Sean Ellis talks about the 3 stages of Growth Hacking Success

good morning everyone so I am the second Californian that's speaking in a row here but unlike Nick I actually for me it's a bit of a homecoming so I I was born in Sydney and I have not been back for 30 years so yesterday was my first day back so it's really exciting to actually come back and see see the just the energy around startups and around growth in in Sydney it's it's inspiring and and something probably a lot of the rest of the world misses because you guys are pretty far away but so a little bit of context on my background I spent the first 10 years of my career working in in a couple of companies that from basically customer 0 through IPO filings on NASDAQ and through these two companies I learned that in the later years pretty much all of the execution that we were doing was based on things that we figured out in the early years and so getting the early years right the traction years right instrumenting the analytics figuring out who the heck needs the product in the first place and why they need it and how they use it and all of those things are really critical for long term being able to grow business and the challenge is that if you're successful like look Nick that you you kind of go through that upfront stage and then you then you are done with it and for in my case I had four years of sort of executing on that foundation twice and so what I what I missed was really experienced in that upfront area and so for the next few years I worked at companies like Dropbox and Eventbrite and look out where I was the first marketer going in and really being able to learn the cycles of what matters and when does it matter and how do you what are the similarities between companies and what's completely unique and is there some sort of systematic way you can bring a company to market and I figured out a lot of stuff and I'll share some of that in a minute but then most recently I kind of took all that learning and then applied it to my own company and starting and a company that has gone through some of the twists and turns that a lot of startups do and getting to that point where you can ultimately make the business really grobel but I'll be sharing a lot of the lessons that I've learned through that and through these other companies and ultimately hopefully it's something that will apply back to each of you in terms of whatever stage you're at if you're in the scale stage or in your in the early traction stage some of those lessons hopefully will be very relevant to use so the big takeaway that I had from from all of the experiences is that growth really comes in stages and if you if you fail at the first stage you're going to probably fail at the second stage and you'll probably fail at the third stage so you have to get the first stage right which is product market fit before you can start to lay a foundation of really stacking the odds for being able to grow the business and once you do those things then you can start to scale the business and so most of what I'll talk about today is is around scaling growth but I wanted to start with some of the foundational pieces they're really critical so product market fit is something that I'm sure a lot of you guys have heard about and but it can be a little bit confusing so for me that just at a simplest form it's about creating a must-have product for a market that's that's big enough to be able to scale a business into and the way that I figure out if it's a must-have product is actually just ask people and so I know that I have something that's scalable when enough people who I ask when I want to say how would you feel if you couldn't use the product anymore if they tell me they'd be very disappointed that's a really good sign that means I can I can we've created something of value for them but if they say they'd be somewhat disappointed that's that's a nice-to-have product if they say they wouldn't be disappointed or they haven't even they don't even use it anymore that's a really bad sign and so what I found is that if 40% or more of the people who you ask that question to say they'd be very disappointed without the product then you have a good fighting chance of being able to grow that business of course that's just a predictive metric the reality is if they say this and then they stop using the product then that that's probably even more important than their answer to that question but most people who say they'd be very disappointed without the product continue to use the product so assuming that you have product market fit then then the next stage that you need to focus on is the stage that called stacking the odds for growth and stacking the odds for growth really boils down to figuring out who are those people who can't live without your product learning everything you can about those people well why did they sign up for your product in the first place why did they why do they keep using it what's the key benefit that they're getting white why is that benefit actually important to them how are they using the product when they when they start using it what do they do first and what's the difference between the people who don't care that much about the product and so when you start to find the signal of the product being a must-have for those people if you can really learn everything you can about those people then you have the formula to try to repeat on future people so that's really what the next piece is really about figuring out that value delivery machine that you learned on that first group and applying it to another group of people so setting a relevant but powerful promise a lot of people get so caught up in a be testing these days that they they ultimately end up creating a promise that's totally inauthentic to what the product can actually do because it tested best on an a/b test and got people kind of in the next step where ultimately if if the promise that you put out there is one that is based on what people love about the product then the people who convert on that promise are actually going to be really happy when they go in and use the product and realize that it is capable of delivering on that so strong powerful promise is really important and then streamlining the delivery of that must have experience getting as many people into that must have experience as possible having them get to that point where they just they get all of the benefit of the product as quickly as possible they're much more likely to say stuck into the product in and not churn out in a short period of time so once you have that value delivery engine once you have the value in the first place now you're ready to scale a business and then you can focus on actually doing the things that lead to long term growth and and we'll be focusing on that for the rest of the talk here so the one thing to start with is that scaling even with all those things it's not easy it's it's actually really hard and it's getting harder and so this is a graphic that James Courier the really successful marketer in Silicon Valley put together where it's just sort of the life's life cycle of channels and just when you figure out a channel it stops working and even even a channel so something like MySpace disappears completely but Facebook the things you did a year ago and Facebook that worked don't work anymore and you need you need to figure out the new things that work in facebook so with changing algorithms you need to keep adjusting to those so that's that the shelf life of a channel is really short now so you got to get in there quickly and the fact is that most marketers miss that little window within the channels and they if you're in six months after a channel becomes a viable channel you probably miss the window and it's not going to work anymore so you have to kind of change your whole planning an annual marketing plan to figure out what you're going to do just doesn't make sense when things are emerging and crashing in really short periods of time and then to exacerbate the problem more is that a lot of times marketers have the responsibility to get people to the front door and then the rest of the team is worrying about how do you how do you actually drive conversions all the way through the funnel and keep people locked in and if you have inefficient conversions and problems deeper in the funnel you're going to have a really hard time being effective externally in acquiring customers because your allowable customer acquisition cost becomes a lot a lot lower so the lot of the inspiration that I have for figuring out how to deal with the challenges for scaling a business taken from my own experience but I also looked to some of these companies so despite despite all of these challenges that are happening with these short shelf lives which have channels you have some of the fastest growing most valuable businesses in the world that have emerged in in just the last in the last five or ten years and so what is it that these guys are doing differently than then everywhere everyone else to get to this point of building so much value so quickly so one of the things that they're doing is that they've introduced this concept of growth teams and what you really have just about every fast-growing company in Silicon Valley today has a growth team and we'll talk about what a growth team actually does in a minute and so I've gone out and met with a lot of these teams over the last few years and compared notes on what was working for me and that's one of the things that's really great about Silicon Valley that I see in Sydney and hopefully you know across Australia and it can get better is that you got to collaborate you got to talk with other people you got to share what's working and and learn what's working and so by there's this great open culture among marketers in Silicon Valley and so I've been able to go and spend time and understand how they're organizing their teams and the processes that they're using and and it's and it's pretty interesting so I'll be sharing a lot of that today so one of the big takeaways is that and it's kind of obvious when you think about it but but surprising that people don't really adjust to this reality too much which is that growth is holistic so everyone across the company affects growth not just the marketing team yet the marketing team is really the only one who tends to be real metrics oriented in trying to quantify their impact on on growth but the but the truth is that whether the product team knows it or not everything they do is is touching an important growth who ever liked like engagement and retention and even revenue each each of those and the customer support team is affecting things and and the business development and sales team and really all of those touches together equal what your growth rate ends up becoming so you need to be good at coordinating that and that's one of the things that the growth teams have done is actually they often sit inside a product organization instead of the marketing team and as a result they have access to the development resources they have access to some of those deeper levers but they have to coordinate their actions either in some teams online marketing is actually sitting in that growth team and in others it's a it's a it's a separate team but it still needs to be super coordinated on what's happening there so another key finding and this one's not surprising for anyone who's been doing online marketing but it's a testing drives growth and if anyone's familiar with the game of Battleship it's just a great analogy and one that I borrowed from Brian Balfour who's who's the head of growth that hub spots sidekick product and it's basically the idea that that there's kind of two sets of tests there's there's tests to discover and then there's tests to optimize but the tests to discover are essentially every time you test something you learn something even if that test isn't successful and in the game of Battleship when you're when you're making a guess at where the other person's battleship is if you don't hit their battleship you drop in a white peg but you learn something and over time you have a board that has a lot of white pegs and red pegs and you start to have a good feeling of where the opportunities are to hit the other person's battleship so it's the same thing here inside one example with koala Roo I had a test where we actually acquired koala Roo from from KISSmetrics and it was called kiss insights at the time and part of the acquisition was I was sure that they were doing the freemium model wrong and that they needed to make a much more valuable free version of the product and so it's a product that had already gotten a billion views across all these other sites and I figured if we made a really valuable free product it's going to expand that base and take you know if I'm if I'm getting 4% of a curve that looks like this that's not nearly as interesting of a business than if it's growing like this and I'm getting 2% and so it's about trying to accelerate that adoption and and footprint and we invested a lot in getting the product aligned to that vision and it totally failed kind of sucked cuz we actually paid a lot of money for the business only to have have it not work out the way we thought but we learned something what happened was when we made it free there was almost no additional demand that we made a really valuable free product there's almost no additional demand for the product so what we learned was there's way less price sensitivity than we thought people either need this or they don't need it if they don't need it the price really didn't matter that much to make them need it so we went the other direction and we started building integrations and we added a lot more features and we had surveys that could have branching and multiple steps and we 10x the price over time as we added each of these things and all that premium pricing has been a really important part of our growth so that that failed test actually learned it helped us learn something so that so that's really test to discover is is pings maybe in new channels tests to optimize this just the realization that everything you're doing there's a better way to do it and you just need to figure it out but you have to have a baseline and you try something and it's and it's constantly trying to improve upon that baseline this one's kind of counterintuitive but you actually really need to focus on velocity more than the quality of the test so a lot of times people spend so much time trying to come up with the perfect test idea and you know let's let's let's try to hit a baseball analogy not sure if that's going to be understandable but let's try to hit a homerun with this test and put a lot into it and it turns out that that you're running one or two tests a month and that doesn't move the needle very much in the business because she just it's really hard to predict what's going to work and what's not going to work so you have to focus on trying to do a lot of tests and as you do a lot of tests you start to learn more and more with each tests and so this is an example of what happened to Twitter when they were they used to run two tests per month and you can see growing on a small number is much easier than growing on a big number but they their growth almost stopped in 2010 when they were running only a couple of tests a month then a new VP of Product came in and and pushed for ten tests a week and you can see that that change in the curve once they were running ten tests per week so doing something like that is not easy that's that's where you actually need a team that's built around managing that testing and so when you're thinking about building a growth team if you don't already have one the most important role is the person who organizes that team so it's kind of like agile development if anyone is familiar with agile development where you've got a scrum master who's is basically organizing the sprints of development here you're organizing the testing sprints you have one person who's got to manage those testing Sprint's that's the critical first hire and over time if you if you say I want to run three tests per week and you're having a hard time hitting those three tests per week with borrowed resources from other departments then start to plug in your own people so you can't get time with the designer try to get your own designer but don't go out day one and just assume you're not gonna be able to get I'm a designer because what's going to happen is that you'll waste months building out a team only to find out that you've got redundancies and you didn't need to go hire all those people so just start trying to hit a goal and plug people in as needed to hit that goal so these are some of the some of the roles growth engineer designer analyst this is some of the kind of the typical roles that sit within a growth team and then this team is really designed to execute a high-tempo testing process so this multiple tests per week and if they're not doing it they're not doing well part of being able to to hit a high tempo is to have a process that's a repeatable systematic process for doing it and it looks looks something like this so you know in your business the process may be a little bit different so it's not about there's there's a one-size-fits-all formula but this this formula has worked well for my business and it seems to mirror a lot of what what other successful growth teams are doing and it starts with just unbridled ideation a whole bunch of ideas one of the things that I found when I was running growth in marketing and in various companies is that I had a lot of pressure to always come up with the ideas of things to test and that was when if you figured out something that worked usually it worked for a year or two today it might only work for a couple of months and so there's this constant pressure to come up with new testing ideas but if if they don't work very long you it's really too much for one person to do so you need to have a lot of people participating in generating a big backlog of ideas then you have to have a good process for prioritizing which ideas you're going to run we each week so it's just this relentless prioritization and trying to figure out what is the very best ideas that I can run from among the ones that we have on our list and then launch tests so when I talk about running you know three tests per week it's not necessarily saying that they have to start and stop in that week but it's about constantly launching three tests sometimes you're gonna have to run them for a month sometimes you can run them for a week depending on your traffic but you just you want to keep getting that learning from from running those tests and then you need to capture the learning so I'll go into the details of each of these so for unbridled ideation might as I mentioned my team we've we've opened the door to where MIT I think maybe I have a hundred ideas of the 500 in our backlog and including the ones that we've pushed through the system which would probably take it up to 750 800 ideas I am we're not reliant on me coming up with the ideas and the truth is I I actually have a slightly higher hit rate on my ideas that they tend to work a little bit better than everyone else but like bear-like like 5% higher 10% higher than some of the other people so it's there's no reason for me to be the person that comes up with all the best ideas and then inviting participation from the whole company so someone who's working in support is going to be really familiar with challenges people have getting started with a product somebody working in sales is going to be familiar with you know how people's decision sorted before they get into your funnel might look like you know what how are they even considering a product like yours depending if you're a consumer product or a b2b product so basically each each of the groups that I've actually had in our engineering team some ideas that we would never have been able to come up with if you didn't have like a deep understanding of how how some of the the newest emerging technology works so everybody can can offer some pretty unique ideas into the system so it's not just internal but also external there's a good chance that you don't have anyone on your team that knows how to effectively use YouTube like Nick was talking about or how to how to effectively use Pinterest or Instagram but that there might be good opportunities in your business in those channels so you want to make sure that you're seeking the content that talks about that and seeking the experts and trying to try to learn from them and bring them into the ideation process with you as well and then ultimately all these ideas are just noise if you don't organize them in a consistent way so one of the things you see across all the leading growth teams is that they all have this concept of the experiment doc where they're actually pretty scientific about it and they've got a hypothesis saying exactly what metric they think they can move and how much they think they can move that metric by and why they think that idea is going to move that metric and any research that supports their their guesses around what's going to work and where in the funnel is it focused and even tagging up is it is it you know Twitter focused is it you know is there different channels the more information you can layer on that the better and then we use something we call ice for our scoring but some people use pie it's just some some way of being able to compare ideas so for us ice is about impact confidence and ease so impact is if if this works is this a game-changer is that something that's going to work really well in my business or is it a is it something that's just going to move the needle a little bit but we're really confident going to work in it super easy to do so why wouldn't you do it but so impact kind of with scale of one to ten ten being really high impact confidence is do I have a lot of evidence that my existing users are saying that they spend a lot of time in this channel so we should be experimenting in that channel or it's a best practice you see a bunch of across a bunch of other company any of those things can help you to increase your confidence score and then ease is it something that I'm going to show you an experiment that we ran that we got a seven hundred percent increase in email signups and it took like five minutes to implement that experiment super easy to implement it so sometimes sometimes ease is really important I mean I've had other things that have taken three or four days to implement and they didn't work so ease is an important thing to consider when you're deciding which experiments to run and so once you've organized all the information around an idea this way it becomes easier to start to prioritize the backlog and so for prioritization so you can do this in Excel what some of the screenshots I'm showing are actually a product that we have in private data called canvas but you can you can easily do it in Excel you can do it in Trello it doesn't it doesn't have to be this product but in this case so once you once you've scored them on ice you can actually sort them by really about if you're if you haven't had a win in a while sort by really high-impact stuff so that you've got people on the team are more likely to buy into the process if you're getting lots of wins if you are really tight on resources find the really easy experiments to run just so you can hit your tempo goal or if you're if you want to take a big swing and go for something really big then sort on impact but we're often just trying to balance across a number of different criteria as where as we're figuring out what to do but we do it through weekly team nominations so for the growth team basically each person is using these these types of criteria and we usually have a focus area as well and so we're using these scores and the focus area and each person is then nominating a couple of tests so we we nominate the tests and then we have a weekly growth meeting they'll go into in a minute where we're actually looking at each person has 30 seconds to pitch their idea to the team and and it's often not their own idea it's an idea that they've picked from someone else I mean fact we have a rule on our team that if you're nominating two ideas only one can be your own idea so one of them has to be somebody else's idea so that's where you do a quick 30 second pitch and then as a team we're essentially saying if we like it or not and then we narrow down and so in a pretty short period of time we're able to say okay these are the three or four or five tests we're going to try to try to launch this week and so launching the test then is is what we do in that meeting where we actually then choose the tests as I said it's the challenge in this meeting is that it's really easy for creative people to go down a brainstorming rat hole and so I'm going to show you an agenda on the next slide that we really work hard to stay within that agenda so that we don't we don't end up going down a brainstorming rat hole but each each of the ideas has a project manager who's assigned to it so that if that idea doesn't get launched now everyone has someone that they can go to and actually or at least the growth master or product manager of growth has someone they can go to and say hey why didn't that launch or where are we with this so somebody has ownership doesn't mean that they're doing the whole thing the coordinating a designer maybe they're coordinating some other people but there's they're the go-to person for that and then and then the PM of growth or the growth master is actually then assisting to help them get that test out and then the last step is capturing the learning from the tests so you want to make sure that it's really easy imagine looking back at that Twitter if they're running ten tests per week so HubSpot I know someone's gonna be speaking from HubSpot later today that their growth team on sidekick runs 30 experiments per week so if you haven't documented really well what you're doing what works why it worked if it didn't work why you'd think it didn't work there's a good chance you're going to be repeating the same tests over and over again or a new person gets added onto the team and they just they're not going to enter in any kinds of ideas because they just don't know what's what's there so you want to make sure that you make it really easy to access any of the information around what's already been tested what worked what didn't work because that's the whole process is is really being done to maximize learning so that weekly meeting that I mentioned so basically the benefit of doing it every week is that it establishes this growth rhythm where each day for my team we know at Tuesday at 11:00 a.m. we're getting together and we're going to go through the tests that we were supposed to launch last week and if we actually launched them what learning we got look at our KPIs where really everyone on the team knows that that's our that's the day we hold ourselves accountable for what we did over the last week and it's in this meeting that we're continuously improving the process so that's why I said it's not a one-size-fits-all process in your organization it might be a little bit different and you might need to figure out something that fits better for your organization but you want to constantly be tweaking things and figuring out if we're we're not getting our tests out is it because it's not a priority or do we not have that skill set within the company or is that person just so overwhelmed whatever it might be you want to try to understand those things and keep keep making the adjustments so that you can hit your testing targets and and really maximizing but maximize the learning that you're getting through running the tests in our meeting it's not just the core growth team but we actually do include the CTO the the vice president of product myself the benefit of including all of us first of all it's one hour a week and if the CEO doesn't think a growth meeting is important enough to spend one hour a week then that's that's probably a little bit of a bad sign for your organization because the CEO is ultimately most important thing most CEOs can do is is deliver growth to their investors and so important for CEO to participate the vice-president product it's important because if they're not part of that decision of what tests gets run there's a good chance that they're going to say oh well that that conflicts with the experience that we're trying to deliver let's we can't do that one they're going to overrule things but having them be a voice in those meetings can be really important CTO so that you just there's going to be a lot of technology decisions associated with this it's important to have each of those people but again that starts to become an expensive meeting and that's why you want to make sure you don't go down the rat hole of a of a meeting that goes for three hours because you get in a creative brainstorming session and so this is an example of the agenda that we use in our meeting we're literally our product manager of growth will tell me hey we're through the KPI section yeah I just want to give you a heads up if you want to keep talking you can but we're going to probably go over on this meeting then and so through that gentle nudging we're able to where be able to stay on schedule and she is able to you know ended at a time and it all it takes is a couple meetings going way over for other people to say I can't I don't have time for that meeting so this is this is how we organize our growth meetings so I'll take you through a few examples I mentioned this is an example of a test that so for for growth hackers one of the important things we have is an email that we send out each week that's the top growth stories of that week and the day we send that out tends to be our biggest day of the week we do it early in the week and it pulls momentum through the site for the whole week of driving interaction and engagement on the site and so our weekly growth letter is it's really important for our growth so being able to grow the the list of people who are subscribing to it is is important and all we did on this test was we move that collector for the you know get the top growth stories to your inbox each week we moved it from the bottom of the page to the top of the page super easy Optimizely tests but you can see it increased our monthly email collection by seven hundred percent and we did a whole bunch of tests off the back end of this but one of one of the tests like we call DoubleDown tests are the best best indication of something that's going to work is based on learning that you got from something else so for example are our top idea contributors actually our analysts because he's the first guy who sees what's working and what's not working and he's living in the data and he's constantly adding new tests as he's processing that information so one of the double down tests that we did then is once we have their email it doesn't mean that they can participate on the site the camp vote necessarily they can't comment they can't do anything that adds value to the site but they're halfway to an actual registration and so that was our double down test was just to give them this message that just says your you're one step away from the full growth hackers experience and being able to give that message to them boosted our weekly signups by another twenty two percent in terms of people who actually could sign up and participate in the community so that was an important double down to us but we had a bunch of winning tests that came off of the back of that first one and what you can see is that we actually didn't start this growth process until January about about ten months ago and part of it was that even though I've been a head of marketing for a lot of companies and growth for a lot of companies CEO role is really different and so I was pulled into fundraising and product development and HR issues and all kinds of other stuff and so I had to rely on a team to do these things and the only thing I could really do was keep telling them hey we need to do more testing we need to do more testing and they took it as oh yeah we better get another one out this month and by not putting a quantitative goal around the testing they weren't doing that much but the minute that I said I don't care what tests we do I don't care how big it is how small it is we needed a three tests per week that's my only goal I put out to them and as soon as I held them to that three per week we turned the corner and and everything else all of the process and management of the tests and everything else was a function of trying to hit that three per week but it's a three per week that ultimately delivers the results and again as I said the sidekick team does 30 per week the Twitter team does 10 per week so holding yourself accountable to a to a number of tests per week can be a huge huge driver on being able to hit your growth objectives and then this is just an example of you can find inspiration anywhere if you just look at at Dropbox for example if you pull down from the menu bar if you if you use Dropbox you'll see a number of things that they've tested and optimized so there's an upgrade button right there and you know the color of that the the how bold that is what what else could go in place of that upgrade button you know is it just upgrade or is it upgrade to get more space like what's what's the benefit that they tie there so a bunch of tests there then they also know the other reason people upgrade is they run out of space so they've got a whole promotion around here's why you don't lose your pictures add your pictures into Dropbox and they know that if they can push people to add their pictures they're more likely to hit the limit and then upgrade and then finally each time you save a file it shows you the five most recent files if you hover over that there's a share link right there if you click on that somebody else then is going to see a Dropbox promotion when they go to retrieve that file and so there's this is just one place where you can see three different growth drivers in there and as you start to just look at different sites you'll realize that there's a lot of in product growth opportunities that you can get inspired so some of the best ideas that I've I've run were things that I saw on other sites and kind of tied them together and ultimately led to big breakthroughs so at the end of the day I get a lot of people essentially say hey give me a growth hack have a growth hack yeah but but you know what I do have is that I can give ideas my ideas are probably not as good as somebody else's ideas who's closer to your business who knows your customers who knows your product who knows where your customers spend time and so by understanding those things you'll get the ideas and then you need a process and a team to actually manage and execute those ideas and prioritize those ideas and it's through that that team and through that process that you actually get the Silver Bullet that leads to growth and so it's not any one individual idea it's actually a process that helps you identify a bunch of ideas and figure out which ones are going to ultimately yield the results so that's it thanks everyone

Channel: StartCon

Views: 111,629

Keywords: startcon, sydstart, technology, tech talk, startups, growth, sean ellis, startup lessons, scale growth, market fit, high tempo testing, ab testing, a/b testing, growth master, growth hackers, conference, early stage startup, scale, growthhackers.com, internet marketing, product management, growth sprint, startup, growth hacking

Id: i0nxESR9sFs

Length: 34min 13sec (2053 seconds)

Published: Wed Oct 19 2016