Video: The product principles that have helped guide Gong’s success | Duration: 1352s | Summary: The product principles that have helped guide Gong’s success

Transcript for "The product principles that have helped guide Gong’s success": I'm excited to have you here. Today, we have the privilege of speaking to one of the most prominent product leaders and my hero, Elon, the CPO and cofounder of Gong. We're really excited to cover a topic that I think everyone here is going to find super insightful and actionable, which are the product principles that Elon has adopted in applied at Gong. So a little bit of housecleaning before we get started. If you have any questions at all, feel free to submit the questions through the q and a tab, and we can definitely get to it at the end. If there's not enough time to get to it, then we'll definitely, follow-up in via email. Finally, there's no pressure to attend the entire webinar. This is recorded. So you can also receive the recording after the webinar has ended as well. Great. So let's get started. So I'm really excited for this conversation again because Gong is one of my favorite products to use. We use it every day. All of my entire company uses Gong. And so really excited to hear what your opinions are on some of these, topics. So to get us started, can you walk through your career journey leading up to where you are today? Sure. And thanks for inviting me. I'd love to kinda meet everybody here in the, in the crowd. So indeed, today, I kinda I'm one of the founders and head of product team at, Gong. But earlier on, I'm a software engineer by training. So kinda had back to door and master's degree in engineering. Let the let the I was software developer for a while, led engineering teams, and made a transition to the product, in a total product role somewhere in the beginning of the century, which is a little bit yeah. Probably now you know how old I am, but it is what it is. Wow. Yeah. I guess that is relevant in the beginning of the century. So you know what? Go for it. Live live for live confidently with that. Great. So what are some product principles that you and the rest of the leadership team decided on in the beginning, and how did that end up helping or hurting Gong over time? Yeah. So when we when we started Gong, it wasn't obvious that, the idea that people can we we start nowadays, Gong is is essentially a a whole platform for revenue organizations to be, efficient to drive productivity. But when we started, it was mostly around understanding what the field does, and the way we've kinda instrument this understanding is by capturing the conversations people had with customers. And and that means to start with recording video calls. And then we added much more recording phone calls, emails, and many more many other more things. But, initially, it wasn't obvious that people don't wanna are gonna wanna be captured. Every every sales person on earth which uses who uses Gong for the first time is like, oh, big brother, somebody's gonna be listening to my calls. Like, no, dude. Nobody wants to listen to your calls. Just people want to be better or understand what's going on. So, in order to facilitate this kind of trust with with the people, the first product principle we use is actually make it very, very valuable for the seller. Up until now, sellers love Gong. And it's, like, not necessarily intuitive. Right? I think in a equivalent software in a call center that gives you it's like nobody like, the agents don't care about it. Right? So we put many capabilities, many UX things, many kind of things in a product, in order for a, sellers to like us, b, of course, sellers to find value in it, which is kinda the the second same thing is kinda like us. And the third thing is bringing them tons of value from almost from day 1. So the time to value is is is pretty fast. So so that was the that's where we started back in 2016. Okay. That's really interesting. And yeah, because I feel like before you guys came along, any type of call recording was more, like, as you said, like, big brother, but and you guys had a bit of a cultural jump to, like, teach everyone, like, okay. Get used to this. We're not spying on you. But now I feel like Gong is one of the few companies that sellers have, like, an extremely strong emotional reaction to because they love it so much. So it's really interesting that that was, like, your approach, and it it also was clearly very effective. And now everyone just expects Gong on every calls. It's not even odd anymore. Exactly. And and, you know, maybe a couple of maybe a year or a couple of years after we started, we heard that people were interviewing the company. They're like, do you have Gong? If not, I'm not coming. It's like, okay. That's a pretty extreme place to be, but it it was true. That's how people perceived it because it brings them value. It's like, you know, personally go, you know, go tell me to do my job without I don't know what it is, like, laptop, smartphone. It's kinda the thing you wanna will have in your life to be productive, and it became that, but it wasn't the obvious in the beginning. No. It's totally true. And even now, I actually can't imagine our sales organization without Gong at all. So, that that completely makes sense. And I guess now as Gong's product has matured and now people are more culturally and not only, like, used to Gong, but now, like, advocating for Gong, have you changed your product philosophies at all and how so? Definitely. As we go, when we started our main audience, I I wanna say in kinda common terms, maybe call it mid market companies. Companies, I don't know, like Gong or, I know, a 1000 people in change and maybe at the time, you know, other companies were 500, 300, I don't know, 1200. But it was mainly kinda what I would call mid market. Ever since we started getting, I don't know, fortune 50 companies, many of them. So it it obviously has an effect on on the product. Right? So you still wanna keep it simple. You still wanna have even employees at any company or kinda human beings. They wanna have good software that helps them, easy to use, fun to kinda interact with. At the same time, the level of configuration, especially, has become, like, way way higher. Because when you start in a in a mid market, it's like, oh, this is what it is. Maybe you can change a couple of things. In an enterprise, like it or not, there are, like, regulatory concerns around permissions. There's things that they don't wanna roll out to every employee. Imagine, like, a feature you roll out, I don't know, in your organization. You can roll it out. Everybody's gonna figure it out. You roll it out in a big telecom company, it's like 5,000 people getting it on day 1. They have no idea what to do with it. So much more configuration around whether how the features get rolled out, how do they operate exactly, you know, how much does the revenue or operations team, which is the one implementing Gong, have control over, how things operate, and then many more kinda workflow ish things that not just drive insights, but actually drive action in a very particular way. Got it. And how has that been for you as a product leader adjusting your mentality from, like, having a really simple great experience for, like, an IC to now, like, an organizational, like, point of view? Because it definitely is a little bit different. So my my you asked me about my background. So I started as a software engineer, but in a company that's sold to, like, very, very large organizations including, like, I don't know, military forces and whatnot. So I'm kinda used to this environment. Actually, it was it was a nice change to sort of be in a more, like, a b to c ish type of of of a company. I I truly believe there's no reason for b to b kinda software to be, you know, clumsy and and, you know, not friendly. So it was it was kind of fun fun to say, hey. Let's do b to b, business in the same way to b to c companies operate. Our brand was, like, you know, different and and trying to kinda come up with this, like, new approach. At the same time, when we did need to put those kinda layers of, you know, tell me what emails I can ingest, tell me how you run your forecast organization, tell me how you're kinda prospecting workflows work, I think we're ready to do it. Obviously, there's investment in it, but in the end of the day and especially enterprise customers really appreciate that because, otherwise, they can't roll out, stuff across dozens of, employees. Yep. That totally makes sense. And switching gears a little bit, so you guys were actually the first company to or first one of the first that I remember to have, like, the dot ai, like, domain name. And you also embedded AI in your product, like, from the very beginning. I remember you guys would have, like, some insights based on the calls and any information on, like, how the transcripts were. So now, that AI is super, super popular, what, like, what are your what's your point of view on AI in product and when it's, like, a good investment versus not, how to experiment with it? What are your thoughts on that right now? I think it's a timing question. I think when we started back in 2015, we founded the company. It was funny because I send this email to my family. It's like, I found a stock you should invest in and that was NVIDIA. It was, like, 2 2 bucks a pop or whatever. Oh, my god. You're rich now. No. Because I sold it along the way, but I I did obviously, I did invest in it and it was it was a good investment at the time. But it was kinda we foresaw the, sort of the the the the AI kinda taking kinda more and more active role. Of course, we didn't know it's gonna be called GPTs and LLMs and whatever, but when we started, we had to kinda borrow transcription, software from other people. It was 30% error rate, meaning 3 out of 10 words is, like, not the right one. Right? We have right now, we have our own one that is sometimes on without any errors. Like, I don't know, 5%. Like, one out of 20 words misses an s, like, folklore. Really, really good. So we we we didn't know these are gonna be the specifics. So we invested like we, but that was the assumption. So it's kinda in many ways, it is great to sort of see the the almost like technology adapting to where we hoped it it would be. I think every AI is this is not a hype. This is not like cryptocurrency. This is actually changing our life. The fact that you can go out of a Gong meeting and have a summary, have a next steps, even Gong is gonna write you a follow-up email. This is a huge time saver. I can I literally missed for following up with a customer, a couple of weeks ago because for some bizarre technical reason, Gong forgot to offer me, like, a follow-up action? I'm like, oh, 2 weeks later. I'm like, I'm not even used to taking notes or doing follow ups on my own anymore. I think it's a huge thing, but I don't think you should use it for everything if you have just, I don't know, You know, many software, systems don't necessarily need AI, but when it comes down to automating stuff, especially generative AI, you know, write something, create something, it saves people time, and then it increases the value of your software. And and, of course, you're gonna be expected to do this even if you don't. Yep. Totally. And especially right now, like, I feel like there's so many different composable there's a lot of different pieces of composable software. You can, like, get ChatGPT, and then you can also purchase, like, you know, different like, you can use Stripe, and you can use Twilio. You can use all of these different APIs in order to construct a new product really quickly. How do you think between, like, buying versus building in house? Like, what what are usually the trade offs for those 2 different decisions for you guys? For us, it might be a little bit different than the sort of your typical organization. So because we have a a large data science team, so we invested in, I don't know, a few dozen different models. But I would say most companies, you wanna start with an off the shelf LLM. This is also there's, like, plenty, so you gotta pick. Do you want this company? I mean, have they have slightly different characteristics. So just test it out. Like, say you wanna have, in our case, like, a summary of a meeting. Right? Try 3 or 4, see what the quality is. I mean, you're gonna do some prompt engineering, blah blah blah. But, eventually, you're gonna get something that you're gonna be happy with. With a reasonable price, you don't have to pay for the latest, I don't know, GPT 7 or whatnot, because eventually, it does become expensive and start there. Now if you want, like, in our case, with some point, we wanted to increase accuracy, and that's when we went and said, hey. Let's next steps, you know, GPT gives us 50% accuracy. Maybe you wanna tag, label, fine tune, do our own models, and you get 80%. But you don't have to start there in most cases. But you do have to assume that if you're widely successful in a certain area, like, there's something that you're gonna do on our users do on a very regular basis, they will want it to be kinda accurate fast, and then you might wanna improve over time. Yep. That makes sense. And right now, I feel like there's so many different ways that every company can bring their products. Like, there's there's a lot of new ideas that are possible now with AI and then also with composable software. How do you end up deciding what, like, what to build right now and when? Because I think everyone's kind of having that issue, especially with, like, much larger customers that have, like oh, sorry. Much larger companies that have a lot of customers. They're kinda yelling at them to build certain things, but then you also have, like, your new road map of or, like, moving to enterprise or moving to new markets. How do you end up making those trade offs, especially with limited resources? That's always a product a tough product question. Right? There's always more requests than than humans to fulfill those requests, and and you have to some point to kinda balance those. And I think it's usually it's trying to come up with just, like, magic numbers of saying, hey. We're gonna invest whatever it is, like, 25% on, you know, call it keeps the lights on. Right? And then 25% on whatever it is the current product set that's, like, tactical improvements. And then 25% on, you know, innovative stuff, but within what we do right now. And then maybe 25% on new net new products or new capabilities. And that's some somewhat of an overall situation. And then within this, if there's AI features, they become even more complex because, like, oh, how are we gonna know if because customers are not asking for, could you help me write an email after the meeting? It's just the acknowledge driven innovation. Right? So it's like, let's see if we can iterate on those, come up with something, maybe pilot it ourselves, figure out a way to quickly determine if it provides value, how much value does it provide, and then roll it out gradually so you can assess the value over time and iterate. Got it. How much right now of your road map is driven from guy or data and data? And how has that changed over time? Driven from, like, data coming from from customers or Customers, like pipeline. Yeah. We always like, we've always been customer centric in the sense of, like, hey. There's a customer that wants feature x. We think it's on the road map. Let's see if we can accelerate it. We've done it in 2016 where we had, like, 5 customers and and maybe, I don't know, one more customer I wanna be, and now that we have over 45100. So I think it's less around data per se. It's, like, 30% of customers ask this. It's like, yeah, I don't think that's because he can't tell, like, how many are actually gonna benefit. But we are trying to I would say probably 50% of our road map right now can be attributed to at least one logo that is asked for it. Doesn't mean that this is the reason we've developed it, of course, because not like we don't reactively develop stuff. But we actually said, hey. Let's do it in this q3 now. Let's do it in q3 versus q4 because we have a big customer that's gonna leverage it. I almost believe, like, we're, you you know, there's there's, this summer has been sort of famous for the Olympics. Right? So it's, like, almost like judo. It's like, you wanna leverage to the other forces to kinda give you give you benefits. So it's like if a customer is pushing, it's not, hey. Do I play along? It's more I kinda leverage the fact that there's an excited customer, you wanting to roll it out. There's a bunch of people there who are gonna be your 1st alpha customers and raving fans to actually just accelerate stuff. Give them a version that doesn't have to be the public generally available version, but use it to kinda expedite, expedite the road map. I I love this. I'm probably on the extreme of using customer forces to expedite stuff. Same. We only build things that customers want. Otherwise, it's a little bit hard to get it tested and then also just validate that it's it's working as you were hoping for it to work. What is a more controversial product decision that you've made that ended up working out? That ended up working out. Wow. That's a great question. I I think many. Even even, we launched the forecast product. Right? In a couple of years ago, we launched a sales engagement product. It wasn't trivial. We could have gone on in in our traditional conversation intelligence capabilities made it smarter. But, we our our assumption was, and I think it's the right assumption, that companies are gonna wanna have a single revenue revenue Forrester calls it revenue revenue orchestration platform, but whatever it is, like, one single pane of glass where all of the sellers are gonna work with same datasets, same interface for sellers, managers, ops, other people in the organization. And I think if we had stated conversation intelligence, we still would have been the number one there. But I think many companies would have been, would have preferred to just have a single platform, maybe kinda maybe less strong in a certain area. Whereas I think where the market is heading towards is the single platform that has all of it. So we did have to sacrifice some innovation ideas that we had, and that's build a forecasting product that is kinda more well established base, and then we try to innovate within this, but it's still a problem set that customers were familiar with. Yep. Wow. K. And I I love the sales engagement products. I'm glad you did release that. That worked well. And so I actually talked to your, former CMO for a while ago to get advice when I when we were looking for a marketing leader. And one thing that he said that I loved was you can't be the best and the cheapest, and Gong is one of the best product experiences. So what do you think that you guys did differently that just, like, catapulted you guys to be the best? I I think from a product standpoint, we're really kind of we're focused on on driving what we call raving fans, but really, like, customers who users and customers that like the product very much. So we were obsessed with, hey. They're not using this feature, and then let's figure out why and kinda make it more more useful, more valuable, more discoverable, all of these things that you can make make the product more better for for, for users and for customers. That was the thing. We kinda call it measure the cell phone. Not everything is measurable. It's like you don't measure customer a lot. But, yes, we did build paid a lot of attention to NPS scores and, I don't know, we had, I don't know, like, 70 at the time, which was, like, obviously, kinda super high. That was in the beginning. And then, I I think some other players in the space are, in my in my opinion, right, I haven't been there, but I think we're kind a little bit to kind of fight a feature war, which is, hey. We did this. We do this. We still have competitors to do this. Like, we do this. We do this. And I think, yes, every software company, you go to a Salesforce or you go to a slower end product, I don't call it the Zoho, they probably have the same check marks and they kinda do the same thing, but the way they do them, the way they kinda create outcomes is very, very different. So, I I think focusing on making sure the customers are successful eventually pays off versus just focusing on and we have all of the capabilities that some somebody might ask for. Cool. Cool. I'm guessing the cheapest phones still do the same thing as an iPhone, but I still own an iPhone. Right? It's like, it doesn't better. I'm being more productive. I'm gonna spend a couple of $100 more just so I'm productive. Right? Yeah. For sure. It's definitely a more enjoyable experience too. Yeah. Yeah. And I guess last question, for, like, this conversation is, do you have any advice for product leaders that are currently making those really difficult decisions because resources are so constrained and the tech market is not the best for companies that aren't in just AI? How do you end up making those really hard trade offs and what do you have any advice for them? I'll tell you one kinda what advice is always to have the customer in mind? So it's sacrifice if, like, your own ideas if this customer if there are, like, a bunch of, obviously, cohort of customers who are, like, passionate about something. Always on the side of making customers successful. Also on the side of making your existing customer successful over acquiring new customers in general just because if you don't have, like, very, excited customers, it's gonna be very hard for you to grow regardless of, like, how many you acquire. You're just gonna lose them. And then the other thing is probably this is gonna be a little bit of controversial. Don't sweat over it. So in a way, if you have 2, if you have a decision between like, if it's a black and white decision, you're probably gonna not gonna spend time on it. Just go with the white. Right? But if it's, like, a light gray versus the dark gray and you're not sure, if you're not sure, you don't have all the information anyway. So it's probably better just to make a decision and move forward for it than than sort of, like, contemplate over it for years. Because, again, if it's a 49 versus 51, the reality is, a, you're not gonna have more information, and b, it's not like a radical game changer. I mean, we look back I mean, my partner and I look at back at some acquisitions we didn't make, for example. Right? Like, hey. Could you buy this company? No. It's like, I don't know. We don't like that to do that. And then and then we kinda have this discussion. Like, should we have acquired them? It's like, I don't know. Whatever. Would it have made a strategic difference? And we both concludes, like, it doesn't like, yeah. It could have been a big investment at a time, but it would not have made a big difference to the company overall. So it wasn't a bad decision or a good decision, but we weren't sure at the time. I don't think we're gonna be sure now, and I don't think it's really, really, really important. Just yes, no, move on, or, you know, develop x or develop y, move on. The one thing you gotta pay attention to is the one way door as Amazon calls. It's, like, very, very irreversible decisions, but 95 out of a 100 decisions are not, like, as irreversible as you might think in that moment. Yep. That's great advice, and I love that. I totally agree. Cool. Awesome. So I think that's it for today. Thank you so much, Elon, for coming to our webinar. We really appreciate it. This is really great for me, not only me, but hopefully the audience as well. So, thank you so much for your time. Hi, everyone. Thank you so much for sticking around. So we wanna take around 5 to 10 minutes to answer some of the questions that have come through the chat. So let's start with this one. Sorry. I have to scroll up a little bit. So what are the KPIs that you at Gong use to measure the success at of Gong as a product? Likewise, what are the KPIs an individual PM can use to measure the success of a specific module or feature that a PM handles? Yeah. That's can can you can everybody hear me? Good. So, that's a good question. I don't think we have any, like, super surprising or super sophisticated means. I think ideally, if you can measure the impact of of the system, that's great. In our case, you know, if you can measure that easy Gong increases your win rates, that's the best situation, or reduce your run time. Usually, it's very hard to measure consistency across customers because there's many many many other factors. I mean, obviously, we're gonna say everything is easy to go on, but the market changes, their product changes, whatnot. So I think this is sort of more on a customer success end and sort of, like, value engineering. Product, we tend to have to focus on more consistent metrics, which, you know, like it or not, usually end up being more user space in the sense of, like, somebody's performing an action in the system, presumably part of a workflow that presumably generates business outcome. So both for products and for features, we have this sort of a funnel of, you know, somebody's actually used the system, they're using it maybe kinda across the company, and then maybe they're using it consistently. So the different metrics and the definition of the funnel changes being per feature or product. For example, we have a sales engagement product. So we're saying, hey. Good usage if more than half of the people who bought it are actually using it on a weekly basis. And then we wanna say companies, like, good usage if people are using it 3 times in 4 weeks or consistently in some other fashion. That's the key metrics we've all set to come. The perform knowing that if somebody uses the system, they're gonna get more leads, become more productive. So I'm like, oh, it boils down to Samsung finally, which is customized for each feature in Chrome. Cool. Awesome. Are there any other questions to have come in as well? Or does does anyone else have any other questions? Alright. Well, thank you everyone so much for joining. If there's any other questions, yeah, feel free to email us right after as well. We're always happy to follow-up, ask Elon any questions that he may have. Thank you everyone for joining us.