Product. Conversion. Scale.

If you don’t have a lot of time, this is what I’m trying to say with this post: prove your product in the small, then prove your business model in the small, then aggressively rinse and repeat in the large. Think about Facebook starting at Harvard, then going after a few other colleges, then going large.

People, including me, often want to go product idea, press, product development. This is exactly what I did with CrowdVine, which first appeared on TechCrunch as a un-validated idea backed by a half-working product. I got exactly one positive out of the TechCrunch post: I felt validation that I was, in fact, running a startup.

Post coverage, we changed our logo, our design, and pivoted from social-networks-for-anything to social-networks-for-events. None of the users that TechCrunch sent us ever converted to users in the new product positioning. None of those users even contributed to the information we used to make the pivot. That first round of press was a complete distraction and still, four years later, I get calls from people who read the old positioning.

The Startup Pyramid Model
So what is the optimal go-to market approach? Sean Ellis, from the world of Lean Startups, gives his take with the Startup Pyramid model, described on his blog and pictured above.

The approach is to get product/market fit first, then figure out conversion, then pour money into scaling. It’s not hard to understand, just hard to have the patience to follow.

Sean is from the world of Lean (and this post is from that perspective as well), but this idea of focusing on product market fit isn’t Lean specific. I read it first on Marc Andreeesen’s old blog before I’d even heard of Lean Startups or Customer Development: The only thing that matters.

The benefit of this approach is potency. During product development you get to be completely focused on figuring out your product, without the distractions of interviews, inbound calls, or scaling. Your work on conversion will be targeting a (mostly) stable product. Your money spent on scaling will be supporting real users (rather than imaginary users) and your marketing will be spent on channels that have been verified to convert profitably.

The opposite of this approach is guessing and flailing, often referred to as “spray and pray.”

Product Market Fit
What it is product market fit? Paul Graham has the best definition, “make something people want.”

Just to double check that you’re not fooling yourself and that you’re capable of understanding such simple advice, Sean Ellis has a quantitative rule of thumb:

Survey your users and ask how disappointed they would be if your service went away (very, somewhat, not at all, already stopped using your service). If 40% of respondants say they would be very disappointed, then you’ve achieved product market fit. Remember, it’s a rule of thumb.

See survey.io for the gory details on how to do that sort of survey work.

Not everyone needs to be so formal. We could feel viscerally that Twitter was powerful when it had less than 100 users. The trick is just to know the difference between the feeling other people get when they use your product versus the feeling you get when you think about your product concept or when you give your product pitch.

Once you have this “fit” you can move on to figuring out how to get users.

Get Efficient Conversion
Getting efficient conversion isn’t about getting tons of users, it’s about figuring out how to get tons of users. That’s a key difference, which is confusing because this phase almost always involves getting at least some users.

If you are looking for paying customers then you need to stay here at least until $1 of spending converts into $1 of revenue. Even if you don’t know your business model, stay here for as long as you are finding easy boosts to your conversion numbers.

This is where you might break out Dave McClure’s AARRR model (Acquisition, Activation, Retention, Referral, Revenue).

At this point in your company you might be dying to have some sort of vanity metric to brag about, like users. But the volume required to tweak these numbers is not high, in the 100s or 1000s.

The idea during this time is to turn your company into something that can scale. How does $1M in spending translate into growth? When you know that, that means you have something that can scale.

Scale
Assuming you get here and you know how to spend money profitably, now would be a good time to raise $100M. Why raise so much money? Because you know how to turn each of those dollars into profit.

I don’t have any other specifics for this phase, and why would I? If you did the product/market fit and the efficient conversion phases then you’ll know everything there is to know about scaling your company.

Gotchas and Edge Cases

1. Stealth mode. There’s a difference between stealth mode and nobody-cares-about-you-yet mode. Usually you can operate in the latter. That way you can gather as much feedback as you want without worrying about the distractions that come when people start really caring about you. With Lift, I didn’t think we had the option of operating in nobody-cares mode, but I still didn’t want to be in stealth mode because it makes it so much harder to talk to people. Also, out of respect for CrowdVine, which is still an operating business, I wanted to make sure I could get a clear message out. If that’s the case, you may get early press (we got fifty articles based on two blogs posts saying essentially that we’d formed a company, chosen a market, and had a landing page), but you want to make sure you remember that your priority is earning product/market fit.

2. Network Effects. People sometimes say that this model doesn’t work for a network effects business since you can’t tell if you have product market fit until you have scale. Nonsense. You just have to do your product development with access to a single high-density niche market. This could be less than 100 people if they mostly know each other.

3. Business Plans. How do you know what you’re working on is worthwhile? It’s possible that you’ll get product market fit and then realize that you can’t make any money off of it. There is a step zero (repeated when evaluating pivots), which is to apply a filter on anything that doesn’t have any possibility of being profitable.

For example, I was working with a company recently that thought it would be valuable if their technology were offered as a for-pay widget to bloggers. But no amount of optimistic business modeling was able to come up with a version of that business that even made $100k per year. So we looked at other business models until we found one that looked like it could be very profitable. The company still needs to prove everything in the pyramid above, but at least we’ve pruned out some of the obviously wasteful ideas.

4. Checkpoints. Each phase of the pyramid is a checkpoint at which you might turn around or stop. You might find that people don’t like your product concepts or you might find that there isn’t a way to scale.

For example, CrowdVine’s usage grew 70% last year, almost entirely off of word of mouth. But I haven’t found any profitable ways to generate additional revenue through other marketing means. CrowdVine is a profitable, growing, healthy company that let me live bicoastally and travel all over the world. But CrowdVine can’t pass the checkpoint for entering the scaling phase. That’s all right–it can still have a long, productive life, but it isn’t a company that can accept an investor’s money.

5. Getting Test Users. I often hear people rationalize a PR push as the only way they can get enough users to test product market fit. I’m extremely suspicious of this. If you are passionate about your idea then you’re going to be telling nearly everyone you know about it and those people (your friends mostly) will be begging to get in. Usually people who feel like they need a PR push are people who don’t know anyone in their target market. That’s crazy! The solution isn’t PR, it’s go to some events and make some friends in that market.

6. No Business Model. The point of this post is not to start a war over whether it’s acceptable to start companies that don’t know how they’re going to make money. CrowdVine made money from day one and the startup pyramid is the model I would have used to understand what we were doing. My new company, Lift, doesn’t know how it’s going to make money and the startup pyramid is also the model we’re using, with two tweaks. One, we need to be very aware of our ability to raise investment and two, in the efficient conversion phase we’re mostly interested in conversion to usage, rather than conversion to revenue.

7. Press. Technically, you do want some idea of how press converts into usage, so you do want some press in your efficient conversion phase. But it’s still distracting and you’re still figuring out other parts of your positioning, so it should come as late as possible.

Feedback
Like all second-time founders, there’s a lot of things that I’m trying to do differently. Most of them come from the world of Lean Startups (see The Lean Startup, Startup Marketing Blog, Four Steps to the Epiphany, and Lean Startup Wiki). This Startup Pyramid model is one of the ideas that I reference most frequently and that I want to be able to articulate to people we work with. Let me know if it makes sense or if you’d do things differently.