min to LISTEN
May 14, 2019

Growth Loops: The End of Traditional Funnels?

Brian Balfour
Brian Balfour
CEO and Founder

It’s time to stop viewing your business as a funnel.

If you want your business to grow more efficiently, it’s time to start implementing growth loops into your business’s growth model.

Brian Balfour, Founder/CEO of Reforge, and previously VP of Growth at HubSpot, joins the show to discuss how businesses can identify impactful growth loops and the first few steps of implementation.

Listen to this episode:


Everyone explains that making your business different is vital — but NO ONE (not even experts) explains how to actually do it... Until now.

Just click on that big fat red button, answer a couple of questions, and learn to stand the f*ck out in a no-bull, super-practical way:

"When are you going to do something in French so I understand it?"

Mr Grenier
My Dad

"You're literally the only marketer I can stomach."

Braeden Mitchell
Security Engineer

"A terrific celebration of marketers and marketing in all its forms."

Cindy Gallop
The Michael Bay of business

We covered:

  • Why not all companies function as a traditional sales funnel
  • The 2 biggest mistakes you can make when looking at your “word of mouth” growth loop
  • Why it is important to continue to layer new growth loops into your business model
  • Examples of commonly used growth loops
  • The 3 major factors that create a successful growth loop
  • The easy first step companies should take when analyzing growth loop models
  • Choosing between highly diversified growth loops or expertise in one growth loop
  • How to identify your growth loop’s most impactful point of leverage
  • Some insightful career path advice for marketers


Full transcript:

Louis: Bonjour, bonjour and welcome to another episode of EveryoneHatesMarketers.com, the no fluff actionable marketing podcast for marketers, founders and tech people who are just sick of shady, aggressive marketing, I'm your host, Louis Grenier. In today's episode, you learn why traditional funnels suck and how to leverage growth loops instead.

My guest today is the founder and CEO of Reforge which is a suite of selective programs for experienced professionals in marketing product data and engineering and he previously led growth at HubSpot.

He started multiple VC backed companies, grown user bases to millions of daily active users, he writes daily, not daily, detailed essays, nearly daily, essays on growth and user acquisition, has been featured on Forbes, Hacker Monthly, On Startups, is quite known in the growth world and so that's why I'm super happy to have Brian Balfour on board.

Brian, welcome.

Brian: Hey, thanks for having me.

Louis: So why ... I just started by saying the traditional, you know, funnel sucks. And that's really not me saying that, I just merely repeated what you wrote in one of your brilliant essays. So why do traditional funnels suck? And perhaps you can tell me more about what you mean by the traditional funnels out there?

Brian: Yeah, so I think you know, a great framework that was created and kind of popularized a number of years ago by an excellent marketer, Dave McClure, was kind of this whole pirate metrics, kind of the (AARRR) acquisition, activation, retention, referral, revenue.

I thought that was, obviously was a great starting point and very useful, but I think what we've realized kind of as we learned a lot more about, especially how software and technology companies grow, that you need to be careful of to not too kind of view your whole business as a funnel.

The reason is, is that if you look at kind of the, some of the fastest growing companies at the center of how they grow is more of a loop dynamic, meaning in a traditional funnel, you're really thinking about, "Well, I put so much, a certain amount in at the top and a certain amount comes out the bottom."

So you kind of get into this mindset of like, "Okay well I need to like, either just keep putting more in the top, or by the bottom, get more out of like for every unit I put in at the top, I get more out at the bottom."

But a loop construct is much more about ... zooming out a little bit, in saying, well how do I view my business as a loop in the sense that, whatever comes out at the bottom, I can sort of reinvest to generate more at the top.

And what it ends up looking like is more of a self-sustaining system, so I'm sure everybody's kind of heard the quote that you know, Einstein said like, "One of man's greatest inventions is compound interest" and it's kind of true in this case too right?

In the sense that when you think about it from a loop construct and you kind of get this self sustaining system, you get this compound interest effect.

And so for every kind of thing that you put in at the top, you actually kind of get more out over time and so you know, some very easy examples like on the consumer side of the business right, would be a product that I'm sure everybody's familiar with, like Pinterest where they take a new and returning user, a certain percentage of them kind of create new content or-re pin new content on their platform.

And then they use that content to create all these pages that are indexed in Google and people stumble across this page and that creates more new and returning users. And this is kind of a self sustaining system, right? And in continuing that front.

But it also works on the B2B side whether you're thinking about bottoms up companies like SurveyMonkey as people kind of create surveys and then send them out to people, get people to be exposed to their surveys and they end up becoming survey creators themselves.

Or even you should be thinking about sales as a loop as well because sales is all about basically how do you add people to the sales team, there's salespeople kind of work a number of leads, generate customers and then you take the revenue generated from those customers to hire more salespeople.

And so you don't really have a scalable sales team in a scalable sales growth model unless you're able to get that loop dynamic, you're able to take that money that's generated and reinvest it more into the thing that's helping you create new customers.

And so as a result, you know, something that we just like teach at Reforge, and kind of how I viewed all of my projects is just very much, kind of looking at this and constructing all different types of growth loops, right?

And within those growth loops there's a set of funnels that you can optimize, but you need to be very careful about viewing your entire business as a funnel, because once again you just kind of get into this endless mindset, of "I just need to pump more into it at the top to get more at the bottom" and that ultimately kind of hits a ceiling and is not super scalable.

Louis: And you also have this feeling if you look at it from a funnel perspective that you just put a lot of money out and what's in and what you get out doesn't really matter in a sense that the one at the bottom, do they end up paying? And then that's the end of the funnel, right? They just pay and then that's it, you forgot about them.

That's something I've been dying to ask you actually for the last while since I read your essay a long time ago now, is the, surely the simplest growth loop out there that probably works for any business if you think about it is, I buy your product, I like it because it's a great product on it's own therefore I talk about it to others, those that are, speak about it use it, et cetera, right?

It feels like, it's something I've been talking about the podcast a lot, which is the best marketing possible is to have a great fucking product in the first place, right? It seems like this is the simplest growth loop out there.

Brian: Yeah, it's probably the one that's been around for the longest which is this, like, word of mouth basically. And so like, one of the keys that we talk about is, well once you identify the steps, the loop, then one of the questions is, "Well, how do I, what are the pieces of the loop that I can control or I can influence, right? And help improve and increase?"

And so while almost every great product starts with a phenomenal word of mouth loop, I think one of ... and every loop kind of has pros and cons, advantages and disadvantages, and so one of the cons of looking at a word of mouth loop is that you don't ... the leverage you have to pull on that, the things that you can control, are either a more ambiguous or just a little bit outside of your influence.

But to take the ... but I think the biggest mistake with the word of mouth loop is just like saying one of two things, right? Either one, I have no control over it, that's just definitely not true. In the second thing is that, don't make the mistake in the first one of saying that you don't have control over it, but I'm blanking on the second thing, but it'll come around to me-So let me actually start with the first one.

Louis: Exactly.

Brian: So the first one is like, well we need to actually think about, well what can you control in that word of mouth loop, what you can control is basically, it's the careful balancing game between who you're building the product for and who you're exposing it to.

So for example, I see a lot of these product launches, for example, it's just like, "Oh let me get a TechCrunch article, let me get it on Product Hunt, let me game Hacker News" and like all of that kind of stuff, and what you're saying there is actually, well, "I'm just going to like release this product that you hopefully have built for a specific person in mind.

We're just going to let everybody and anybody in and sort of kick the tires," and the problem with that is what you end up happening is this massive spread of people that you haven't built the product for coming into the product and having a bad experience, not because you built a shitty product, but because you've exposed it to the wrong people.

And so all of a sudden you have more people having a worse experience than the people that are having a good experience. And as a result that has a bunch of negative impacts.

It actually decreases this word of mouth loop, it increases the number of people who aren't promoting your product over the people who are. And a lot of people look at that and say, "Well people will know that like the product just isn't for them, I didn't build the product for them, or that whatever," that's not how people think, right?

Like that's not how consumers think they just kind of go in and try, they get frustrated, and they're like, "Ah this thing sucks." Right?

And so that's kind of ultimately what you're creating when you kind of do those types of product launches, but it has a bunch of other negative effects in the sense that you just kind of like, really kind of muddy the waters on your data and your qualitative feedback. And all of a sudden it becomes a little bit more, less clear about like how do you, who do you listen to, who do you not listen to? Things of that nature.

And that's some of the most important stuff in the early days of a new feature or a new product, or a new business because, like, the more mud that you kind of have to sift through, it's just like, the longer the learning cycle, the longer a time, it just decreases your chances of building and growing something truly great.

Circling back, the second big thing that I was sort of blanking on before is, also looking at word of mouth and just saying like, "Well, all I have to do is build a great product, and like word of mouth will happen."

And that's not necessarily true, I think like what we just talked about is that, well the other side of that equation is kind of like who you're exposing it to, when you're exposing it to them, like how you're exposing it, these are the types of levers that you have control over.

And the other part of that is some products kind of lend themselves to word of mouth more than other types of products.

So this is the other thing about growth loops is that some growth loops work with some products and some don't, right? And so you need to identify those things and so a product that lends itself ... you know there tends to be this pull or this push model, in the sense of like, this thing is just, has this 10x experience and as a result I want to go tell everybody in the world, but then some products also have a pull experience on word of mouth.

Things like, big purchase ... consider purchases in our life, think about things around having babies and weddings, and real estate, right?

You always tend to have those few friends who are your like go to people that you go to talk about this and it's just some products have that pull strategy where it just, people to naturally have conversation about their product in their life, right?

And so you need to understand these social dynamics so you can understand, well what can you play to, what do you need to play to and how do you, what do you have influence over and what can you control and so, but absolutely, that's the foundation, but it more than just like hey people are telling about other product, you just kind of need to understand it at a slightly deeper level.

Louis: And the product should be visible as well. So I guess one of those would be higher if it's a product you buy, you wear all day, every day. You won't necessarily talk about your ... the fact that that's the worst example possible, but I have to say it, let's say you bought a sex toy, you're not going to necessarily talk about it to your family, like word of mouth or this type of product would be more difficult, right?

So I'm trying to think the worst example, so you're absolutely right in putting a lot of nuances into this one, but it sounds like from what you mentioned, that you can have multiple growth loop in one business, right? And it's absolutely okay to leverage a few, right?

Brian: Oh, yeah, I mean for sure, I mean I think like, I mean you look at the most powerful businesses over time, then the ones that, like Facebook and stuff, they're the ones that have mastered kind of laying on new growth loops over time because any growth loop you have has some sort of ceiling.

We call it the maximum scope in Reforge. So you can think about it like an S curve, right? And so as that loop kind of hits exhaustion there's a bunch of reasons why it might exhaust.

Things like audience saturation, like competition depending on the loop, there's different levers that just tend to get worse over time than better. And so as it kind of hits that top of that S curve, right, you need to start to think about, "Well what are the new, what are the things that I need to layer on over time."

And so if we thought about HubSpot as an example, right? They very much started with this dynamic of, they had an inside sales loop paired with like a content loop that sort of generated leads. What they then did overtime is that they layered on what a lot of people would call a VAR program, a value added reseller, we call it our partner program.

And so it was a new type of sales loop rather than like a salesperson generating new customers, salesperson generated new partners and those partners would get to go out and sell new customers and that was like a very powerful mechanism for HubSpot.

We then over time started layering on new products that enabled new types of like viral loops, like that was one of the big things that I worked on and so that product expansion the new loops that those new products have sort of enabled, have really fueled the next stage of HubSpot's growth.

And so the biggest companies are the ones that figure out how to layer these things on over time. The mistake is like at the beginning. I think the two mistakes are one is like, either not having a hypothesis about what your growth loop is and taking more of a like, "Well I'm just going to test everything and see what works," right?

But that's not really how it works because when you think about a growth loop you have to think about your product, your business model and a bunch of other factors and how they all kind of seamlessly fit together. And so that's the first mistake, is like just not having a hypothesis.

But the second mistake is to kind of go on the other end of the spectrum and say, "Well like I have these five growth loops," right?

And typically if you're like at a starting point where you're like, "Well to get growth I need all of these six things to line up", that's a really tough position to be in because typically teams can only really execute and start with like one to two at a time, really nail it, get a ton of growth out of it and start to layer these things on over time.

Louis: So from your experience, let's ... you mentioned a few, but I'd like to maybe list them and maybe we can pick one or two and tie them into it and give folks a way to do that themselves. So we mentioned a few, we mentioned the word of mouth growth loop that is one, it seems to be the basic one, you mentioned viral growth loop, can you give me a few more examples of the typical one you see happening over and over again?

Brian: Well the other one I mentioned earlier was kind of Pinterest's user generated content loop. And so that would be another one. There's multiple types of other loops, so basically word of mouth is just another type of viral loop.

It's just that the spreading mechanism and the reason why people are spreading it are a little bit different, like if you think about something like, you know Ubers or Lyfts, you know, instead of viral loop which is like a give, get sort of campaign. Or you think about that, it's the same exact dynamics except the motivation in the flow around how that's spreading is a little bit different.

Right so, I mean of course we've also mentioned other ones, I just mentioned in HubSpot right? Like just like a core sales loop, a lot of people call it content marketing, but basically we call it like a company generated content loop and I think a big here is like there's a difference between having content marketing and a content marketing that works as a loop.

So you can write a bunch of blog posts, whatever, that doesn't mean it's a loop. The thing that had worked for HubSpot was that they really figured out that the distribution mechanisms were either social or SEO, they really hammered down on like SEO, and then like as new visitors would come in, really get people to subscribe and so like every new piece of content that they generated would basically get spread out to more people, bring in more new visitors per average piece of content, generate more subscribers and so that's the compounding effect that you have.

You know, that's not necessarily the case for everybody company right? Like just because you have a blog doesn't mean it's like a compounding loop. So those are some of the general ones, but like, in Reforge our program, we actually ... we've identified like over 20 of them and we kind of go down and break them down and I don't think they're actually that's all it. Right?

I think like there's probably more and really understanding the dynamics behind these things? Like that's the powerful thing because then you can kind of create new ones that are custom to your business.

Like I said, in some businesses a dynamic works as a loop and in some businesses it's not really a loop, it's more of like a, kind of like a linear funnel. It just kind of depends on your product context.

Louis: And that's ... the main difference between in some businesses it's a funnel and in some businesses it's a loop, is it the fact that companies stand to engineer and optimize those loops on purpose? Or is it because just as you said, the context is different and you can't make a loop happen even if you would like to?

Brian: Both actually. So I think like, you know, in some context it's like, they're just not thinking about it this way so they're not actually like connecting the dots between how ... and the simplest questions is like, "How can I take this thing, user money, or a piece of content that's being generated and like what are the steps in reinvesting that back into you know, generating more at you know, the beginning steps of the loop."

And so there's like a lot of people that, like some of the products that I've invested in and advised, they've already had some of the dynamics in place, they just weren't necessarily what I would call, "Closing the loop," Right? And so just kind of building the pathway for users to, you know, it's just like anything, right you've got to pave the path for them if you want them to do it, right?

And so, just kind of figuring out how you kind of close the system as part of it. But the second reason is just like, it's different dynamic. So let's go back to that HubSpot example.

So the reason that you know content really worked for them as a really powerful compounding loop is due to a number of factors in terms of their timing in their target audience, in their business model. So let's go through each one of those individually.

Timing, they entered the market where actually like, it was pretty ... content marketing was pretty cream filled, right? And so getting ranked on Google and part of that was easier you know, 12 years ago when they just started versus today, now you have to go up against HubSpot and their domain authority of 98 and their page authority of 92, right?

And so there's a question there of like, well how do you compete in like a saturated environment like that? Well you've got to take a totally different approach or look at a totally different distribution channel, right. So that might be like doubling down on something like video and YouTube, or something in that nature. So that timing was part of it.

The second part was like their target audience. So their target audience was marketers and so when you think about marketers and you think about the part of the loop where you generate content and then that content gets amplified out to others.

Part of that was search, but part of that was social, well marketers as everybody here listening knows is like, we share a lot of content and so, but if you thought about, let's say HubSpot was targeting chief financial officers and controllers, right? They're not tweeting things out in articles, right? And so, that would make that type of loop such less powerful for a product that is targeting that type of audience.

And so, yeah, so we went through timing, target, audience and the third was business model which is there's a certain cost inherent in every single loop, right? And so the major cost in like that type of content loop for HubSpot was obviously the company funding the generation of that content. And so they needed to make sure on the other end of the spectrum the product that they were selling actually had a business model that supported that type of cost.

And so our ACV was about ten grand per year-

Louis: What's ACV? Just define it, please.

Brian: Oh, sorry, like Annual Contract Value, so basically just like, how much are you making on a customer per year on average. And so, but if I had a company where my, you know, my ACV was like, 500 bucks or 1000 bucks, right?

Like somebody targeting like very very small businesses, obviously that dynamic becomes much much harder, right? And so it was kind of a perfect fit from that dynamic as well.

So that's kind of what makes it ... there's other factors, but those are the three major factors where it's like, well in some markets and in some products it works as a loop and in other markets and other products it doesn't necessarily work as a loop because you can't really get it to compound over and over.

Louis: So you have timing, target audience, the model, it sounds like as well, your strength as a business, like the type of team you have potentially it could be something. I'm thinking; go ahead.

Brian: That's a great thing, like going back to the HubSpot example, it was marketers writing for marketers, right? Like that's pretty easy, right? It's much harder to hire some content marketer who can you know, create amazing content, for I don't know like, you know engineering managers.

Right? Like that's super hard. Like that's really really hard, right and do that was certainly part of it as well, absolutely.

Louis: So let's try to deconstruct that and help folks listening to perhaps identify one or two growth loops they can use in their business to engineer them, optimize them a bit more than what they currently have.

It sounds like, one of the first step you want to do if you're listening to this episode right now is almost auditing what's going on. Auditing like the dynamics at play in your business, in your business right?

So you said that a few minutes ago, you talked about it, like you help companies that you invested in to identify the dynamics and put that together. So how do you do that and how do you advise people to do that?

Brian: Yeah, the simplest thing to do and one of the first things I do with all of my companies is basically, I just basically get up on a white board and I ask ... or actually the better exercise is this, is like, you have everybody draw out a visual picture of how they think the product grows and everybody do that for like, ten or fifteen minutes and then put all those pictures up on a whiteboard, and what you'll probably realize is that everybody has a different picture and that's crazy, right?

Most teams are probably like, "Oh yeah. We're definitely on the same page." But then you put that up there and it's like, wait a second, somebody thinks the business grows like this, and the other person thinks it grows the business like this, and so that makes it like really hard to do anything else, right? Like talk about initiatives and all that kind of stuff.

So the very first thing is like just getting people on the same page and just drawing a visual map, right? Of like ... and of course the guiding framework is like, "Well, yeah what are the steps of our loops?"

As we generate newer, returning customers what do they do in order to generate more newer and returning customers? And so that's just a very simple exercise of just like, getting up on a white board and visually mapping the answer to how does our product grow?

And what you'll realize ... and so like if anybody listening, I hope some of you actually go do this, you'll probably get up there and you realize actually, this is a harder thing to answer than you would think. You would think you could just draw this out in five minutes, but it actually takes some pretty deep thought.

And so like, if you struggle a little bit that's okay, but actually getting to a map that everybody is on the same page on is actually really important, but that's just like the most simple thing that I would do, is like first start to try to answer, how does our product grow?

And if you're drawing these loops and you're having trouble how they connect like, one question I ask is well, how does like one cohort of users lead to another cohort of users?

Meaning, if I acquire a hundred people this month, how ... what are the steps, what are the things that happen that take those hundred customers and generate another hundred customers like next month? Or two months out from now?

Like what are the dots in between those two points? And that starts to start hopefully like reveal some things about like what's ... what potential like loops might be for the business?

Louis: So you get people in the same room, you get like the marketers, the product managers, the engineers, the designers, who do you like to put in this room like when you work with a new company to flesh that out?

Brian: So who ... yeah that's a great question. So, well one you want the entire management team kind of, initially involved in the conversation, the high level map, right? And so, you want all of the functions represented whether that's depending on what that is in your company is that product? Is that marketing?

Do you have a sales team? Do you have a customer support team? It's important that everybody kind of has a general sense of who that is.

So that's where I would start, but I think the most important people that to be in that room, would be certainly product, certainly marketing, if you have ... if like one of your things is sales, then sales needs to be there as well. Right? Like those are like the critical functions.

But I would try to get like the entire once again, like everybody kind of in the business needs to be thinking about this in somewhat of the same way. Then you would take that from there and you would start to like, communicate that to the rest of the team.

And you would probably take pieces of it and dive into it a little more detail with individual teams, because individual teams obviously they're going to be focused on different parts and different initiatives that target different parts of you know, this map basically.

And so you want to make sure that they understand those parts and how their part kind of contributes to the overall, the entire picture, right? That entire picture is kind of like how they fit into the grand scheme of things. So I would start with the management team with those major functions and then sort of expand out from there.

Louis: So let's say everyone is writing like on the whiteboard and they're only growth loops, maybe marketing is saying, "Okay, actually you know, we grow through Google Ads and we invest a bunch of money in the start, we get a bunch of people in, they sign up, they talk to their friends and that's our loop."

Maybe the designer would say, "Actually, no it's by using the product, so they sign up to the product, they like it so much that they show it to their friend and actually there's a small widget in the app that makes people share it and that's how they share it.

So you might have all of those growth loops, or maybe things that are not growth loops to be shared with you right? And once you have that and once you have the team pitching in and you have all of those growth loops in front of you, what do you do then? Like what's the next step in your mind? What do you start to do?

Brian: Well yeah, then the next biggest step is to identify what is your point of leverage? Meaning like, what is the most impactful part of that, that entire map. And so the question there is just like, what piece of that map if you make an improvement to has a ... generates a larger output or a larger impact than other parts, right?

And once you identify that, that's what I would call your point of leverage or like your strength that you really want to double down on. And once you do that what you realize is in a lot of cases is that you've got a ton of initiatives and a ton of resources that aren't really connecting or aren't really targeted at that point of leverage.

And so then you really need to ask yourself the question, "Well, like if there's other things that I can be doing to help improve that point of leverage, then it probably makes more sense to put those people and resources on that." Versus whatever else that they're working on.

And so like, that's typically like the next step and so that's kind of a strategy question, right? Is like, well you can't really answer that question without having the model itself. Or having that kind of visual picture. And so there's a lot of things that start to fall out of that.

Whether that's like your roadmap changes, or you priority changes, what people on the team working on changes, right? And so that tends to be kind of the next step in the process in what ends up falling out and what ends up being like, the super actionable piece of it.

But of course like, that changes over time. So you need to revisit this, So you basically, you know you basically ... you work on something for a while you'll help improve it, right? And as you improve it, basically, the biggest impact part of your model, your point of leverage, it changes right? And then so you need to like revisit it. You'll also learn a bunch of things.

You'll be like, "Oh we thought it worked like this, but it actually works like this." Right? And so you kind of need to revisit the map and that might change things.

So, anyways, there's all these things that once you have the map it's really becomes this strategic question of like, "Well what is my point of leverage? What is the biggest impact? And how is my team aligned to it, how are my resources aligned to it? How are all my initiatives aligned to it?" And that starts to create a bunch of, you know, really actionable things.

Louis: So I want to dive in a bit more identifying this point of leverage, but before that I just want to make a quick pause on the Pinterest example and please correct me if I'm wrong, but as you said, right now, Pinterest uses the user generated content to rank on Google so that more people discover it, more people use it, start to create their own boards, they rank on Google.

And correct me if I'm wrong, but before that they used to ... the same loop, but in sort of SEO and Google, they deliver it social media, right? But that started to die out. And so they had to change their loop to the one that is SEO based.

Brian: Yeah, that's correct I mean like, you know in the early days of Pinterest like they started during like the whole wave of the Facebook platform and the thing that they really tapped into was, you know, the feed API right?

So like publishing peoples pins and saves as stuff to the Facebook feed and that brought in, like new and returning users, but then as everybody knows, Facebook kind of just like clamped down on that over time and like really kind of killed that loop.

And so this gets to the dynamic that I call of like product channel fit, which is that we need to build our products to fit with the channels because the channels don't mold to us, right? Like they define the rules and so we've got to play on their turf right? They're not playing on ours.

And so, in other words they're defining the rules of the game so we better fricking learn them. And so, yeah that was like a case where basically that concept of product channel fit broke because the channel changed the rules and as a result Pinterest needed to go through this whole transition and this is something that, Kaycee Winners, who was a big part of the growth team during that transition led and they needed that transition to where they're at now was like, a multi year process.

Because when you do that transition, you're not only changing the channel, but the onboarding experience of the product, the change ... they had a bunch of technology and engineering and product that they had to build around like all these indexable pages and like, there's just a bunch of things that changed.

And so it just kind of highlights once again that these things move ... and then to the point that we were just talking about then their highest leverage area or their biggest impact area, right, you know, really fundamentally changed as well and so they're one of those that kind of succeeded through that transition, but a lot of people don't succeed through that transition.

You know I was part of that ecosystem but slightly different, I was in social gaming at the time and as Facebook started to clamp down on it and there was a channel shift to mobile, there were a number of gaming companies that went from you know, just printing money to going bankrupt, like going out of business.

And because they didn't make the transition when that happened, and so, you look at the biggest companies, Zynga, it took them years to make that transition, upon years, right?

And so like, yeah, when these big shifts happen or these big point of improvement change, like it's important to recognize them and go all in because once again, we don't ... they define the rules, we don't. And so we've got to play their game.

Louis: It could be quite a risky strategy to rely on one channel as a growth loop without any other in the background right? It sounds like you could really leverage it for a long time and then it could die out, and as you said, Zynga is one example of that-

Brian: So that's a hard question I think that's debatable ... like this is why like our industry and our craft, right I think it has a lot more depth to it on the surface, right because you know, if you think about if you're a fast growing company and you've got a loop working and your constrained on like time, resources, time, money or people, you know, it's this big question of like, "Well, I've got this thing working, I got limited time and resources, you know, am I ... if I put all of those resources on this one thing I actually know that I'm going to get growth faster.

I'll end up at a faster point six months, twelve months from now."

Or do I give up some of that growth to try to pursue diversification, right? And the answer isn't 100% clear, like I think, you know, it really depends on the context and the industry, and like all that kind of stuff.

I mean, certainly if like you were in the early days of Uber versus Lyft, right? And just like that heated environment, you know, that was about just like grow, grow, grow as fast as possible, double down like at all costs.

And of course that created some other bad dynamics which we don't need to get into on this podcast, but yeah it ended up being a question, but the pursuit of diverse.

I think I more see that people think that they need to be highly diversified at their early strategies, meaning like have five or six channels, or whatever it is in the very early days and that's not true actually. You typically are spreading yourself too thin and not building an expertise around anything going really deep on anything.

And so I tend to see that more often than being like, too siloed in one, in like one channel or the other. So it's definitely ... it's a good question to be asking, but it's not always 100% clear.

Louis: Right, it really depends on your vision as a company, whether you're VC backed or not, whether you're here for the long run, or whether you're here to sell in three years, and there's so many questions right and depending on the context of your business.

So going back to the point of leverage, which you started to explain, after people draw their maps in front of you, the growth loops, and you said you would basically identify the main point of leverage, the one that will have the highest impact, but I know listeners listening right now are thinking, "But how do you actually do that then?"

Because I've mapped out with my colleagues the growth loop, which one to leverage should I focus on then? How do I know?

So how do you do that like, and I know we have a few minutes left and there's plenty to say in your next part in this, but if you have to select the 20% that will lead to the 80%, like what are the things you would tell people to identify this point of leverage? What should they start looking into?

Brian: Yeah so I just want to make sure I understand the question, what you're asking is exactly how to understand at what the most impactful part of the model is, is that correct?

Louis: Yes, that's right.

Brian: Yeah. So just like anything it's part qualitative, part quantitative. So you can start with just like the qualitative discussion and this is a very valuable discussion to have internally because once again, my guess is, everybody's going to have a different opinion on what the most impactful piece is.

And if everybody thinks the most impactful piece is something totally different, then once again you're not on the same page. And you need to get on the same page. And so you start with that discussion and so, typically I think a lot of companies have somewhat of a qualitative sense from of like, you know, "If I work on and if I push on this area, I know I get a lot more out."

But the more precise way to do it is, is more of a quantitative exercise where you actually build a quantitative representation of this growth model, which is like basically building a giant Excel spreadsheet.

And like, every step in your loops is basically a variable in that model. There's variables like, basically produce some output, whether that's whatever you're solving for, new customers, new revenue, retained users, whatever it might be.

And then, you know, once you have ... it's very similar to building a financial model except you're not thinking about kind of like P and L and balance sheet and stuff like that, you're just thinking about the input and output of users and customers.

So it's slightly different, but it's a similar exercise, so anybody with like financial backgrounds, I think you know, tend to be able to this really well.

And then once you have that, then you can start playing with that model more specifically say, "Well if I move this step from X to Y, what happens to my new customer number, you know, six months from now?"

Versus, "If I move this other thing from, you know, Z to T" right, whatever, and start to compare it in that way, and of course, there's like something in the middle, you could do all sorts of scenario modeling while low, medium, high, right?

But that's how you would get more precise around it, no matter what you do it's not going to be perfect, right? You can't perfectly predict the future, right?

But the point is, is like, going through this exercise, you're making a much more informed decision than just kind of throwing a bunch of things against the wall and hoping it generates more new customers.

So it's very roughly what I would do and it's like one of things that we teach in Reforge, but it takes us 25 hours to teach it so- Explaining it in a few minutes is a little bit more difficult. Yeah.

Louis: So let's simplify it a bit, right? And I know, obviously, 25 hours versus 10 minutes is a difficult task, but I'm pretty sure we can explain it briefly when it comes to this Excel spreadsheet in a very simplified model. Correct me if I'm wrong, but each row would be a step in the loop, roughly? Each row of the spreadsheet-

Brian: Roughly, yes, that would be correct, yeah.

Louis: Right. So if you have ... if we talk about a Pinterest example a user coming they create their first board that gets ranked in Google, people search for it, they click it, they sign up, they create a new board, et cetera, so all of those five things I mentioned would be a row in the spreadsheet, right?

Brian: Pretty much, yeah.

Louis: And then what do you put in the columns then? Like what types of things? Do you put volume data, so like in term of how many people come in?

Brian: Oh yeah, so the columns are just your time period, right, so you drag it out, and then in the individual cells, once you have those things identified you look at historical data and you say, "Well on average what percentage of new users, you know activate on my product?"

And you put in that percentage and then you say, "On average per month how many pieces of content does an activated user create?"

And then you say, well on average, if that creates a certain number of index pages, on average, how many visits does an index page get? And then on average, how many of those visits convert into new users and so then you basically have all of that data and then you can drag it out over a time period and then you can just start to change things.

You can say, "Well, I don't know maybe we increased our conversion rate from, you know, Google visits to sign ups, what does that do over time?" But yes essentially, each row is a step in the loop, the columns are your time period, you look at historical data to plug in kind of historical averages.

Louis: And you also mentioned something where you could swap steps, right? In a sense. It's like maybe the row ... step number four, what if it becomes step number two in the loop, or did I misunderstand that?

Brian: Sorry could you repeat that real quick?

Louis: So let's say you have, you know, the five steps. And let's say step number four is a referral widget where people click on it and refer their friend.

Brian: Mm-hmm.

Louis: What if instead of putting that in the thank you page once they bought, we put this referral widget once they just sign up. So at the very start- Instead of so, do you also play with the order of the steps in the loops, or do you stick it to just-

Brian: A hundred percent. I actually think as in industry like, we probably become too siloed, too focused on and too narrow-minded on just like, looking at a stop and just optimize, optimize, optimize, right?

Optimizing is one strategy, but it's one out of a bunch and so like, actually sometimes when you look at the most impactful piece of your growth model their right strategy is not optimization, but it's more innovation. It might be like a new feature.

And in HubSpot's case we got to a point where actually the biggest lever on our model was creating a whole new product. Like HubSpot Sales Pro and customer support product and all that kind of stuff.

And so there's a whole spectrum of like pure optimization to pure innovation. And a bunch of strategies that live in between, right?

Yours would be, I would say like, not in the middle of the spectrum, but not purely on the optimization, it's probably somewhere in the middle, like a quarter on the spectrum, it's like, "Well, I'm not just like changing copy and optimizing on page, but I'm looking at this flow, and I'm like let's start to switch some things around, or let's like actually close the loop in this area," or something like that.

And so, yeah, we kind of have to like zoom out a little bit and get ourselves out of the habit of like, optimization is always the answer, because it's not always the answer.

Louis: So optimize, you can swap, you can create new product, like innovate. Any other strategies?

Brian: I mean, pretty much anything. Like you can create new features to like improve, you can create new products, you can tweak your like pricing and your business model. These are some of the bigger changes, right? Like also on this there's like a small to large spectrum as well, somewhere around this.

And so there's a, I'm trying to think of what else, like I mean even things like we were going back to kind of rooting this in the conversation earlier, things like, you might start off with an initial target audience, you might expand the target audience a little bit.

So that might be, well actually we need to spin up some new landing pages to appeal to this audience and all of a sudden we're capturing a larger percentage of the traffic that's already coming to our site. Like it's things like that, right?

And so like, those are probably like the largest ones of just like pure optimization. But yeah, make sure you understand like there's a spectrum of possibilities don't get narrowed into one.

Louis: And so the point of leverage, the one thing you really need to focus on the most is the step in this Excel spreadsheet that will have the highest yield if you increase it, right? So for example-

Brian: That's correct.

Louis: If we manage to make more Pinterst users create, I don't know, twice as many boards as they're creating right now then we are in a, we are like growing much faster-

Brian: Oh, way faster. Like explosion, right? Like, yeah.

Louis: That will a compounding effect in a sense then then twice-

Brian: Oh yeah.

Louis: As many boards are created therefore twice as many search results that's going to show up in more search results. There might be dynamic in play for Google, the more index pages your domain has, the higher domain authority, therefore you rank more, it just keeps on giving.

So what you're saying is really the root of strategy, picking the one thing, the one strength, the one thing that matter and just going all into it.

Brian: A lot of times yeah, it looked like, you know doubling down on that strength or, you know a lot of times these points of leverage are really hard problems to solve. But that's the point, like if you solve them and it's the point of leverage, you get massive results and those things tend to be harder to copy or harder for competitors to really go after you know, things of that nature.

But I think once again, like these concepts are really important for all types of marketers to understand because I think we marketers we kind of you know, sometimes get kind of put in a silo or put in a, you know, this little side bucket, but if we understand the bigger picture and can have the conversations around that, we tend to have a lot more influence over the direction that we can go.

Louis: So Brian thanks so much for going through all of this with me and the step by step, even though I know the Reforge program has way more depth to it, it doesn't really do justice to quiz you on this, but I think people got a lot of value out of it, so thank you.

What do you think marketers should learn today that will help them in the next ten years, twenty years, fifty years?

Brian: Oh that's a good question. It's one that I've been thinking about actually, look I think the marketing roles, like paid marketing, that tend to be very manual processes are either going to be automated, or they're going to turn into product teams. That's kind of like what I'm thinking.

That's what I'm seeing actually, is like a lot of paid marketing teams are actually becoming product teams and it's like leveraging data, engineering to scale up paid marketing. And so I think as marketers you need to move to one end of the spectrum, you either need to like move more to like the product end of the spectrum and learn like some of the technical skills and things of that nature to work on those types of teams.

Or you need to move to the end of the spectrum of like, the pieces of marketing that are still actually still really important that require just like, not a lot of manual processes.

And I think that's like ... look I still think things like branding in category creation and things of that nature are really, really powerful and impactful.

There's a huge lever for HubSpot and I think marketers that are kind of like, trapped in the middle here, that's a dangerous place to be in, so if I was to give any advice to people who are thinking about their career for the next ten years, kind of move to one of the ends of the spectrum because a lot of this stuff is getting automated.

Even a lot of the stuff like personalization, if you like take a look at tools like what ClearBrain is doing, and what MadKudu is doing on lead scoring, and what Amplitude has done on the analytics side of like figuring out, you know just building in like correlation and causation stuff into it, these are all like things that we used to do manually as marketers, by you kind of drop some code in and plug it into your analytics system or something like that.

And they're doing it automatically, super powerful and compelling tools, but yeah, that's what I would be thinking about is maybe not necessarily about the skill, but like how ... what direction you want to move in on that spectrum.

Louis: So branding your product, you need to pick your battle here.

Brian: Yeah ... I wouldn't just bucket branding just specifically, but I think there's actually ... there's room for very technical, data-oriented, engineering-minded marketers.

But the question is that just a product manager at some point? And so I don't know, like, kind of like what Geeom from Drift + Segment has been working on, like he's an awesome person to follow if you want just like, crazy awesome mad scientist marketer, he's incredible.

And so, I would say either product or more of the creative elements I think are probably the two places to really double down on.

Louis: Besides the Reforge program, it's a fantastic resource one of my colleague attended it and he vouched for it many many times over, so you have multiple programs besides this ... what you offer, what are the top three resources you recommend our listeners today?

Brian: Yeah.

Louis: Could be anything.

Brian: That's a good question. I would just choose a very small group of really like actually practicing and leading and just follow them. So the people I love and learn from the most are Kaycee Winners, Shawn Clouse, and certainly like Andrew Chen who I've been friends for a long time, Elena Verna who was at SurveyMonkey and is now at a company called MalwareBytes, Fareed Mosavat from Slack, he's amazing,

Darius Contractor who is at Dropbox and now the head of growth on Facebook Messenger, these are all people that, I just like, if I could just like hook a suction machine up to their brain and get it all out I would, and so you can follow ... some of them write, some of them tweet, some of them do speak, so I would just kind of follow them and learn whatever you can from them.

Louis: Nice, and Brian thanks so much once again for your time. Where can listeners connect with you, learn more from you, and learn more about the Reforge programs you run?

Brian: Yeah sure, just you know we do a spring and a fall cohort every year, just sign up if you want to be on the early notification list, we've got a bunch of email sign ups on Reforge.com, other than that, I wish I blogged daily like you said in the beginning, but it ends up being more like bimonthly and that's at BrianBalfour.com and I would love for you to join me there.

I am off social media so don't really do much tweeting or anything like that these days.

Louis: Yeah you picked the right growth loops.

Brian: Yeah.

Louis: Thanks Brian.

Brian: Alright thanks.