“Where the #@!*% is Part 2?”

That’s what I’ve been hearing since we published Part 1 of our rare interview with Sean Ellis.

Here’s part 2.

In Part 1, Sean discussed what you do before product/market fit: how to get there, how to measure it, and how to survey your users so you can improve fit.

In Part 2, he explains what you do after fit: optimizing your positioning, implementing a business model, and optimizing your funnel — all so you’re prepared to acquire users quickly and profitably.

If you don’t know Sean from his blog or tweets, he lead marketing from launch to IPO filing at LogMeIn and Uproar. His firm, 12in6, then worked with Xobni (Khosla), Dropbox (Sequoia), Eventbrite (Sequoia), Grockit (Benchmark)… the list goes on. 12in6 “helps startups unlock their full growth potential by focusing on the core value perceived by their most passionate users.”

This is the first time Sean has done an interview on the record. I’m really psyched he’s making his insights public — this interview is a must-listen.

SlideShare: How to bring a product to market, Part 2
Audio: Interview with chapters (for iPod, iPhone, iTunes)
Audio: Interview without chapters (MP3, works anywhere)
Transcript with highlights: Below

This inteview is free — thanks to KISSmetrics

We’re bringing this interview to you free, thanks to our sponsor KISSmetrics.

Sean is an advisor at KISSmetrics and we interviewed their CEO, Hiten Shah, in How to measure product/market fit with survey.io. KISSmetrics built survey.io with Sean — now they’re collaborating on KISSMetrics, a new tool for funnel optimization that we’ll discuss in an upcoming interview with Hiten Shah.


You’ll get more out of this interview if you also read:

  1. Part 1 of our interview with Sean.
  2. The startup pyramid.
  3. We use several phrases interchangeably in the interview: Growth = scaling = acquiring customers with a known ROI. Preparing for growth = after product/market fit = optimizing promise + implementing economics + optimizing the funnel.


Here’s an outline and transcript of Part 2.

  1. What comes after fit?
  2. Prepare for growth
  3. a) Put the metrics in place
  4. b) Optimize the funnel
  5. c) Optimize the messaging
  6. Prepare for growth as quickly as possible
  7. Remove bottlenecks to preparing for growth
  8. Preparing for growth is not a low-burn period
  9. Preparing for growth doesn’t require growth
  10. Use a business model to grow quickly
  11. Grow and create new channels
  12. Nail the first user experience while preparing for growth
  13. Leave no room for the competition
  14. Marketplaces are an exception to this model
  15. Why Sean decided to focus on startup marketing
  16. Just scratching the surface


Music: The Equator by Tortoise

What comes after fit?

Nivi: Right. One thing you talk about is, once you have the product/market fit, trying to get through the next few steps as quickly as possible. So it would be great if you could talk about that, and also, for people that don’t know what the pyramid looks like, what are the next few steps?

Sean: Sure. As I said, if you don’t have product/market fit, you want to be obsessively focused on getting there, however you’re defining product/market fit. I think the easiest kind of definition to work toward is trying to get 40% of your users to say they’d be very disappointed without it.

And during that period you’re trying to stay very low burn and very conservative on all of your expenses within the business. And once you find 40% of your users that say they’d be very disappointed without the product, then you’re in a position that you have a business that can grow now. So then the question is: do you try to grow the business right then or are there some things that you can still do that are going to make it even easier to grow when you’re focused on that? And what I’ve found in working with so many companies at this [post-fit] stage is that you’re definitely better off waiting a little time until you try to grow the business.

Nivi: And what does it mean to grow the business? You’re talking about spending money to acquire users, essentially.

Sean: Acquiring users, for a lot of businesses, means starting to spend money, but it might also mean really trying to crank up the virality of the business, or it might be SEO that doesn’t require a lot of spending.

Nivi: But it’s time and money that you require for people… [interrupted]

Sean: Yeah, and focus. The time piece, you could say, has time or focus. But it’s essentially saying…

Prepare for growth

Sean: Is now the time to orient — especially the CEO and the marketing group and maybe a sales group — is now the time to build up that group and just say, let’s step on the accelerator and do everything that we can to get very aggressive about growing this business? And I would say, not quite.

The reason being, one, up until this point you shouldn’t have really focused on really trying to get the metrics into your business, so one of the ways that you can effectively grow a business today is, you can experiment in lots of different areas and figure out, in which areas do I invest a dollar and get that dollar paid back the most quickly or make a really strong return on the investment? But you can’t do that if you don’t have the measurement systems in place. So one of the first steps that you want to do is specify what metrics you really need to be tracking in order to be able to grow this business. And you want to then work to implement those.

So, to some degree, Google Analytics can help you, but the problem with Google Analytics is that you don’t really have tracking on a very user-specific level. So maybe you can track with Google Analytics or with Website Optimizer, even with Google AdWords with action tags, you can track a first transaction through dollars spent, but what you can’t track is lifetime value. And ultimately, expected lifetime value is what you should be basing your customer acquisition investments on.

Nivi: With the per-user tracking, is that what KISSmetrics is for?

Sean: That’s what KISSmetrics is working on. For other companies that I’ve worked with, it’s basically been database driven reports where we’re cookie-ing users when they’re coming in to track them back to the source, and we’re recording that information in the user’s permanent record in the database, and that way any revenue that’s generated from that user, we’re able to tie back to the money that was spent on that referring origin, coming in.

The reason why I started working closely with KISSmetrics is that all of my projects were six months, and that type of a system that I just described tends to be fairly complicated, especially as you start to track where you’re losing users along the way, and some of the other pieces. It was taking a big chunk of the time that we had together to implement that, and there were a lot of bottlenecks around it, so I was trying to find someone who could build that type of a solution as an off-the-shelf type solution and KISSmetrics was on their way to doing something like that. So I’ve been advising them for a couple of years to try to get the full vision out there of something that can really give the metrics that any business needs to be able to drive and manage growth.

Optimize the funnel

Sean: So metrics is just one part of it. Obviously, as I started to touch on right there a second ago, most companies initially have a very inefficient customer-acquisition process. So it’s not just how effectively can you spend money externally, but how can you convert people once they get to your website to having a gratifying first experience, and then ultimately buying your product? And metrics are a good way of determining experiments that you run along the way — how can I get more and more people to actually convert?

That can be through landing-page testing. You can be really effective on that front, or all the way down to just testing purchase prompts. There are all different areas, so you want to have a pretty sophisticated measurement system in place to do that.

I’ve worked with businesses where, at the start of that process to the end of that process, in a matter of just a couple of months, we’ve seen five times as many people purchasing just by more efficiently on-ramping them into the product and converting them into being paying customers, which means that when we then focused on trying to grow the business and buy growth, we could spend five times as much money to get someone to the website at the same return on investment, after that process.

And when you can spend five times as much money to get someone to the website, there are a lot more viable marketing channels that are open to you — a lot of times channels that would not have been effective at all before. If you’re putting a dollar in and you’re only getting 50 cents back, after you’ve gone through this process and you try that channel again and now you’re putting a dollar in and you’re getting $2.50 back, you’re going to put as many dollars as possible into that channel, and you would have cut it previously.

Nivi: And the expected LTV of the user is equal to your allowable acquisition cost per user. Is that right?

Sean: That would be your allowable acquisition cost for a break even. It kind of depends on what your objectives are. If you’re trying to do market share you might even be buying users at a loss initially. If you’re trying to generate a profit, you probably want to build in some profit on that.

Optimize the messaging

Nivi: Right. So, we put the testing in place. What other steps are we going to do before we grow?

Sean: So messaging. You want to make sure that you’ve got good messaging. So part of that was to get the product/market fit. You might be doing some positioning and messaging. But this is where you want to do a lot of landing page optimization and fine tuning just to see, what’s the best way to convert users coming in?

Nivi: In terms of positioning.

Sean: Ultimately in terms of converting them, positioning is part of that. To me, positioning is sort of the qualitative side and then testing and optimization is the quantitative way of getting to the best result. Hopefully, where you end up is somewhat consistent with what your positioning usage led you.

Prepare for growth as quickly as possible

Nivi: And what are your thoughts on just going through this whole process as quickly as possible?

Sean: That’s a good question. The mistake that I’ve seen a lot of companies make, particularly those that struggled to get to product/market fit, is that the whole mentality while you’re trying to get to product/market fit is to be super conservative and not spend very much money, so a lot of companies kind of [say]: OK, we got there. Now let’s prepare to grow.

And they, again, are really conservative and they’re taking six months to…. A prime example would be: I need to spend $100 a day or $200 a day to know that this is the best-performing landing page, or home page even, or I need to send 1,000 people or 10,000 people through there, and most people will space it out over two weeks. They’ll say they don’t want to spend more than $100 a day while we figure that out, so we’ll space it out over two weeks.

In reality, if you have some place where you can spend the money — let’s say you’re doing that through a Google campaign and you’re not getting the cap on that, but you’re saying, I just don’t want to spend more than $100 a day, you’re essentially going to spend exactly the same amount of money to get to the answer if that’s a good page or not. One way it’s going to take you two weeks; one way is going to take you one day. If you consider that time is money as well, then you’re closer to being able to accelerate the business after one day than you would have been otherwise and it costs you the same. But it’s just really hard I think, a lot of times, for people to go through that mentality.

Remove bottlenecks to preparing for growth

Sean: Another example would be if you are tight on graphic design resources internally, if you test 50 landing pages or 100 landing pages or 1,000 landing pages, you’re going to be so much more efficient at being able to drive growth in your business that it just doesn’t make sense, if your big bottleneck is on graphic design, then you want to bring in two or three people. Pay them a little bit more. It’s a temporary spike in costs to get much faster to the point where you’ve got a lot more landing pages in there.

So you combine those two things. We’re going to spend more to get people through, and we’re going to be able to test five landing pages at a time rather than just two. So being able to do all of those things… The goal should really be….

Preparing for growth is not a low-burn period

Sean: You have two low-burn periods in the business. The first is pre-product/market fit. You’re trying to spend the least amount possible.

And then the second low-burn period is once you accelerate the business you’re working within the parameters of your allowable acquisition costs, so you can actually go to being cash-flow positive pretty quickly at that point.

One of the companies that I worked with, once we got through this period of — it took us about four months to transition to being able to really accelerate the business and we got it to the point where we could scale marketing to close to a million dollars a month with a fast payback on those marketing dollars.

And at that point we had to raise some more money to be able to fund those campaigns. We raised the money on some very proven metrics and went cash-flow positive the month after we raised the money. So it shows that if we had taken 16 months to go through and really optimize that experience, it probably would have taken a lot longer to get the cash flow positive, but we were very focused and aggressive on it through that period of time and could really accelerate that business in a very cash-efficient way once we got through that period.

Preparing for growth doesn’t require growth

Sean: The mistake that most people make is they’re trying to manage growth and optimization at the same time. For this period of time, growth isn’t important. That’s the other thing that’s really hard for people to consider. That is not the objective — efficiency is the objective.

So for the first part of the pyramid, the base of the pyramid, you’re just trying to create a product that’s good enough, that people want, so growth is definitely not important.

A lot of companies still pat themselves on the back if they grow faster than expected through that period. You shouldn’t, because what happens, especially if you go to your VCs and you pat yourself on the back publicly to them, then you’re essentially saying growth is important to us, and you’re setting expectations that they should want more of it. So basically, say: Hey, we grew. That’s not what we were trying to do yet, but it’s positive on the hope for the business, but before we really focus on growth we want to get our product/market fit number to this.

Now, in this case, basically we want to be able to grow as efficiently as possible when we’re focused on it. And so what we want to do is build in efficiency in the business and the customer acquisition monetization.

Use a business model to grow quickly

Sean: So this is the time when you put in the business model as well, because you can’t spend real aggressively if you aren’t confident that you’re going to get that return on the investment.

A lot of companies make the mistake of… at least I believe it’s a mistake. This is probably one of the most debatable things that I say; a lot of companies believe they can grow faster without a business model. To me, having seen companies that can arbitrage dollars to grow and do it really effectively, you can just accelerate the business so much with that, why would you not spend money to grow? But you can’t do that if you don’t have a business model in place. So that’s why I tend to want to put a business model in place at that point.

The mistake of trying to put a business model in earlier, before you have product/market fit, is that you don’t know if you’re charging for the right things in the right way. You may be charging for things that are totally irrelevant for people and giving away stuff that is really important to people. So that’s one of the big reasons I like to delay the business model until, one, I have product/market fit, and two, I know why I have product/market fit — I know why people love the product.

Grow and create new channels

Sean: Now I’m in a position to put the business model in place, work on all the efficiency on the customer acquisition side, and do everything so that when it comes time to focus on growth, the marketing group or the CEO or whoever is going to be managing that process can be completely focused on driving growth — they’re experimenting with lots of channels and finding those that work and killing those that don’t, and basically just as quickly as they can, adding as many new customer acquisition sources to the mix.

And you shouldn’t focus on that until you know how to efficiently convert people and until you have a product that people really need. So that’s why this pyramid just gives you the idea that until you have really achieved that level, and if you’re trying to do multiple things at the same time, you’re not going to give things the necessary focus to actually accomplish them.

And your goal is ultimately to be growing a product that has very efficient conversion, that you’ve got all of the metrics in place to really manage that growth, you’ve got a product that people love, and at that point you’re starting to become a regular company. You’re organized into product teams and marketing and sales and finance and all of those things that are needed to create an efficient, high-growth company that can manage that growth and not implode under that growth.

That’s really why in that center period of time, it’s exciting. What should be driving you, day in and day out, is that we’re close. We’re right to the point where we’re almost going to accelerate the business.

Nail the first user experience while preparing for growth

Sean: So, one other piece that I think fits into this is that a lot of times it requires a lot of experimenting with how you introduce people to your product. What do you show them first? What is that whole first user experience?

The mistake a lot of companies make is that they try to get that first user experience right before they have product/market fit, so those companies are essentially pulling limited resources off of their core product and saying: we’re going to focus those resources on getting that first user experience right. Then that means that they’re not really dealing with the problem. They can maybe get twice as many people in to experience the product, but it still sucks. So they’re not really dealing with the issue.

But now everybody who comes in, or a big chunk of people who come in, are saying that they’d be very disappointed without this product — there’s a lot of love for that product. At that point, basically hitting the pause button on core-product development for two months isn’t even going to hurt the business because people already love it, and taking all of those resources that would have been focused on honing that core-product development, and refocusing those on: How do you introduce people to that? What does that first user experience look like?

The mistake that a lot of companies make is that they’ll take one person off to do that, and it’s going to take them a lot longer to do the development to get that right. But now you’ve got that core-product experience right, taking the majority of your development resources to get through it in a few weeks is going to get you there faster and that is going to put huge dividends on your business.

Nivi: You’ll become master of your own destiny much more quickly. Right?

Sean: Yeah. The way that I’ve been able to get development teams excited about that is, one, I give them the context of the whole big picture, and then I essentially say this is not something that we’re going to ask you to keep coming back and refining the first user experience. Once we’ve really got it right, backed up by numbers, backed up by saying twice as many or three times as many or ten times as many people are getting in and experiencing that core-product experience, then you can go back into just working on that awesome core-product experience that you’ve created, and evolving that and just continuing to create a great product experience.

But otherwise, we’re going to have to keep asking you for help for the next six months, until we get that first user experience right. It’s just much better for everybody, for you guys, to put the pause button on that, help us get this right, and then we’re going to be able to grow the business and it’s going to be a lot more exciting.

Leave no room for the competition

Nivi: And once you’re growing the business, what do you think about growing quickly to eliminate slack in the marketplace and basically leave no room for the competition?

Sean: Once you have product/market fit, and once others see you having traction, you’re going to attract a lot of copycats into the business. And they should be trying to come in because, like I said, the product/market fit is kind of the hardest thing to get to, so once somebody sees that you have a lot of passionate users who really love it, it’s the easiest way to create a moderately successful business — to knock that off.

So you have a short-term advantage, which is you’re able to respond to all the real product feedback that you’re getting from users who are coming in and experiencing that, so you can start to evolve your product better.

So partly what you want to do is just buy every related keyword, maxed out, and be very efficient on those keywords so that you can spend a lot more money on them acquiring users. So, the next guy can’t even consider spending keywords, so you’ve killed that channel for them. You just want to basically make it so that by the time someone can respond, it’s going to take them several months to be able to come out with something that’s close to your product. By the time they come in there, there’s just no market left for them to play in — you’re really redlining that market.

It’s interesting. In companies that I’ve been with through the growth period, the products that come easiest are the ones that often get the least focus because, oh, we’re hitting the targets on those numbers. We have to work harder on these other numbers. But it’s always the products where the growth comes easiest where the competition comes in because we didn’t take the slack out of the market in those products, and then suddenly competition that we didn’t expect, and then it’s not so easy to grow any more.

Marketplaces are an exception to this model

Sean: I think all of the things that I’ve talked through are a snapshot of what I think is best right now. There are so many exceptions in the different types of businesses that are out there. For example, marketplaces do need to focus on growth a lot earlier. Their business is a function of the people that are on there and how they’re using the product, and the experience changes for everybody else with more people on the product. eBay would be a good example of that.

I think each of these things are some of the guidance that I wish I had in the startups that I was in, but don’t just take it exactly as-is and think you can just plug it right on your business and that it will work. You need to sort of interpret what this means for your business.

Nivi: Do you have any experience applying any of this theory to marketplaces at all?

Sean: Yeah, Y Corp. is a company that I’ve worked with that is a marketplace that I’ve definitely learned a lot and learned that it’s different than a lot of the companies, like Dropbox or Xobni, where it’s a more specific product that, with a million users or fifty users, there would be experience where the product doesn’t change that much, where for Y Corp. the experience changed a lot.

Nivi: So how do you still take a thoughtful approach when you’re building a marketplace?

Sean: It’s just a lot more complicated. I think, for me, one of the things that’s kind of come out, in my head, on marketplaces is that you probably need deeper pockets for it. You don’t have the luxury of focusing quite as laser focused on each of the steps in the pyramid that I talked about — that you actually do need to think about growth earlier on, and that even when you’re then focused on growth you need to make sure that you’re constantly finding out if the love is moving, if it’s on something else. It’s just a lot more dynamic to try. It’s fun. It’s interesting, but it’s definitely harder.

And I’ve had enough conversations with other people from dating sites to anything that’s got the experience based on the mix of people that are there, that it’s really something that affects any businesses like those. So that’s one piece that, don’t just feel like you can plug this in directly. There are probably exceptions for a lot of these things.

Why Sean decided to focus on startup marketing

Nivi: Why did you decide to focus on startup marketing, and why is it important to you and how did you come to that conclusion?

Sean: It actually happened when I was coming off of LogMeIn and figuring out what I was going to do after LogMeIn. I knew that the company had gotten big enough that it was not really the size that I wanted to be working in anymore, and that I wanted to go back to early stage.

So it was more just thinking, OK, I can jump into another startup and hopefully I can be fortunate enough to get into a startup where it has the success. But just looking at realistic startup success rates, LogMeIn was the second in a row of two startups that I had run marketing from launch through a NASDAQ IPO filing, and I just knew, realistically, that there was not a good chance that I was going to be able to do that with a third one.

And so partly what I was looking at was just in general…. I kind of went through the thought process of: did I just get completely lucky, which certainly had something to do with it, or were there some things that we did that actually led to that success? And when I really started thinking through both of the companies, I came to the conclusion that the hardest part and the part that really mattered in building a successful company, was what we did in the first year, and really just figuring out everything about the business: Who really needs this product? Why do they need it? How are we going to charge for it? How are we going to acquire new users?

All of these things that needed to be figured out, once we had them figured out then it was much more about, OK, now how do we get more users? But a lot of the moving parts had stabilized at that point and we could really focus on just getting more new users. And so I thought, if that’s really the most important part, I thought that it was also the most challenging part, and so future success that I have is really going to be based on how well I can do that first part.

And then I realized that in 10 years, across these two startups, I’d only spent one or two years in that really challenging stage, so if I really wanted to get good at that I was going to need to figure out how to work in that stage a lot.

And the problem with startups is that every person working in a startup full-time, basically has an option vesting period that’s usually about four years on those initial options. So you work really hard in the beginning and you get some level of success and then it takes another four years to be compensated for that success, so there’s no way you’re going to create all this success and then leave the company right away.

So for me it was more about: how can I work really hard in that stage and actually be compensated for it? The only thing that I really figured would be the best way to structure it would be as an interim VP marketing role, and I was introduced, from one of the investors in LogMeIn, to a company that was in his portfolio called Xobni. They were still in private beta, and he said just talk to the guys. So we had a conversation that ended up extending for a couple of hours and we really bonded over some things.

And I got excited about the product and they got excited about potentially working with me and tried to bring me in long-term, full-time, and that was not something I wanted to jump right into. I was really passionate about figuring out this early stage and doing cycles there. So we decided that it made sense to pursue an interim VP marketing role. So I basically went there and thought, I’ll do interim VP marketing role after interim VP marketing role, for the next several years — essentially negotiate my exit before I came in so that I’d be able to keep a reasonable amount of options. That was the initial plan, and the whole idea was just to get the experience, up front, and practice, up front, to get good at that and to really document what’s important and what’s not important.

I think the challenge that I figured out was that, really, on the marketing side, there’s a lot of bottlenecks to get a lot of the things done, like optimization and trying to get resources to do a lot of those things. Everybody’s scrambling to get the overall product out.

So I knew a lot of what needed to be done, but there’s just a lot of waiting around until you have the tools that you need to do that. So where I’m looking at everyone around me working 14 hours a day, and wanting to put that time in myself, there just wasn’t that much to do. So I almost felt guilty that I wasn’t putting in the same amount of effort that everybody else was, so I still spent the time in the office, but I just found myself doing a lot of things that maybe weren’t that important.

That was why, after Xobni, I decided that I was going to do a more leveraged model where I could work with a couple of companies where they actually had somebody inside to manage the execution of things, but that I would actually be more on the outside, being able to not spend more time in the company than I needed to spend. So that’s why I did Dropbox and Eventbrite in that sort of way, where I was balancing two companies at once, and that worked pretty well.

Just scratching the surface

Sean: I feel like I’m still just scratching the surface on what can be learned in early-stage startups. There are not a lot of people that have deep enough experience and frequent enough experience in that really critical up-front stage to get good at it. It’s been fun to try and figure it out, but there’s a big variety of companies. There are certain things that I think are applicable to all businesses.

One of the most important discoveries in the last couple of years, since I’ve been focused on this stage, is figuring out this very-disappointed number and just understanding that trying to grow a business that doesn’t have users that would be disappointed without it, or don’t consider the product a must-have, is going to be a difficult and frustrating experience for anybody, including me. It’s humbling. I can’t do it.

I can’t do it very well, and when I see companies doing it I have a lot of admiration for them to be able to grow a business that is weak on that [fit] number. Those are the hard-core marketers that are kind of too good for their own good, because they’ve basically allowed the business not to focus on what the core issue is that’s holding them back. But those people would really benefit from taking a step back and tightening up some of these things.

I’m doing an experiment with Webs.com, for example, that is a later-stage company. They’ve been around for a long time. They’re a top-100 website and have a ton of user passion and a lot of the things that I’ve done to help prepare companies to grow are things that Webs hadn’t done, so we’re working together to see if we can retroactively put some of these things in place to accelerate their business.

And for me it’s all about learning and just experimenting in companies to find out what’s unique to startups and what’s unique to later-stage companies and what applies to everything.

Nivi: Great! That’s great. I just want to say thanks very much, Sean, for talking to us.

Topics Customer Development · Interview · Podcast

14 comments · Show

  • Sean M Everett

    I think this two-part series has some real gems in it (not just because I love the name Sean). LTV is a basic marketing concept that most start-ups pay absolutely no attention to, but is important in determining appropriate investment in the business.

    Through my various start-ups and working with clients of all sizes, the one consistent message I continue to hear is the absolute lack of appetite for spending any money, regardless of the start-up stage. As is stated in this interview, sometimes you have to spend money to make money. Many times people can’t conceptualize ROIC as it relates to their business, especially the part where spending $1 can result in $2 of income. I’ll take that bet. Every. Single. Day.

  • Matthias

    Great interview, thanks a lot for sharing these very valuable insights, Sean! I have a few questions:

    1. How do you organize positioning testing? Do you try to think of possible positions, test them and stick with the winner? How do you know, you’ve found the best position?

    2. LTV: In most cases, startups won’t have any history data about LT of the users. What’s your experience concerning handling this problem?

    3. Efficiency vs. Growth: I think I understand what you mean by that. If you’re trying out some acquisition channels and some have a positive ROI do you simply hold them off and you continue testing and try to be even more efficient? And only when you seem to not make any big achievements in efficiency anymore you’re going for growth and start spending in all channels that have a positive ROI?

    Again, thanks a lot for your help!

    • Sean

      Thanks Matthias – below are the answers to your questions as numbered.

      1) I get most of the ideas for positioning from the open ended survey on Survey.io and then start narrowing them down. One of the key projects that I offer is positioning and messaging optimization so I won’t go into the exact process that I follow — I’ve still got to feed the family :-). But based on how you asked the question, you seem to be on the right track.

      2) For LTV, I start with some very conservative assumptions around transactions per user, average transaction size, churn, etc. As the numbers roll in for users from each campaign, I update those assumptions with real numbers. I’m almost always too conversative in my assumptions — which is a good thing. You can burn a lot of cash if you significantly scale something that turns out to be a money loser.

      3) When I talk about efficiency vs. growth during the middle part of the pyramid, I’m really talking about your primary focus. If you discover a channel that generates positive ROI, it’s OK to leave it running. However, be careful to stay focused on improving conversion efficiency and don’t let yourself get sucked into trying to develop and manage channels.

      Hope this helps.

  • Michelle Klann

    Thanks Sean for sharing this information! I can very much appreciate your stance on solidifying the user experience and establishing solid conversion rates before pushing for large growth. It gives me a new, fresh perspective on savoring the challenges of the first year (to market) – especially when you think you may have found a “niche.”

  • Tristan Kromer

    Hi Sean,

    Do you put any effort into optimizing (or perhaps a better word is reducing) the length of the sales cycle? How would you go about this?

    Reason I ask is that I would imagine you can fund a lot of growth with short term debt instead of raising equity if you’ve got the metrics to back it up and you can routinely turn over debt. I would also imagine this helps building up your corporate credit with financial institutions.


  • Gabe

    Can you explain a bit more behind the thinking behind keyword buying – specifically in regards to “buy every related keyword”? What exactly does that mean? Can’t you just bid on keywords and can’t a competitor simply bid on the same keywords that you’ve bid? I confess I don’t know a lot about Adwords so I know I’m missing something. I just don’t see how one builds a competitive moat using keywords that a competitor can easily bid on?

    “So partly what you want to do is just buy every related keyword, maxed out, and be very efficient on those keywords so that you can spend a lot more money on them acquiring users. So, the next guy can’t even consider spending keywords, so you’ve killed that channel for them.”

    • Nivi

      Gabe, I think he answers your question in the quote you pulled out. If you’re more effective/efficient on keyword ROI you can outbid the competition in the long run. But in the short run anyone can spend enough money to outbid you.

  • Gabe

    Thanks Nivi!

  • Dave


    That was a very useful interview.

    Would you mind clarifying how you get the initial customer base to survey for market fit?

    Our startup will provide a monthly subscription service to brick and mortar retailers. We are targeting one specific sector where we believe our product/service will initially work best. We’ve been told by multiple prospective customers that our service is attractive so we think we’ve passed that gate. The next step is to get customers using the service and it sounds as though you’d say “do it for free…just get the users and get real feedback.” We and our Board of Advisors already feel that way, but the challenge is that each customer costs us $150 in on-site hardware so I want to think carefully about that.

    I’d appreciate your thoughts. Thanks.

  • Mike Volpe

    Thanks for sharing! The observations hit home based on my experience at HubSpot. It does seem like there are some tradeoffs between making the product perfect, and capturing the attention of the market, especially in a case where the problem you are solving is to create one tool that brings together a number of tasks into one place.

    Sean’s advice seems perfect to companies with a narrowly focused product, like Dropbox. I wonder if the right path changes at all for a company trying to create a new platform or more complicated system/product, like Salesforce.com. I think Sean would probably say to delight customers with some smaller piece of functionality first and then build on top of that later to create something bigger. I wonder if Benioff would agree? Or is the right path in that case to message to the market about the whole solution and then build it like crazy even if the early customer have lower rates of “love” than what Sean like to see. My gut tells me Benioff would have a different opinion, it would be cool to see a debate…

  • Daniel Sim

    Awesome article. It’s painful to see startups that have coded in a cave for a couple years, never touching their customers. I’ve seen friends in this stealth mode paranoia who’ve ultimately launched something that no-one wanted.

    As an engineer (and bit of a perfectionist) it took us three months to get the first Plug in SEO out. Short by some standards, but way too long. Since then we’ve been releasing at least every two weeks with tangible benefits from the feedback loop it creates.