPLs they bring. Something I think was important for us is that the objective for us was pipeline and revenue. If marketing is not bringing pipeline so that then sales can close it, it's not making good work.
Martin Gontovnikas (08:53):
By making marketing have ownership, or KPIs until pipeline, meant that if leads didn't convert, we sucked, basically. We needed to have them convert. I think that starting with having pipeline as a shared objective, that was the objective for marketing, meant that we needed to create good quality leads for sales, because if they didn't convert, they weren't going to be useful at all for us.
Martin Gontovnikas (09:19):
I think that another thing that also we did a lot was, in the beginning, it was all inbound, but we were continually experimenting and trying new things to increase the number of leads, but we also cared a lot about what's the quality of those leads, meaning, did we get meetings? Are they using the product? How much are they using the product? What does it mean?
Martin Gontovnikas (09:40):
So, we focused a lot on that as well, which also I think helped increase and make the relationship with sales better. We were also very honest. When something was not working out or something was a problem, we were always the first to raise a hand and say like, "Hey, this week maybe leads are not great. Hey, this week this is happening," or something like that. So, we were also transparent about what was working and what was not working.
Gabriel Lim (10:06):
Nice. Again, earlier you said that, in the early days, a lot of the growth was fueled by inbound, right. Just to make sure that everyone here is on the same page, in your opinion, what's the difference between inbound marketing and outbound marketing?
Martin Gontovnikas (10:20):
So, at least in the beginning of Auth0, how I see inbound marketing is when people are somehow, organically, so without paying, coming to your site, and then they are either trying your product and eventually talking to sales, or they are clicking on "Get a demo," or "Talk to sales" or something like that.
Martin Gontovnikas (10:38):
In the case of Auth0, in the beginning, it's a tool for developers. It's about adding authentication to a platform, and the developers are the ones who are adding that authentication to the platform itself. At least for us, it was a lot of around content. Everybody says, "Yeah, it's content." We had a big strategy on content around... We actually hired, in the beginning, engineers who are writing content. We have no writers. They were engineers who were actually talking to other engineers, and only a part of their job was writing, but the rest wasn't. It was like learning new technologies and doing open-source projects and stuff like that, but it was a big part.
Martin Gontovnikas (11:22):
The other part was we are always on the lookout for trends, like what's happening? Why? So then, we could talk about them. There was, back then, one programming framework called Angular JS that it was starting to become popular, but it wasn't that popular yet. So, I made the bet that was going to become a popular framework, so we thought about what if we become thought leaders here. We started to create content about Angular JS, but we also went to every conference around the world that spoke about Angular JS, and not because of the people that were listening to us, but actually, because we were drinking beers, drinking wine and going out with the Angular teams.
Martin Gontovnikas (12:07):
We became friends, so then, because our content was good, they were sharing our content, and we ended up becoming thought leaders of Angular, where anybody who had a question about authentication in that community was actually asking us. That's how we started, at least, with people coming to us and inbound.
Gabriel Lim (12:26):
I love it. That's clever. Sorry, continue.
Martin Gontovnikas (12:32):
No, it's a lot. The other thing about something that's outbound for us is when we didn't sell to developers. We were selling to directors of product, VP of product, CTO, enterprise architect and similar. For those, we did do typical B2B demand gen, like we had multiple touchpoints. We had webinars like this one. We got trade shows. We did direct mail. We did nurtures. So, more typical, I would say, B2B demand generation.
Gabriel Lim (12:59):
This is, I suppose what, for you, most of the demand real quality inbound leads... Sorry, not inbound, real demand for the sales team, right? I can't imagine many salespeople want to jump on a call with a developer, right?
Martin Gontovnikas (13:13):
The thing is that the developers... So, how we nurtured those inbound leads is, once the developer got to us, they started to try the product. Once they tried the product, the developer was bringing it up to their boss, engineering manager, director of engineering, and then they were talking to them.
Martin Gontovnikas (13:30):
At Auth0, we actually got to 25 million in revenue with no outbound. We had only a handful of salespeople, and it was all through the developers that talked to their bosses or similar, and then they eventually came to us, and they closed deals. It's a bit of both on the sales side.
Gabriel Lim (13:49):
Got you. Oh my god, that's awesome. I'd love to double click more on that, but I'm starting to see questions coming in from the audience, so I think maybe we can take a few of them as we go in. I think, interesting question is from this guy called Reuben. So, hey Reuben, thanks for asking the question. He said that, "I agree that marketing should be held accountable for the quality of leads, however the quality of sales will always be the variable that marketing will not have control over." How do you think about this? And, I think this is kind of a perennial problem with most sales and marketing teams, right?
Martin Gontovnikas (14:25):
I actually think it has to be a partnership. So, couple of things, one, we own the team pipeline not revenue, so that once the opportunity is created, that's what we own. Then, if the sales team could not close it or something happened, that was something else that we need to inspect. What pipeline drove is that, for example, ADRs, or BDRs, or SDRs or however you called them, are a big part of the conversation, and they are right in the middle. Something that was important to us is they were the ones who were helping convert to pipeline, because they were talking to the leads, they were having conversations, having meetings and then creating opportunities.
Martin Gontovnikas (15:04):
Having pipeline meant that we had a very strong relationship with the BDRs team. To give you an example, we even have two people on the team that were writing copy, writing the nurtures for the BDRs, writing scripts for the calls, and they all live in marketing, because we needed the BDRs to work. Saying, "No, we cannot help to do what sales does," I don't think that's true. It has to be a partnership. It has to work together, and once you own pipeline, that means that we do deeply care about how the BDRs are working and what's working or not.
Martin Gontovnikas (15:40):
When some BDRs were not working out, we proactively gave feedback to sales, like, "Hey, these two people are not working out. Have you checked what's going on?", and stuff like that. I think you have to be proactive if you want to make sure that it works.
Gabriel Lim (15:54):
Awesome. Back at Auth0, the BDRs and the ADRs reported to sales, right, not marketing?
Martin Gontovnikas (15:59):
Gabriel Lim (16:00):
So, technically, you did not really have "full" control over them, but I love how proactive you were that you created messaging, collateral, writing the email content for the BDRs, which most of the time if left to their own devices, you're going to start... I mean, number one, they don't have experience. Most of these, they are like fresh graduates out of college. And second, you start to see this messaging drift, which is what happened at my previous company. We see different people are talking about the same product in 100 different ways.
Martin Gontovnikas (16:30):
Exactly. I think you have to put a big focus into how you work with the BDRs. We've done multiple experiments with them on, for example, do we pick the leads for them and they have to talk to specific ones? Do we give them a pool where they can pick from which ones or not? Even something that was one of the most important things for Auth0 when working with BDRs for these outbound campaigns was selling to them on those calls.
Martin Gontovnikas (16:57):
Like, something that was fascinating to me was, in Europe for example, direct mail didn't work. In America, it worked great. It was one of the biggest ones. When we started talking to the BDRs to understand a bit of what was happening, all of the BDRs in Europe did not believe in direct mail. So, if they do not believe in direct mail, of course it's not going to work, because if you don't believe in it, it's not going to work out.
Martin Gontovnikas (17:22):
So, since then, we've put so much effort now in selling campaigns to the BDRs on how and why they're going to bring a lot pipeline. They are going to get a lot of money, because they're going to close deals, and what's different about this one and what's there to gain. Since we started doing that, for example, we started getting so much more traction and success with the BDRs. I don't think you see enough marketing teams doing that, because they do exactly like what the question was saying on, "No, this is not our job, so it's up to them."
Gabriel Lim (17:52):
Awesome. I love it. So, basically, it's about taking internal marketing as serious as you take external marketing, right? This is kind of similar to this guy, Dave Gerhardt, what he mentioned on one of his podcasts. A lot of marketing leaders, they make the mistake of only messaging outwards, and they forget to market internally.
Martin Gontovnikas (18:10):
Exactly, 100% agree.
Gabriel Lim (18:12):
Nice. So, I mean, if you were going to brass tacks, how exactly do you do that? How exactly do you get them fired up? Do you do a Zoom call and do a presentation, "Hey, this is what we're going to be working on." How exactly do you get your BDRs fired up about these campaigns?
Martin Gontovnikas (18:30):
First of all, we actually have projections on how many meetings we think we could get from each of them, how much money that meant for the company, but also how much money it meant for them. We also talked a bit about what are the creative parts of the different parts of these campaigns compared to others and why we think that's going to work. Like, if we do a direct mail where we're sending an iPad that is more expensive, then why do we think it's going to bring bigger ASP and what does that mean for them? Why is that going to make a big change compared to others?
Martin Gontovnikas (19:03):
We also did talk very energetic on the meetings. We actually did a lot of work on that.
Gabriel Lim (19:09):
I can see that.
Martin Gontovnikas (19:10):
Exactly, being energetic, talking passionately about it and stuff like that. That was definitely a big part on how we talked to the BDRs on those calls as well. And then, finally, we also talked a lot about specifics on where we thought they could target and where we thought they could add their own flavor and why so that they also understand how their own personality would be a part that they could help on some of those campaigns.
Gabriel Lim (19:39):
Got you. Makes sense. Nice. I guess, just to wrap up this thread. I'm sure that there's been challenges, were there ever moments that you couldn't generate enough quality conversations to keep the BDR team "fed" so as to say. How did overcome that? What's your approach? Okay so, how do you overcome that and what's your approach to overcoming that?
Martin Gontovnikas (20:03):
We have, I would say, multiple times that we have problems and different types. One thing that I think is important is, one, bring it up to the sales team almost immediately, "Hey, this is what's going on and we're doing research." Or, even trying to do research together with them. To give you an idea, there was some time when we had a lot of leads in the small and medium business and mid-market but not enough on enterprise. We did a deep dive into what's happening with enterprise.
Martin Gontovnikas (20:32):
So there, we learned, for example, that sometimes we were not getting enough leads. In other cases, we were getting leads, but the problem is that they didn't qualify with BANT, so many we needed to nurture them for more time, because they were the right person and we just needed to find the timing to get the right budget or the right project. It was something that we were not doing before.
Martin Gontovnikas (20:52):
So, it was a lot about doing some research into the specific things. Other thing is, for example, when we know that we're not going hit a metric, like form fills for us was a metric, which means how many people click and talk to sales and come to us. When we knew we were not going to hit it, we gave them a heads up two weeks before, "This month we're not going to hit it."
Martin Gontovnikas (21:12):
We did a team, together, to investigate on the why. In some cases, for example, one was an A/B test that we had run that it wasn't significant yet, so we kept on running it, and it was killing our numbers, so we realized that and we killed the test, and it started to improve again. But, we were only able to do that because we were constantly checking, and we were also being honest about, "Look, it's not working out," or it is working and why. That was another example that I think was very key to how we thought a bit about that.
Martin Gontovnikas (21:47):
Finally, another thing that we did a lot. When we moved from inbound to outbound, it was mostly for two reasons. One, we needed predictable revenue, because inbound, sometimes these calls, people come, people don't come, et cetera. The other one is we did a lot of math understanding of how inbound was growing month-over-month, and what we saw is that the growth was starting to decrease. We started to foresee that in one year, or one year and a half from now, we were not going to hit our numbers if we continued that growth perspective, so we needed to do something different.
Martin Gontovnikas (22:20):
We actually met one year before, to start thinking about what can we do so we can actually hit the number next year or in six months.
Gabriel Lim (22:27):
Martin Gontovnikas (22:27):
So, some planning and strategy about that is key, because in our case, doing the outbound campaigns, they didn't work at all in the beginning. It took us nine months to get them to work. Another example that I have on this is, outbound campaigns, we did in the beginning the typical industry, persona and then pain point. We did some emails and nurtures, webinars, et cetera.
Martin Gontovnikas (22:52):
A lot of them did not work. They just didn't. We actually then paid some of the people that we were marketing to to see if they could give us feedback on why it wasn't working out. We paid them money just offering that, "Hey, I'm not going to sell you the product, it's done, but I'm going to give $100 if you tell me why, what's going on?".
Martin Gontovnikas (23:12):
What we learned there is that some of them told us, "Auth0 looks interesting, but we do not have a project for this right now, so that's why I didn't even answer." By looking at that, we actually went back to all of our Salesforce data that we had, and what we realized is that people didn't install authentication just because. They had to have an initiative that went with it. That initiative could be privacy or compliance regulations. It could be they are moving from a monolith, on-premise, to the Cloud. It could be they are integrating multiple services.
Martin Gontovnikas (23:45):
Then, once we realized that, we started to target companies based on those initiatives. So, for example, if they are moving from the on-premise to the Cloud, usually those companies are hiring a head of growth or a head of digital. If they have a privacy or compliance, that means that they either just hired a CSO or just hired somebody in privacy, or there's a new compliance that is going out.
Martin Gontovnikas (24:05):
We started to wield these manual scores where we looked into companies to try to guess if they are going through one initiative, and we did outbound to those. That, for example, was a big success eventually, but it came from the complete failure of - we tried from six months, and nobody replaced through none of my [wording case 00:24:24].
Gabriel Lim (24:24):
Oh my god. Thanks for sharing. So, I see, actually, two interesting themes here. The first theme is that, number one, the culture of intellectual honesty that you have built around your marketing at Auth0, and I think one of the biggest shear tensions between marketing and sales, or even marketing and leadership, is sometimes a lack of intellectual honesty, right, around what's working and what's not working. Marketers, sometimes we tend to get defensive about our campaigns, and that actually creates alot of friction with sales, right?
Martin Gontovnikas (24:57):
100% agree. I think that's a great summary.
Gabriel Lim (24:59):
Yeah, and the second interesting thing that I see is this whole culture as well around experimentation. I see that you're approaching problems very systematically, which probably is because you have roots as an engineer. You're approaching it like a scientist, "Hey, we are facing a problem, what's going on? Let's decompose the problem. Let's get down to the ground truth, talk to the people and do firsthand research," instead of just staying behind our computers and just looking at the charts and numbers and then coming up with all sorts of crazy theories.
Gabriel Lim (25:31):
I really love that. So, I think that may be one of... Also, the idea to reach across the aisle and say, "Hey, we are stuck, we are having a problem. Let's sit down together, as sales and marketing, and try to figure this challenge out."
Martin Gontovnikas (25:47):
Exactly. I'm a big believer in qualitative and quantitative feedback, but a lot of people talk about data and they only talk about, "Look at the number, look at the ratios," which we do, on call-to-conversations, conversations-to-meetings, meetings-to-SQLs. You should look at that, because it gives you some insights. But, talking to people, like talking to BDRs to understand why something is not working. Talking to prospects to see why something they didn't reply actually gives you so much more information firsthand, that maybe just looking at numbers is not enough.
Gabriel Lim (26:19):
I totally agree. There's something similar that we went through as well. Last year and the year before, we were positioning ourselves as an AI assistant that does “yada yada yada”, but we were getting so little replies on our outbound emails, and were attracting a lot of non-qualified leads through our website. What happened was that I picked up the phone, I called 10 of our customers and then I reached out to over 80-plus prospects and asked them, "Hey, what's going on?". The predominant theme that I kept hearing was that, "I don't understand what the hell you guys do. I don't know why I need an AI assistant. What's an AI assistant?". This is terrible messaging.
Gabriel Lim (26:55):
So, we went back, we went changed that, and almost overnight, you see a huge spike in the number of responses and qualified leads they get in. It's as simple as just using the vernacular and the words of your customers or your prospects.
Martin Gontovnikas (27:12):
Exactly, I think that's a great example of how overnight success takes a lifetime. You have to do the research. You have to look it up and stuff like that. Another thing that you just mentioned that I think is key is we actually record a lot of our customer conversations, and something that we pay a lot of attention to is the words that they use. Because, you need to talk in their words, so if they are using specific words to talk about their business or authentication or identity, we need to use the same. We looked at that per industry, or globally, because that makes it feel so much more approachable. So that, I think, which you commented, is also a key thing that everybody should look into.
Gabriel Lim (27:51):
Awesome. So try out Gong, maybe?
Martin Gontovnikas (27:54):
Gabriel Lim (27:56):
Nice. So, let's take another question from your friend, Clayton, so he's asking, "I would love to hear Gonto's take on product qualified leads. What's the progression of learning how user behavior within the product qualifies them as potential customers with high intent? What's your advice towards early stage setups in terms of instrumenting user behavior and beginning to understand the PQL model?".
Martin Gontovnikas (28:20):
I like, a lot, thinking about looking at behavior in the product. I personally, don't like calling them PQLs, because that's reporting on a number that is not real, meaning it adds no real value. Having said that, I am a big believer on having a score, or a machine learning something, that will help you better understand how people interact with the product.
Martin Gontovnikas (28:43):
Something that I think is important when you think about product behavior is two things. If you're looking at product behavior, that means you're a product-led growth company, so people are doing a trial or a free account or something like that. There's two objectives there, either you want them to be retained, meaning you want them in 12 months to continue using the platform, even if it's for free, you don't care. Because eventually, if it's for free, they might go to another company and then they’ll buy you.
Martin Gontovnikas (29:09):
Second, is you want them to convert to an opportunity. Those two are, sometimes, completely different objectives, because if you look at product usage that will drive into opportunities, sometimes they don't have that much usage, and they just do a demo or something and they try it out. So, you need to well understand what are the patterns that drive a product account to convert into opportunities, and then look at those into sending them to sales or something like that.
Martin Gontovnikas (29:35):
And then, for those that are retained or your focus is retention, you also need to find a way to send them emails or push notifications or help them out to be successful in the product. It's not only about selling, but it's about getting people to be retained by helping them understand how the product works and helping them be retained with the platform.
Martin Gontovnikas (29:56):
I do think it's key that you always think about both, because SQOs are great, short-term, but if you only drive towards SQOs, you are maybe killing your retention. Retention is long-term, because people who are retained eventually are going to buy from you, so I think you have to always keep both things in mind and try to put a big focus into those and make sure that when you improve one, you don't kill the other, let's say, too much.
Gabriel Lim (30:23):
Interesting. I like that. You are balancing both the short and the long-term. You're not treating your product qualified leads, so as to say, as just a mega funnel that you're trying to churn them into sales qualified leads.
Martin Gontovnikas (30:37):
Exactly. I think a lot of companies don't think about retention, and they always seem to think about short-term, so I do think it's important to think about long-term. One thing on this is, I do think it's very important on who the person that is using the profile on your product-led growth is, meaning if they are a developer, for example, whose bullshit sniffer is very high, I don't think you should try to sell them through a BDR, maybe what you should try is having a DevRep or a technical person reach out, because they will feel more familiar. And eventually, they'll then connect you to somebody who then you can sell.
Martin Gontovnikas (31:12):
However, if it's somebody who's more used to talking to sales, and they're also a user, maybe it's okay. But again, I think you need to A/B test it. We did a lot of A/B tests on contacting signups, not contacting signups, sending nurtures versus sending BDRs, on having a technical person reach out to do demos, or having somebody who was not technical and stuff like that.
Gabriel Lim (31:33):
Got it. Nice. I think, let's get a question from Ash. Ash, you actually asked a series of three questions, but I guess the overall theme is how do you outbound properly? What's the process from your emails to your process route, creating quality leads, Gonto, how are you thinking about this?
Martin Gontovnikas (31:54):
To me, at least, the first thing is you need to try with some companies that you think, just gut feeling, that might be a good target and send them something, like couple of emails, couple of calls and stuff like that. From there, I think then you need to understand, did it work or not? If it worked, great, you were a lucky motherf-er. If not, that means that I think you need to do some digging. That's similar to an example that I gave before on trying to talk to the prospects and get some feedback from them. Try to look at the emails, try to look at the responses and stuff like that, and then figure out what is the magic type of companies that better worked for you.
Martin Gontovnikas (32:31):
For us, as I said, it wasn't about industry and persona, it was about the initiative and the persona. Once you know the initiative, I think now it's time to do more account-based marketing. What we did is look at companies from lower 2000, or something like that, put what are the parameters or columns that you think are valuable, like are they hiring a growth person, as I said. Do they have a privacy person? Do they have more than three login forms? Are they using HTTPS? Is the login form telling you that the username doesn't exist, because that shouldn't happen.
Martin Gontovnikas (33:07):
Then, based off all of that, you put a score, manually, on are they doing it? Yes, no, from zero to five or something like that. We did Amazon Turk or FunnelBeam or multiple services that we have a human look at those companies and look at the information online and fill those numbers. From there, now we have scores on which were the companies that we should contact.
Martin Gontovnikas (33:29):
So now, we have the ability to pick companies based on which ones we thought were more likely to buy us or were more likely benefit, actually, from buying us. It's a real benefit that they would get from us. From there, again, a lot of A/B testing. We, for example, ended up getting to a 16-touchpoint model, which is a mix of LinkedIn, calls and email. We actually had a scale. First week, two emails one call, second week, two LinkedIn, one of this and stuff like that.
Martin Gontovnikas (34:04):
Because we created tasks for the BDR to do something, we actually continually A/B tested. We had 30% doing one thing, 30% doing another thing, and then we looked at which one worked better until we got to a place where we thought we couldn't increase it more. And then, we did 80% that and then 20% still continued testing, continually, new things to see which one could work better if there was something new.
Martin Gontovnikas (34:30):
But, I think that also requires a lot. For us, as I said, we found out that it was 16 touchpoints, and like 90% converted within the first two or the last two. Nobody converted in the middle, so that's why we actually needed to do the 16 touchpoints. Similarly, for us, having a call in the first week in the first two days was really useful. Or, for example, if somebody did a form fill, we discovered that if we contact them in the first hour, it makes a big difference, so we also optimized for doing some of those things.
Martin Gontovnikas (35:03):
So again, I think it's about constantly evolving and trying to better understand how you can target your customers, but making them understand the benefits, showing them value, and also making sure that your emails are not very long. We use BASHO style of emails, but I think sweet, to-the-spot, clear, simple is key as well.
Gabriel Lim (35:27):
Awesome. Thanks. I like what you mentioned earlier about continually testing. You keep 80% of what works and then you're continually testing the remaining 20%. I think that's probably something I'm going to steal as well, but this may sound like a pedantic question, but did you test it out on 80% of the BDRs and 20% of the BDRs for new things, or is it like the same BDRs, but you split their time 80% trying old things and 20% trying new things?
Martin Gontovnikas (35:50):
So, how we did it, it was usually per campaign and per region, so that meant that it was more around that rather than a specific BDR. I do think that you should test it with multiple BDRs, because if you always test it with one or two, if they are great or if they suck, it's not going to give you that much information. So, I actually think that testing it with multiple BDRs is going to be much more useful than just per one.
Martin Gontovnikas (36:14):
For us, it was like, "Okay, Campaign 1, we're going to do it in these four regions, 80% of the people who are running the campaign are doing this, 20% are doing this," and then we randomly send that to specific BDRs.
Gabriel Lim (36:26):
Got it. Awesome. Nice. Does anyone else have any questions for Gonto? Maybe we can take one last question. "Will we be receiving a recording of this webinar?" Yeah, definitely. I'll be sending the recording of the webinar after this. Anyone else has any questions?
Gabriel Lim (36:53):
So, I have a question. For yourself, you basically created this whole framework around demand generation that's very, very experimental-focused, right, and that's something that I think the marketing world is still kind of waking up to. Could you share with me a bit more about your experiment framework? Where do you get ideas from? What were some of the experiments that led to some interesting insights, at least from your point of view?
Martin Gontovnikas (37:28):
So, on the ideas from experiments, I think that's something that's important, is that ideas come from everybody, and they should come from everybody. Everything should be constantly experimented, but sometimes something that we did is, for example, when we did an offsite, we had people from other teams think about problems from another team. So, we had a DevRel think about the DemandGen problem or DemandGen think about the DevRel problem.
Martin Gontovnikas (37:52):
We could get a diversity in that sense of ideas from another group that has nothing to do with that. Then, for example, when COVID started, we did a survey called "F* COVID," and the survey was all about getting creative ideas on how we could test COVID out. What we told everybody is that if they send an idea, and we tested their idea, they will get $500 in return. That was not only for marketing, that was for the entire company.
Gabriel Lim (38:23):
Martin Gontovnikas (38:23):
Anybody could send us ideas on things to try, so I think that's, on one sense, is key. Another thing is being methodical on, first doing qualitative and quantitative testing and understanding, with interviews, looking at data, what are the things that you should try? Another one is being very specific on how you write the hypothesis. The hypothesis has to have the qual and quant data that you find out, has to have a specific time period or a specific number of blocks or something that you're going to try, has to have a KPI, has to have a goal.
Martin Gontovnikas (38:56):
And then, you need to work on that, build it, and then execute it for some time. Then, I think, you also need to make sure that whatever you do, you have to learn something. I always say that the only failure is that if you don't learn anything. So, if you do something, and it doesn't work, but then you dig into the data or you do some interviews, and you learn something that test was awesome.
Martin Gontovnikas (39:21):
There's actually a study from HBR that says that one in every 13 ideas works. That means that 12 of your ideas suck, but if you learn from one to one the next, then one in seven. If it's one in seven, it's much better and it's closer, so I think that is also something that is really important to think about what is the learning and actually keeping a database of learning. We kept database of the learnings that we got from experiments so that then others could look into what have we learned in the past, so we don't have to do the same experiment again or lose it.
Gabriel Lim (39:54):
Where were you guys keeping these learnings, like in JIRA or Google Docs?
Martin Gontovnikas (39:57):
We had it in Confluence and then we actually moved them to Smartsheet. We had a big Smartsheet, like an Excel table, with all of the ideas and all of the things. You could do the same with Airtable or others, so it doesn't even have to be a very sophisticated thing.
Gabriel Lim (40:10):
Martin Gontovnikas (40:11):
Another thing that is important, I think, is the culture of failure. A lot of companies think, "We are experimenting," but then what happens is they try something, and then the manager or the boss will say, "What the f* did you do? Why is this not working out?". That doesn't encourage experimentation, because experimentation is all about embracing failure. If people are afraid to fail, they are never going to experiment. They are never going to try something new.
Martin Gontovnikas (40:42):
That's why I think you have to be very... As a leader or a manager, you have to make sure that when there's a failure, you ask if they've learned something. If they haven't, then you can tell them like, "What the f* are you doing?". But, if they have learned something, that's okay.
Gabriel Lim (41:00):
Awesome. Here's an interesting scenario that sometimes we face, right? You have reports who refuse to admit that something is a failure. I don't know how you approach this, but the way I think about is setting very clear hypothesis and parameters before you go out and run the experiment, or are you thinking about this differently?
Martin Gontovnikas (41:19):
I agree with that. I do think that every time you're going to run an experiment, it has to have the KPI, it has to have the goal, it has to have everything, so that then you can measure effectively, if it worked or if it didn't. Then, if somebody is not accepting failure or hiding failure, you should let them go.
Gabriel Lim (41:34):
Oh, wow. Okay. Nice. Cool. I wish we had more time to go on this, but I see one question from Nicholas DePaul, actually two questions, "What tools do you use to manage BDR tasks and how do you control the process?".
Martin Gontovnikas (41:50):
On the BDR tasks, we actually use Salesforce. Salesforce has Salesforce tasks, so we have Marketo connected to Salesforce, and based on Marketo nurtures and stuff like that, it will create tasks in Salesforce. We also use Outreach, which is the thing that they use for sending the sequences from the BDRs. So, similarly, they will get a task, they will need to enter some personalization and stuff like that, and then they can send it from Outreach.
Martin Gontovnikas (42:13):
So, that's the main thing, and then from there, we did use Tableau to get data from Salesforce, from Outreach, and all of them and look at that together. As I said, the main things that we did was calls-to-conversations, conversations-to-meetings, and meetings-to-SQLs. Some examples of the things that we looked at is if BDRs are not getting calls, that means that the messaging that marketing did is not good, at all. Because, otherwise, at least they should be getting calls.
Martin Gontovnikas (42:47):
If they are getting calls, but then the conversation is not great, because there's not meeting and there's nothing else, a lot of times that's the BDR, because they have the messaging because it worked to at least get the call, but then the conversation was not great. If the conversation, then, doesn't move to a meeting, it might mean one of two things. It might be that the messaging is not concise enough so that they get, and they finish it, but it's appearing on the emails, so you need to improve your scripts for the calls, or it might mean, again, that the BDR did not do a good job, so you need to look into it to better understand what's going on.
Martin Gontovnikas (43:24):
But, I think that looking at those metrics is something that I would definitely look for that. And then, how do you control the process? So, the process around like we have an experiment, we're going to run it, and then the experiment can be used in multiple platforms. Like, for websites, we used Optimizely. Marketo includes control and test groups, so you can do it. Outreach, the same. So, it's multiple tools, and in each of them, you have to make sure that you are looking at that.
Martin Gontovnikas (43:49):
Having said that, for experiments, we actually had a template. We had a slides template on how you present an experiment, meaning what was the phrase that you had to use, what was the data, showing the region and the variation. And then, looking at if you expect a change of 10% and you know the website traffic, and you do an A/B test on the website, you actually can know how much time you need for statistical significance. That was the other part, "Look, we need to do these tests in the home for two weeks to know that it's 10% better. If it's only 2% better, it's five weeks," or something like that.
Martin Gontovnikas (44:24):
All of that was mandatory, and on the smart sheets that we had all of the learnings from, the experiments, we also had the experiment definition since the beginning, and that's where the slides were actually linked from there.
Gabriel Lim (44:37):
Awesome. Nice. So, one last question from Clayton. Would Auth0 be as successful if you guys didn't build a large sales team?
Martin Gontovnikas (44:47):
If we didn't have a sales team? No, I don't think it would have been successful at all. I'm a big believer that you need to have both top-down sales and sales and marketing. Peter Thiel talks about this in Zero to One, like whoever says, "No, the product will get them. No, It will just work," you are delusional. I think that when you have a company and you have the founder that is doing a good job, and people are buying, it's because the founder is a very good salesperson. But, you do need sales. I'm a big believer in having sales and having marketing.
Martin Gontovnikas (45:22):
Having said that, one thing I think is important is understanding that the different profiles of sales, where if you have inbound, maybe it's more about the people who are coming are convinced that they need to change. So, it's more about the salesperson has to explain better why Auth0 is better, in that case, and then get them to convert and to close. Have said that, if it's outbound, you need to first explain that they need to change, then why, then why Auth0, so it's a different process, so I do think you have to think about that when you hire. But yeah, big believer in sales.
Gabriel Lim (46:01):
Awesome. It's something that I think, as former engineers, I used to fall into this trap as well. If you build a great product, that's all it takes, right, and somehow the world will build a path to your door, but more often than not, that's not the case.
Martin Gontovnikas (46:14):
Gabriel Lim (46:15):
Nice. Cool. So, I think we're at time. Thanks so much, Gonto, for this session. I personally took so much away from this, and I think the audience as well. Yeah, for those of you, feel free to drop either myself or Gonto a note on LinkedIn, and we will be sending recordings out to all the attendees.
Martin Gontovnikas (46:34):
Thank you very much for inviting and thank you everybody for attending.
Gabriel Lim (46:38):
Thanks Gonto. See ya.
Martin Gontovnikas (46:39):
Gabriel Lim (46:40):
Bye-bye. Bye, cheers.